December 24, 2024 • 62 mins
In this final episode of 2024, William and Eyvonne dive into a comprehensive year-end retrospective. What is happening on a personal level with The Cloud Gambit hosts? What were our favorite technical innovations? Could it be Google’s Willow quantum chip breakthrough? Or maybe, it’s Elon Musk’s “Colossal” data center build in Memphis. Join us for a fun conversation about the events of 2024, where Eyvonne defends her timing for putting up holiday decor and William attempts to explain quantum computing.
Show Links
- Google Willow: https://research.google/pubs/willow-quantum-computing/
- X/Tesla Dojo Supercomputer: https://www.tesla.com/AI
- NVIDIA AI Infrastructure: https://www.nvidia.com/en-us/data-center/
- Proxmox Virtual Environment: https://www.proxmox.com/
- Gallium Discovery / Montana: https://www.semiconductor-today.com/news_items/2024/mar/uscriticalmaterials-210324.shtml
- xAI Colossus: https://www.maginative.com/article/inside-elon-musks-colossus-xais-world-leading-ai-supercomputer/
- Tesla TTPoE: https://wcollins.io/posts/2024/tesla-adapts-ethernet-for-with-modified-transport-for-dojo/
Follow, Like, and Subscribe!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Eyvonne (00:00):
My kid had a school event last night and I got I kid
you not four texts and an emailsaying that the kids had to be
there at 530 to be ready for theprogram at six.
We pull into the parking lot at520 and the doors are still
locked and there's a line ofpeople around the school.
(00:21):
I'm like, and it's, and it'sfreezing cold outside and I have
this weird thing where myfingertips turn white when they
get cold, like my bloodliterally stops circulating in
the tips of my fingers and I'menraged.
William (00:35):
I need to take a vacation to Canada during the
winter.
Welcome to another episode ofthe Cloud Gambit.
We are approaching the end of2024, and this is actually going
(00:59):
to be the last episode goingout for 2024.
I am William your host episodegoing out for 2024.
I am William, your host, butwith me I have someone that is
so serious about the holidaysthat they put Christmas
decorations up before the turkeyis pulled out of the oven on
Thanksgiving.
That is Yvonne Sharp.
How are you doing today, Yvonne?
Eyvonne (01:19):
Hello, and absolutely I fully own the pre-decorating in
my defense.
We were trying to have aholiday party uh, that that
preceded a lot of the otherholiday parties, so the people's
schedules would be more openand could make it.
I'm not sure that it waseffective, though, because
(01:41):
people are busy, everybody'sbusy now.
We had 25 30 people in thehouse that day and we had a
great time.
Um, you were at your kids hockeytournament, so you didn't make
it, which I will try to forgiveyou for that, but uh yeah, we
had all of our decorations uptwo weeks before thanksgiving in
my house that's very, very.
William (02:04):
You know what?
This is one of my first piecesof holiday decor.
I'm just going to wear thisthing.
Eyvonne (02:15):
I should have brought a holiday hat.
I didn't even think about it.
It's not even the regular one.
William (02:19):
It's checkered.
It's like a flannel.
Eyvonne (02:22):
It's perfect for a Kentuckian.
Exactly there we go.
I just need the actual flannel.
I look like a flannel.
William (02:24):
Hey, it's perfect for a Kentuckian.
Eyvonne (02:25):
Exactly there we go.
William (02:26):
I just need the actual flannel.
I look like a lumberjack.
Give me an axe and some wood tochop.
Yeah, that's great Gotcha.
So, anyhow, so this episode.
So Yvonne and I were talkingand we're like you know what.
It's the end of the year andthis has been a crazy year.
A lot has happened, bothpersonally, professionally, you
(02:51):
know, the world markets.
Everything seems to be a littlecrazy.
So we thought, you know what,why not have an episode and just
kind of look back and just takea time to reflect?
Eyvonne (03:02):
you know, it's our retrospective Love it, so that
is what we are time to reflect.
William (03:05):
It's a retrospective.
Love it.
So that is what we are going todo.
Do you want to kick us off,Yvonne?
Eyvonne (03:12):
Yeah, we thought we'd start with a few personal
updates.
So things have been busy in myworld.
I think most of you hopefullyhave noticed I've been trying to
focus more on creating externalpublic facing content.
So part of that is joining theCloud Gambit with William so
(03:34):
that happened mid-year, soexcited to be here and been
reconnecting with some folks onLinkedIn, been hanging out a
little bit more on X and BlueSky.
So, yeah, I've been focusing onthat and I think for me, I've
also been working professionallyat Stretching and Growing as a
(03:55):
Leader, trying to think moreholistically and to also
navigate times of extreme change.
And how do you lead peoplethrough that and provide enough
guidance so that they know whatto do, but not enough, not so
much that you're constrainingand hindering creativity, and so
(04:16):
I've been thinking a lot aboutthose things.
What's been up in your world,william?
William (04:22):
Well, so first of all, before I talk, talk about my
world.
I just want to get yourperspective on something.
So you're talking about likegrowing as a leader and you have
, you know, direct reports andyou're working for, you know, a
big company.
And one thing just myobservations from in the past
you know having like led teamsbut also being in like the
(04:43):
engineer focus, like when you'rean engineer or you're doing any
of the boots on the ground work, you get so focused, of course,
on details.
You get so focused on all thelike very granular maybe tunnel
vision sometimes.
But when you sort of move upthat org chart a little bit,
you've got to like widen thataperture and you have to like
(05:04):
zoom out and you really needboth of those things.
And I remember, just when I wasin that engineering mindset and
I was doing a lot of the work,that used to drive me nuts
because I would always think youknow what these leaders.
They just don't know whatthey're talking about, they
don't understand, they don't seewhat I'm seeing, they don't
know how important this onelittle thing is.
(05:24):
But then as you move up and youget that bigger picture and
that bigger vision and you startseeing all the things you know,
it really changes yourperspective.
But any thoughts on that.
Eyvonne (05:35):
Well, I do think you know, the more your you know
your aperture increases and thewider the breadth of your
responsibility as a leader, itdoes shift your perspective, I
think.
For me it's trying tocontinually rethink.
You know what is important forme to focus on as the leader.
(05:59):
What is it that I need to besure that my folks have the
freedom and, frankly, theaccountability to focus on as
individual contributors.
And then how do I be sure Ihave an effective feedback loop
so I'm getting and understandingwhat are the real boots on the
(06:19):
ground problems?
And then how do I translatethose to the broader
organization to be sure that myindividual contributors can be
effective, right?
And so it's continually tryingto understand what's going on
enough but then also framingthat effectively so that you can
(06:41):
get organizational buy-in toenable your people.
And it's a very hard thing to do.
And I think sometimes it's as Isee we can not really grasp the
luxury that we have to focusdeeply on a thing and completely
deeply understand that thing tothe exclusion of everything
(07:05):
else.
Because when you're a leaderyou don't have that, in a lot of
ways, privilege anymore to justdeeply focus on a thing.
You've got competing prioritiesthat you have to think about
different stakeholders withinthe org.
Who who's who's.
You know who you need tosatisfy um both just
(07:28):
existentially and for thebusiness.
So it's it's.
It's a very different um, it'sjust a very different job yeah,
yeah, yeah, that was.
William (07:40):
you said that perfectly .
I doubt that Better than Iwould have anyway.
So, yeah, personal here, asidefrom traveling for work and
stuff, my son plays travelhockey and they are crushing it
(08:01):
this year.
Eyvonne (08:02):
So he's 10 and under.
William (08:03):
They've been going around and dominating.
They actually won the EasternRegional Silver Sticks
tournament.
It was so funny too, because wewon the final game and I was
walking back to the locker roomand there was someone griping.
They're like they just got beatfrom some team in Kentucky.
Do they even have hockey inKentucky?
That just made me smile.
(08:25):
I'm like putting Kentucky onthe map Ice hockey.
Who would have thought?
Eyvonne (08:29):
You know, there's this great line in the movies, in the
culturally relevant movie SweetHome, alabama, where the main
male lead says honey, justbecause I talk slow it doesn't
mean I think slow and I feellike there are times when I just
have a moment where I'm likeyou know what, don't let the
(08:51):
accent fool you and it's greatwhen you have those moments, and
especially when it's your kid.
How wonderful is that to seeyour kid have an amazing
experience that he probablycan't even realize yet.
What a big deal.
It is right, yeah, that he'llbe able to look back on and be
man.
We did that thing.
That was pretty awesome, wasn'tit.
William (09:13):
Yeah, exactly.
And now it's on to the Canadianteams though.
So we got to travel for theinternational finals up to
Canada in January, and it'sgoing to be a little crazy.
So they're going to be playingsome of the best 10 year olds
out there, and you know you tryand taper, cause I mean, I
thought the regional one Ididn't think we were going to
(09:34):
win that.
You had teams from everywherethat were really good, but these
kids, just they worked so hard.
Yeah, they're all conditionedWell.
They're all conditioned well,they're all taking it, they're
all eating.
Eyvonne (09:52):
good, it's like crazy, uh, like when I was 10 years old
, like there's no way you'd getme to eat on a certain diet.
To prepare, you didn't takeaway your kids halloween candy
for hockey, did you?
William (09:57):
we just very severely limited it, but he wanted to
limit it though.
Eyvonne (10:00):
That's what's cool about it, yeah.
William (10:04):
So there's all that Just travel lined up and I mean
there's been like at theirpractices.
We've had like a different newsagency there, like WHAS, like
Fox, I think, spectrum News,like they've all been like
running little stories on them,but it's like such a big deal
around here anyway, made for TVmovie.
Eyvonne (10:23):
The mayor tweeted out about it.
The governor tweeted out.
William (10:27):
What was that?
Eyvonne (10:28):
Made for TV movie.
Yeah, Like you know, one ofthese great underdogs.
William (10:32):
I should have been documenting this whole time.
You missed it.
Man, I should have beendocumenting, I missed a moment.
Eyvonne (10:38):
Yeah.
William (10:39):
But yeah, Other than that, I have a new.
We have a new addition to thefamily.
Eyvonne (10:49):
She's yes, over three months now.
William (10:49):
Yeah yeah, and let me tell you she's a good baby.
So my, my first two, especiallymy first one, he, oh he,
everything he's really easy now.
He's so good.
But when he was a baby hedidn't want to sleep good, he
didn't want to eat good.
It just nothing was easy,nothing.
And I remember after, after hecame along, I was like I don't
(11:11):
know about having a big family,I think one's good, you know.
But then we had our second oneand she was a little easier.
But this, you know, our newaddition is like, so easy.
She sleeps good.
Like last night she sleptthrough the entire night.
Um, she stocks up during theday.
She just smiles a lot.
She's very patient.
Like she's just such a goodbaby.
(11:33):
Like when I hear all thesefolks in the past talking about
like they're good babies, likeoh, we, I just go to sleep, or
yeah, no, they really don't crythat much.
And I always used to be sojealous, like where is this.
Eyvonne (11:44):
I had one of those yeah .
William (11:46):
Well, you know how it feels then.
Eyvonne (11:48):
Yeah, well, I've had four.
I had one that was, you know,you could lay him down and he
would lay in his crib for likefour hours, like he would sit
there and play for an hour andthen he would lay down, would
startle her awake and, uh, andnap time was over.
(12:10):
So you know it, parenting likeyou get what you get and you
don't throw a fit, that's whatwe tell the, you know the
preschoolers, uh, in theirlittle church classes.
You get what you get when youand you don't throw a fit.
You know the color of popsicleor the candy, and yeah, it's
kind of that way with the kidstoo.
William (12:31):
Yeah, I like that Less control than you realize.
Eyvonne (12:33):
Yeah.
William (12:34):
Venerable, very good.
Eyvonne (12:35):
I may pull that out with my team sometime.
Leadership, yeah, you knowleadership lessons?
William (12:46):
No, I'm probably not, but it's fun to say yes.
Other than that, um, personally, yeah, just busy, and I know
you're busy too.
Eyvonne (12:52):
You've done a lot of conferencing and a lot of you
know public speaking.
You've had some big speakingevents for yourself this year.
Um we did.
We did some travel, so just acombo work, personal trip.
So I've spent some time inLondon and Paris and my husband
and I got to do that, which wasgreat.
We spent a whole year in ournew house.
(13:12):
If you followed the house buildon Twitter, um, a couple of
years ago, so we've been inthere a full year, if you can
believe it.
So, yeah, it's been, uh, it's.
It's, in a lot of ways, areally great year for the Sharps
.
William (13:25):
Yeah, and what a house too, like the way that the
cabinetry and everything is laidout, the shelves, the space
it's yeah, it's very veryoptimized.
Eyvonne (13:36):
I'm really loving my choices, frankly.
Yes, we worked hard and weagonized and we thought a lot
about it and I'm very pleasedwith my choices.
William (13:45):
So yes, it's been good.
If you want ideas, go over toYvonne's place, if you're in the
Louisville area.
Eyvonne (13:51):
ping me and I will show you around.
You'll have to drive an hourand a half to the middle of
nowhere, but I will absolutelyshow you around and cook you a
meal.
When she says middle of nowhere, folks, it is the middle of
nowhere.
William (14:08):
It's very, very much out there over the river and
through the woods that's wherewe are.
Eyvonne (14:11):
Yep the sticks, as they would say.
William (14:11):
Yeah, absolutely beautiful beautiful drive yeah
over the, over the hollerkentucky lingo just that's right
now um.
So what's next?
The uh geopolitical stuff ohman.
Eyvonne (14:24):
So let's talk about what's going on in our world a
little bit, right.
So William and I were kind ofbrainstorming like what are the
big things that happened in 2024?
I wanted to kind of not leaninto like recency bias, but
there's a lot going on in theworld geopolitically whether
you're talking about what'sgoing on in Ukraine and what's
going on with Russia and Chinaand Korea.
(14:47):
And then you look at ournational political scene and it
seems that those things are,have invaded our consciousness,
particularly in the last year,more than they have really for
me in any memory in my adultlife.
They have been front and centermore significantly, as
(15:13):
significantly really, as COVIDwas a few years ago.
So I feel like that's a bigdeal and it's had a big impact
in the last year and willcontinue to have a big impact in
the future.
As we think about what's goingon nationally in the US, with
you know, talking aboutgovernment efficiency, I think
we're going to see downstreamimpacts of that.
(15:35):
And then what's going ongeopolitically?
It's, I think we're going tocontinue to see economic and
real life implications for thatfor us.
What do you think, william?
William (15:47):
Yes, yes, yes and yes.
And I kind of like we weretalking about before, like you
just said, optimization orefficiency and stuff, and we
were kind of just talkingthrough like how much of a you
know our government government'shuge and it's very rigid and
lots of things that have builtup over many, many years.
And of course you know, everyonce in a while I mean even if
(16:10):
you think of, like cleaning yourgarage you let it go a little
too long but you got to go backand you got to like sort stuff
out and organize it.
And you know that's going to beno different for the government
.
It's just it's coming, I think,at a really aggressive pace now
with the new administrationcoming in.
But really at the end of the day, when you go through and you
(16:32):
start undoing things like nomatter, this is not political at
all.
But whoever does it, whoeverstarts digging through and
trying to make changes, you knowit's going to get worse before
it gets better, because you'regoing to break things or it's
going to be a lot of.
You know the whole fenceanalogy.
You know that Yvonne usessometimes.
You know you don't want to teardown that fence until you know
why.
You know, you know why someoneput it up.
(16:53):
So, yeah, I think there's goingto be some major.
I don't want to useconsequences sounds so terrible
because you, you need to go backand you need to make things
more efficient.
Um, you want to make sureyou're using, you know, those
taxpayer dollars.
Eyvonne (17:08):
Maybe unintended outcomes.
There's going to be all kindsof things that could not have
been foreseen.
It is just when you look at ourgovernment and the geopolitical
climate.
It's so complex that there isno way we could quote Gandalf
(17:29):
here and say even the very wisecannot see all ends Like.
That's how I feel about theplace in which we are right now,
regardless of how you know whoyou voted for or what you
believe is right or wrong.
You know, I think we arestaring down a time of, you know
, significant change in manydifferent facets, and it's going
(17:51):
to be a wild ride.
William (17:54):
Yeah, and even like.
So that's thinking completelydomestic.
Well, foreign and domestic, Iguess.
So, like one thing, I I'm goingto have to try and find the
article and dig it out and I'lldefinitely put it in the notes.
But one of these critical, likerare materials that is used in
like chip manufacturing, amongstother things, is gallium.
I did an episode on it like way, way back before I had the
(18:18):
privilege of you joining the pod.
But the problem is a lot ofthat is brought in from China,
so it's imported from China, andyou can see where the conflict
would be here, because you, youknow, I think the goal is to
have less dependency, you know,on other, you know, especially
(18:40):
China, really.
So, like earlier this year Idon't remember, when a huge
deposit of gallium, like reallyhigh um gallium, over like
thousands of acres I think itwas montana or wyoming, it's one
of the two um was found.
So this stuff is bound to besomewhere.
(19:02):
Like the us probably has a ton,of ton of this stuff buried here
and there, of course, but youdon't know where.
Maybe sometimes you stumbleupon it, but it's a the the
deposit they found was extremelyhigh grade, uh, which is really
awesome.
So that'll.
That'll be good, because theless you know, as we're trying
to ramp up chip manufacturinghere in the us, for you know,
(19:25):
more fabs I think I don't knowhow many more were announced,
but there's more in Arizona nowwe're trying to ramp this up and
get on a, you know, sort of aneven level playing field, which
is thinking about the hardwarerequired to do all the things
(19:47):
these days, especially AI.
It's sourcing that hardware andyou know it is an arms race,
like you you had said earlier.
So pretty cool.
Eyvonne (19:55):
We found some cool stuff in the us less dependency
on foreign nations, so yeah yeah, and as, as we talk about, you
know, there's there's thisnational efficiency conversation
going on.
I think we are going to see,you know, continued impacts in
tech based around thatconversation.
I think the notes that we madein advance really were like tech
(20:17):
is growing up in a lot of ways.
You know, we've had 20 years ofsignificant growth in the tech
industry.
And what were these?
You know, scrappy startups thathad created a very new way of
doing things have now they'vechanged the world, they've
become their own economicpowerhouses and we see across
(20:40):
the board you know Google, meta,amazon, I should say Alphabet
you know making.
You know we started seeingefficiencies in headcount in the
last couple years, but I thinkwhat we're going to see is those
organizations making decisionsthat are more driven by the
(21:00):
bottom line than they have beenup until now.
I think we you know that, andto me this isn't even really a
controversial statement, it'sjust an observable fact that
there's going to be, you know,continued focus on how we
measure outcomes and on profit,and I think we're going to see a
focus not on efficiency, notjust, at least, people talking
(21:25):
about it at the federal level,but also talking about it in our
big tech organizations and Ithink you combine that with AI
and I think in a few years techis going to look radically
different than it has looked forthe last decade.
William (21:40):
Good points, and this is such a compounding thing too,
like something.
We didn't talk about this, butsomething I just thought about
was these big companies almosttravel in packs.
You saw it with thework-from-home stuff, amazon.
They came out and they said,hey, everybody's coming back to
the office.
They went through a mountain of, they took the heavy hits as
(22:05):
far as the initial negativepublicity there being the big
elephant in the room.
But then you saw othercompanies coming out and you
know it's like okay, hey, ifamazon's okay with doing this,
and they're doing it and it'sactually working.
Or maybe it's working, maybeit's not, but you know we're
going to try and do it too.
So then you saw other big techcompanies doing it and then even
(22:27):
feeding down into theenterprise space as well.
Eyvonne (22:31):
You know medium to large enterprise, so it's almost
like a cascading effect, whichis kind of interesting, and I
think like there are going to bedownstream consequences from
those changes and it's going totake a while to figure out how
to reoptimize, right.
I mean, I think we're going tosee, you know, less focus,
(22:54):
frankly, on DEI initiatives.
I think we can see thathappening culturally and we also
just see a focus on revenue andwhat that means and how to be
more efficient.
And I think, as we sort of theslope of the curve for tech
(23:20):
growth changes, we're just goingto see a continued focus on
bottom line, on outcomes and onrevenue, and that's going to
have significant impacts for howthose organizations function.
In my mind, that is the naturaltrajectory of organizations
that do what big tech has done.
(23:42):
They've shown up with new ideas, they've changed the world,
they have recalibrated how theworld operates and now they're
going to focus on operationalefficiency.
That's a pretty normaltrajectory and really shouldn't
be a shock, and there will benew industries that come up
behind it that follow a similartrajectory.
(24:04):
We'll have to see how thatplays out.
William (24:09):
We'll have to see how that plays out.
Yeah, it'll be interesting tosee some of the unintended
outcomes.
If you go fully back in theoffice, you're going to lose
some people, and it might besome good people.
So, just of course, mostcompanies, especially the bigger
ones, know there's a lot offolks that are willing to move,
(24:33):
that want those jobs and theywant to fill those seats.
So maybe the impact won't be asfelt as it would be for smaller
companies.
But even then too, like yourcomment about dei, if you think
about a lot of the DEIinitiatives and just social, I
would say social issues ingeneral are much more important
(24:56):
to people today than they wereMaybe not Maybe importance is
not the right word but more outthere in the public and they
mean more especially and they'remore important to the younger
generation as well.
So it'll be interesting to seeany negative outcomes that hit
(25:18):
businesses because of that, asthey're trying to become more
efficient.
Eyvonne (25:22):
I also think we've seen companies have sort of hit the
boundaries of social engagement.
The boundaries of socialengagement and what I mean is
when you have a very broadconstituency, at some point if
you take a strong social standwith a hotly contested topic,
(25:45):
you are alienating a goodportion of your customer base.
And I think we have started tohit some of those boundaries and
started to see the negativeimpacts of that.
And when those things begin tohappen and actually impact the
functioning of the business,you're going to see
organizations say wait a minutelike, do we really want to take
(26:09):
stands that might be divisive,that impact the bottom line?
I think there was a time whenthat was perceived and probably
even demonstrated to be good forthe bottom line, but I don't I
don't know that that's the caseanymore.
It doesn't mean thatorganizations won't have moral
or ethical stands, that theytake, moral or ethical stands
that they take, but I thinkwe're just going to see
(26:36):
everybody back off on choosingsides as much, because that's
played itself out and is turnedis not maybe the benefit that it
was perceived to be at onepoint in time.
William (26:43):
Yeah, and plus, if you like, say you have two different
scenarios or you have twodifferent options or two
different stances you can take.
You know, stance A and stance b.
Well, if you go and you make apublic endorsement of one or the
other, like you were saying,you could lose 30 of your.
You know it could be an anhyzerwas it anhyzer bush or whatever
um beer I think it wasmanufacturer, yeah yeah, they, I
(27:10):
mean they're they.
They lost a ton of.
Yeah, that was just crazy, uh,seeing that happen.
So that's a very real thing.
And yeah, I mean the businessis there to.
I mean, let's be real, you'reyou.
You're beholden to shareholders, you're beholden to a board, or
you know, if you're a privatecompany, like you have all these
(27:30):
different things, and if you'renot growing or at least
optimizing, or if you're doingthings that are risking, you
know, high risk to the, thebottom line and you, you don't
want to end up on the news andpotentially be in the place of
okay, yeah yeah, this companycould literally die.
(27:51):
We could be facing consequencesof doing this for the next 15
years, if we're even stillaround.
And then okay, what was it forthen?
So yeah, remaining completelyneutral or just not having a
public opinion, yeah, I guess itjust makes sense that that
would be the natural progressionat this point.
Eyvonne (28:11):
Yeah, I think that's that's where it's headed is.
Just, you know, organizationsthemselves not taking as many,
you know, stands on culturalissues.
I think that's where we'reheaded.
William (28:26):
And it would be nice to just not see this stuff in the
news anymore, because you have,you know, company A that's like,
ok, we don't like Trump, or wedon't like Biden, or we don't
like conservatives, or you know,we're against X, y and Z, and
then all these people strike andthen you see about it in the
news for a month and then youknow it happens with another
(28:46):
company, but with the alternateside, and you see the same thing
happen and then both sides justcomplain about each other and
that causes discourse.
Eyvonne (28:53):
So it's yeah, I think we're gonna see less of that
yeah, so what is next?
Oh, you know, we can't getthrough a conversation without
talking about ai william oh you,you brought up AI yeah.
Yeah, make me the bad guy.
No, it's.
(29:14):
You know.
I think you know we continue tosee AI conversations increasing
, but I also think that we'restarting to flesh out more where
it fits, where it makes sense.
We've got, you know, ourfriends like John Capobianco out
there talking about how to doAI and networking and how to
(29:35):
really gain efficiency, and Ithink we're beginning to the
world is beginning to understandit a little bit more.
You know there's still a ton ofhype, but we're going to keep
talking about it.
It's it's it's going to beimportant for a long time to
come.
It's, it's, it's, it's.
It's like saying, maybe, thatthere will be a day when we
(29:56):
don't talk about, like, mobilephones anymore.
And it's like you know, they'renever going away.
They're part of our life nowand we're going to be
continually talking about appsand new use cases and how we're
using that technology to changeour lives in the world.
And AI is going to be the samething Like.
If you're hoping that somedaywe're not going to be talking
about it anymore, I'm afraidyou're going to be sorely
(30:19):
disappointed.
William (30:22):
I totally agree, and with that so I I don't know if
you've heard of this company,google, by any chance.
They have a cloud platform Timeor two.
Yeah, google cloud platform, uh, yeah, so they, they developed
so in.
So one thing I wanted to talkabout anyway.
It was like, okay, if I canlook back through 2024, what was
(30:43):
like something that stood outto me as far as like a gigantic
like that stood out to me as faras like a gigantic like wow,
like tech innovation, um, and sothis sort of fits in neatly
with ai as well.
But hold on, I actually copieda summary so I can just read it.
So google has developed a newquantum chip called willow.
(31:05):
I like that reduces errors asit scales up a major
breakthrough in quantum errorcorrection.
Willow also performed acomputation in under five
minutes that would take asupercomputer 10 septillion
years, demonstrating itspotential for solving complex
(31:26):
problems beyond the reach ofclassical computers.
This achievement marks asignificant step towards
building commercially relevantquantum computers that can
revolutionize fields likemedicine, energy and ai.
And this is just such a coolthing, you know, because if you
think of like we were talkingabout, I think before we hit the
(31:50):
record button, but this quantumcomputing thing is really kind
of a crazy paradigm to wrap yourhead around.
So in the traditional sense,like a classical computer,
everything is ones and zeros.
So things are processed in asense through these things
(32:12):
called logic gates withconditional type things.
So these gates basically takeup to two bits you know one or a
zero is like inputs and go intothe machine and then they
return a single bit as an output.
And when you think of likequantum, there's these things
called qubits, like q, quantumbits, and it's I don't know if
(32:35):
the right word is three, threedimension.
I wouldn't be three-dimensional, it doesn't make any sense.
Um, it's like a uh, like atwo-dimensional vector of of
numbers, like real numbers.
So basically it can hold a lotof information, you know, rather
than just a single bit.
But the problem then becomeserroring out, because if you
(32:58):
think of like an individualqubit, it's almost like you know
.
I heard AI described as like agame of probabilities a while
back.
There's a talk on YouTube AgainI'll try and go back and find
it, but it was a really good one.
But this is almost like adistribution of probability, in
a sense, and averages.
(33:18):
So because you don't have aconcrete value and then you have
, you end up getting into athing of like the observer
effect as well, and this is whyin the article they mention like
the multiverse.
There's a little blurb in there.
Eyvonne (33:36):
Can I read that sentence?
William (33:37):
Please do.
Eyvonne (33:38):
It's talking about the size of the numbers and the
speed of the computation, and itsays this mind boggling number
exceeds known time scales inphysics and vastly exceeds the
age of the universe.
It lends credence to the notionthat quantum computation occurs
in many parallel universes, inline with the idea that we live
(33:58):
in a multiverse, a predictionfirst made by David Deutsch.
So this is next level,otherworldly, literally, kind of
discovery.
William (34:14):
That is a bold, crazy sounding statement.
That is a bold, crazy soundingstatement.
Yes, machine, uh, if you will,um, each of the, the, the qubits
, like the, whatever 64 qubits,they can all kind of like talk
(34:36):
and communicate through thisphenomenon, again known as, like
uh, quantum entanglement.
And that's where you see,honestly, like I was watching
this new uh tv show like thatcame out a few months ago called
dark matter really cool show.
But, yeah, it goes into like,it kind of like takes everything
we're talking about and it'sfrom a book, um, first, and then
(35:00):
the book.
I actually went back and readthe book.
The book's so much better thanthe show.
So, if you're listening, readthe book or listen, listen to
the book, great book.
But it kind of goes in andtakes a lot of these principles
and applies it into a reallycompelling and engaging and
awesome uh narrative um, from a,an author named blake crouch, I
(35:20):
think his name is.
Eyvonne (35:22):
So, yeah, very cool the the other thing to think about,
and and we didn't talk aboutthis in our initial discussion,
william but as you think aboutthe rise of quantum computing,
you've got to think about PQC,which is post-quantum
cryptography, and what quantumcomputing would mean for our
(35:43):
current cryptographic systems.
Cryptographic systems,everything that we have right
now is secured by cryptographythat could potentially be broken
by quantum computing, becauseyou suddenly have I'm going to
say infinitely it's notcompletely accurate, but
(36:05):
seemingly infinitely morecomputing capability, and you
think about everything fromBitcoin to SSL transactions.
What does PQC mean for that andwhat are the long-term impacts
and how do we have to evolve oursecurity measures in light of
(36:26):
this new computing power?
It's really it's a lot to thinkabout and and it's going to be
an interesting evolution towatch over the next few years.
As you know, quantum computingcan begins to become a true
reality at a scale beyond.
You know what Google can provein their lab, right.
William (36:46):
Yeah, and like back to that summary, like it was.
It was like it was like takingsteps towards building
commercially relevant quantumcomputers.
Cause right now it's like I seeall this game breaking research
with quantum computers over thepast five years and it's mostly
like nothing.
Nothing burgers.
(37:07):
There are a few, you know, ibmand so forth that have actually
built these things out.
But when you think of likecommercially relevant, you know
it's like okay, this technologycan be used and has an impact on
the rest of us and that's wherethis is going.
This is just one of those stepsin this big staircase.
Eyvonne (37:25):
But it always starts in a lab somewhere right Big
staircase but it always startsin a lab somewhere right, like
there's always an initial proofpoint and there are theories,
and then there's the practicalyou know, the proof point, the
application in a lab, and thenthere's the commercialization of
that technology, and so we'regoing to see that trajectory
play out in the future and it'sgoing to have significant
(37:48):
impacts on computing at scale.
All of it, you know, whetherwe're talking about our phones
or our communications.
William (38:01):
You think about the reliance that the industry has
on encryption today, about whatthat's meant from a regulatory
standpoint and what happens toall of that when we have quantum
computing power at scale canyou imagine if there was just
some scientific medic or justsome crazy scientific
breakthrough where all thepieces fit together and it was
(38:23):
like, okay, we got to get thisstuff out as quick as possible.
Can you imagine the impacts onso many different things that
would have?
Everything?
Would change everything.
This is mainstream yes, it's.
Eyvonne (38:36):
Uh, I mean, I think we'll get there now, whether it
will be slow and incremental orwhether it will be like a flag
day, kind of a you know crowdstrike moment where we all wake
up and nothing works anymore.
Uh, who knows, right?
Um, and actually that's anotherthing that should have made the
2024 list, but I digress.
William (38:54):
Oh yeah, I just wonder.
You know, honestly, the bigquestion here to really ask
yourself is once this does hitmainstream, will IPv6 be
completely adopted or not?
I think that's the big question.
Ipv6 be completely adopted ornot?
I think that's the big question.
Eyvonne (39:10):
No, it's going to be like the year of the Linux
desktop.
We'll talk about it forever.
William (39:19):
Yeah, good times.
I think we have something else,don't we?
Oh, yeah, a couple more.
Eyvonne (39:27):
A couple more.
We would be remiss if we didn'ttalk about the X data center
and what Musk has done to builda pretty incredible data center
in less than a month that doessome pretty amazing things.
William (39:47):
Yeah, no, you're absolutely right.
It's even got a fierce namelike the Titan or it's going to
bother me now Colossus.
I think that's what it isColossus or Titan one of those
two, the Colossus supercomputer.
Eyvonne (40:07):
that's the word you're looking for.
William (40:09):
Yeah, so it's just um, it's crazy.
So when you think about the,the cost of like one, um h100
and they, they have likehundreds of thousands of these
things and and that's basicallybeing used to train um X AI
model, which they call.
Eyvonne (40:29):
ROC.
William (40:32):
And just the time I mean thinking about.
I mean that's just such a shortwindow of time to actually get
the hardware shipped in.
How do you power all of it,which I think they're still
powering a lot of it throughgenerators that they have
running, that they're shippingin a lot of gasoline for?
Eyvonne (40:51):
Racking and stacking, cabling, the logistical bits of
it, right, I mean, and that'sthat's, frankly, like the easy
part, like the, the architecture, the distributed configuration,
all of that, and it's not andit's not something that's been
done before, so it's not likethey're established
(41:12):
architectural patterns that youcan follow it.
William (41:15):
it really is an incredible technological
achievement yeah, even whenjensen's talking about it, like,
yeah, we us at nvidia, wedidn't think this was possible
and yeah, then someone else justcomes and does it.
You know, and part of thereason they did very
unconventional things, that Inever would have thought that
(41:36):
you would do.
So it's either in Finna band orethernet kind of a thing, rdma,
but like they went ahead androlled, I think, their own
infrastructure, I think they umas they were building it out
like and they're trying to scaleout to all these, these GPUs,
they actually moved an offloadedTCP and into hardware, which is
(41:59):
a very interesting, you know,on the surface, but when you
think about the commoditizationof some of this hardware and
like trying to reach that pointof, okay, we can scale as we
need to, we can just keep adding, adding racks, adding racks,
adding racks and really getting,you know, ramping up, like that
Um, like, if you go through andyou actually read, there was a
(42:22):
presentation from the um, the umhot chips 2024, in which I
think I don't know if it's teslaor whoever um presented at just
as far as like how they viewsome of the things they're
coming out with and I don't knowlike when you I know that the
(42:43):
offloading of tcpip or tcp to Iknow that they did that for
Tesla what they use to trainTesla's um, all these names, not
, it's not claw dojo, they callit, they call it dojo.
So what they do there may bedifferent than how they've
scaled up the um, the datacenter in Tennessee.
(43:04):
I'm actually kind of curiousabout that.
I'll try and find something.
But just these are like verymuch off, very unorthodox
approaches and ideas thatapparently nobody else thought
of, or they would have beendoing it.
Somebody else would havealready done this and nvidia
would have probably said, hey,you should do this, because hey
(43:26):
you what?
We're going to sell more GPUsif you can scale to a higher
number, you know, so it's a veryinteresting space.
Eyvonne (43:33):
The thing that we see with Musk is he has a very
different risk profile than mostorganizations.
He is concerned more aboutaccomplishing the thing than he
is about the profitability ofthat, which, interestingly
enough, has been a prettyprofitable approach for him,
(43:56):
right, apparently yeah.
But he, you know, hey, I want todo this, we're going to do this
.
What's it take to make ithappen?
Screw all the objectiverealities we think we know, like
, throw those out the window.
Let's start at the beginning.
Let's say we're going to dothis, now what do we need to do
(44:17):
to get it done?
And I think that approach hasbeen incredibly effective.
Now there's collateral damagewhen you take that approach that
we've all seen.
That can be true, and at thesame time, this is an incredible
technological feat.
There's no other way to see it.
William (44:38):
Two things can be true at the same time Always, yep.
Eyvonne (44:43):
And in that vein, really the last observation that
we have for 2024 is just thecontinued rise of a new hardware
arms race.
Right, we have seen NVIDIA isstill on top from an earnings
standpoint.
There's a great conversationwith Jensen, the CEO of NVIDIA,
(45:05):
basically saying you know, wehave a situation in which, even
when our competitors make theirproduct free, the cost of
ownership of ours is still lowerbecause it is so efficient and
effective.
And I think you know NVIDIAunlocked a ton of potential and
capability there.
And so we see this new hardwarearms race.
(45:26):
We see NVIDIA doing well, wesee AMD doing well, we see
Broadcom doing well and we seeIntel struggling.
Any comments on the hardwarearms race before I talk a little
bit more about Broadcom,William.
William (45:44):
No, it is an arms race.
It is an arms race, I mean,when you think of the
manufacturing, the, the amountof just land, the amount of
engineering, the amount of, youknow how complicated the supply
chain is.
And I mean even you knowbroadcom.
You mentioned broadcom they'remaking don't they make gpus now.
(46:04):
I think they make gpus now,which I didn't know.
I think I read that.
I'll have to double check, butyeah, it is definitely an arms
race and I think, the ability tohave more efficient, faster,
more computing power.
If you think of AI as like anation-changing technology.
(46:27):
Ai is like a nation-changingtechnology.
Whoever has the most powerfuland innovative infrastructure
that runs it is going toobviously have some sort of
advantage on the global stage.
Eyvonne (46:40):
And that has a lot of impacts.
It makes hardware cool again.
It makes power incrediblyimportant.
It makes power incrediblyimportant.
This wasn't officially on ourlist, but we see the rise of
(47:06):
hyperscalers, investing innuclear and new you know, small
form nuclear and new ways topower, where we were talking
about the effectiveness of GPUsto run AI workloads, and one of
the metrics that we monitor isnot just the utilization of that
GPU but its temperature,because if the temperature is
(47:29):
higher on the chip, we know thatmeans it's working harder and
it's more fully utilized, and sothere's actually things you can
learn about monitoring theperformance of that chip.
It is, how hot is it runningand are we getting the most out
(47:57):
of it that we could.
So it's a whole new world ofunderstanding efficiency, how
best to utilize that hardwareand also how to keep all those
GPUs in sync, because it's notlike packet delivery right,
where you send a bunch of themand they get there when they get
(48:18):
there and then it getsdecomposed on the other side.
That's not how GPU processingworks.
And the other observation I'llmake in about 2024 is with
Broadcom.
In many ways it was the best oftimes and it was the worst of
times the day we're recording.
(48:38):
We saw a huge jump in Broadcomstock price because of their
recent earnings report and, atthe same time, we have customers
reeling from the acquisitionBroadcom has done of VMware and
how they are going to plan forthe future of their VMware
infrastructure.
And I think Broadcom is makingsome efforts to roll back some
(49:01):
of those changes and to be alittle bit more customer
friendly.
I think the thing that we'regoing to find out next year is
whether it's too little, toolate.
Do they do irrecoverable damageto the VMware business?
I think it's different, as weit will be a different
organization than we've everknown it to be.
(49:21):
But what's what's?
What's the future of VMwaregoing to look like coming into
2025?
And where it's going to looklike coming into 2025.
And so Broadcom, for me, is oneof the companies that probably
generates the most cognitivedissonance, because we're seeing
it's just puzzling to see whatthey're doing, both on the
(49:44):
hardware side and on thesoftware side with their
acquisitions and try to puzzleout exactly what the future
looks like with thatorganization yeah, it's wild,
like, yeah, I mean even like, Ithink one I mean vmware was like
channel, channel first, rightum, so just they started they
(50:06):
opened up the wholevirtualization.
you talk about innovation.
Like you know, virtualization,you talk about innovation.
Virtualization really didn'texist at any kind of scale until
VMware, and so, in a lot ofways, everything that we're
doing in our hyperscalersabsolutely is building on their
shoulders.
William (50:24):
Yeah.
Eyvonne (50:25):
Conceptually.
William (50:26):
I remember so I was in tech.
That was actually when I firststarted getting into tech Well,
it was a little after but thewhole physical, the virtual
migration thing was gigantic.
I remember sitting on a CTOtown hall and it was just like
the coolest thing, where theywere like, yeah, we have this
much physical servers today,this time next year, this
(50:50):
percentage of that is going tobe completely virtualized.
So you had this whole numbersgame.
But then you started seeing,like you started seeing it
happening.
That's how I got into vmwareand it was just a huge, a huge
game changer and you think ofhow go ahead?
Eyvonne (51:05):
well, I was going to say it was an incredibly
necessary technology when ithappened in the early 2000s,
because organizations had builtout data centers, they had
expanded to the max capacity oftheir floor space.
Either they couldn't get morepower, which is very similar to
what we're seeing, or they justdidn't have the floor space to
(51:28):
put more gear of what we'reseeing, or they just didn't have
the floor space to put moregear.
And I mean I remember workingfor an organization that was
like we have reduced our entiredata center footprint down to 12
floor tiles right byvirtualization and by, you know,
implementing this consolidation, and the financial benefit for
(51:50):
that was incredible, I think,you know, and that kind of
innovation isn't over.
It's just going to happendifferently, but it was.
I mean, it was huge.
You just can't hardly quantifywhat a big deal that was.
(52:14):
No-transcript Across the wholeworld and I just checked,
broadcom's stock is up 24% todayWow, despite all of the
craziness with VMware.
So you know, regardless of howyou feel about Hawk, his
(52:35):
business from a numbersperspective is through the roof.
William (52:42):
Yeah, it's interesting times interesting times and I
think if they dial back, I meanif they're showing, I mean of
course they're gonna makechanges if it's affecting their
bottom line.
But you know, ripping out thechannel game and then saying,
hey, we're gonna direct sell allthis stuff, definitely, I mean
I, anybody, most people thatwork in this industry could have
said, hey, that's not a goodidea.
But I think they're trying toreverse that now and they're
(53:06):
trying to do some reversals ofsome of the bigger oopsies.
Because, hey, vmware has greatcompetition out there, the
technology is out there.
You've got Proxmox, which is afantastic hypervisor and all
that it takes is a lot of thosealienated partners to wrap up
and scale up their businessmodel to start, and then it
(53:28):
turns into a rising tide, raisesall ships thing because the
partners are winning.
You have some sort of things onthe Proxmox side.
More money is going to go intoProxmox, making it a better
product, licensing and such, andit kind of would almost become
the new VMware in a sense.
Eyvonne (53:48):
So yeah, and I'm biased here as a former VMware person,
but I will still argue that foryou know, enterprise
on-premises virtualizationtechnology, there are
competitors out there, but asfar as a holistic solution I
(54:10):
don't think there's anybody whocan still hold a candle to
VMware today.
But they are gaining and I thinkthat's going to continue to
happen today, but they aregaining and I think that's going
to continue to happen and, withpublic clouds, very competitive
.
That's a different paradigm.
Sure, you can run VMware inpublic cloud.
Lots of folks do it.
(54:32):
I talk a lot about it.
Google has a great solution forit, for example.
But we're also just seeing ahuge paradigm shift from the
center of gravity not beingOS-based virtualization hosts,
it's container-based workloadsor it's other higher order you
(54:53):
know, platform services, and soyou know I think that shift is
as the industry matures, assoftware continues to be
developed.
That's the direction theindustry is moving Full stop.
You know there are still goingto be VM workloads for decades,
just like there are stillmainframes out there doing their
(55:13):
thing.
I mean, I work ininfrastructure and so a ton of
my career is still centered onthose kinds of workloads and
they're incredibly important.
At the same time, you know thetime stops for no man or woman
and things are going to continueto march forward.
William (55:31):
Yeah, I just I hope mainframes don't go away in my
lifetime.
They're so pretty, the zerodown times from IBM.
They're beautiful inside.
It's like looking into OptimusPrime, like the Matrix, like
those things are so cool lookingbut I would never want to run
them anymore.
Eyvonne (55:49):
Okay, all right, you appreciate them for their
aesthetic beauty.
William (55:55):
And they do crunch some major numbers.
They do.
Eyvonne (55:58):
Well, and you know we're back to the what's the
business justification forreplacing this thing, right?
And I know of more than oneorganization in which lots of
people had resume generatingevents because they planned to
(56:19):
replace a mainframe and itdidn't work out.
They spent a ton of money,found out that they couldn't
really do it and then they endedup, you know, staying on one,
and I'm not saying that that'salways the right choice, because
it's not.
Sometimes you absolutely haveto modernize, but I think it's
smart to say, okay, what's thisreally going to cost, what's
(56:40):
it's really going to take, anddoes it really make sense for us
to put all of our efforts here,or should we put that
innovation effort somewhere else?
And it's a smart leader,technical leader, who does that
for their business and their usecase, because there's no one
right answer.
I know organizations that arerebuilding everything to replace
(57:02):
mainframes that are beingwildly successful both
technically and in themarketplace, and I know
organizations that are keepingtheir mainframes that are being
wildly successful and innovative, both technically and in the
marketplace, right?
So there's no one answer forthat question.
And I think we're going to seethe same thing play out for VM
workloads.
We're going to see likeeventually there's probably
(57:25):
going to be a delivery mechanismother than a container.
I know I'm insulted.
William (57:35):
No, just kidding.
No, you're right, though youknow it's just going to keep on.
I mean it's funny becausecontainers are kind of like re.
You know they're modernizedjails of old.
You know they're kind of like akind of the same technology.
You know it's the whole.
You know rule 11 thingrepurposed, reoptimized for a
(57:56):
new world and new workloads andnew application designs and
things.
Eyvonne (58:01):
It's a different, more contained abstraction that can
be more managed and deployed incode, and it provides more
consistency and less dependenceon the underlying infrastructure
, including the operating system.
Right, um, and and that's agreat thing like we love
(58:24):
abstractions right, we just wantto be sure we're using the
right ones we.
William (58:29):
You know what we ought to do.
I know we probably need to tieup, but where we need to get a
cobalt developer.
We need to get a cobaltdeveloper on here to kind of
discuss, or somebody that's veryin tune with mainframe
architecture and the tie-ins toinfrastructure and all the like.
What, what are the?
What would it actually requireto get rid of mainframes?
(58:51):
Because they're still in thedna of a lot of our core
infrastructure and cobaltdevelopers are aging out?
Eyvonne (59:00):
I would also love to see somebody do a.
A.
There's a word that I'mthinking, that I'm that I want
to use, that I can't pullmagical musical chairs?
Yeah, no no, you're not helpingshoe manufacturer.
No, you're not helping.
Shoe manufacturer, I would loveto do a discovery of all of the
(59:25):
abstractions throughout thehistory of computing, like on a
timeline, right, you've gotbinary, you've got assembly, and
then on top of assembly you'vegot your programming languages,
and then on top of yourprogramming languages, you've
got your compilers andinterpreters, and then there's a
code level abstraction, andthen you've got like the
(59:47):
hardware abstractions, right.
I think it would be fascinating.
William (59:52):
Because then when my stuff moves to software, they,
you know, offload certain piecesof it to hardware again and
then like it's yeah, stuff, thatwould be a great conversation.
I'm down, yeah, if, uh, theworld is listening, if you know
anybody that wants to have thatconversation and is an expert it
takes about like five peopleand then you can talk about
networking abstractions.
Eyvonne (01:00:12):
Right, a whole nother kind of abstraction.
You start with the bits on thewire and then you talk about
protocols and encapsulation andthen encryption and then, yeah,
the osi model like it's.
There's so much fun stuff youcan talk about yeah, that sounds
like a nice round table that'sa project for 2025, william
(01:00:32):
sounds good.
William (01:00:33):
Well, we better wrap up .
We went over the normal time,but this was a lot of fun, a lot
of unplanned, just off the cuffconversation, and I just want
to extend, you know, happyholidays to all of you.
You know, no matter what you'redoing, try to find time to
relax.
Um, if you're in cold climates,find a good warm tea or hot
(01:00:54):
chocolate, uh, whatever, relax,read a book, you know.
Get off the screens for alittle while, maybe do something
, play some board games.
Eyvonne (01:01:04):
Watch a Christmas movie .
William (01:01:06):
There you go.
A Christmas Story is myall-time favorite.
I love that movie.
Eyvonne (01:01:11):
It's a Wonderful Life is one of those nostalgic
favorites, and I might be asucker for a Hallmark Christmas
movie or two.
I can't watch too many of them,but every now and then there's
something comforting about thatand you can cringe all you want.
It's fine, go watch people blowstuff up.
William (01:01:32):
Yeah, I mean those Hallmark movies are great,
though, like if I'm reallyhaving a hard time sleeping I
can just crank one on.
Eyvonne (01:01:40):
I have a few of those too.
Yeah, hey, there's nothingwrong with that nice, nothing
wrong with a predictable storywhen you know what it is and you
need just to feel good for aminute.
William (01:01:51):
Yeah, yeah, all right whatever you got to do to
naturally generate that dopamineright there you go, exactly, go
, exactly, all right.
So, yeah, thanks for the time,merry Christmas.
Eyvonne (01:02:04):
Happy Hanukkah.
William (01:02:05):
Yeah, enjoy, enjoy, enjoy.
Eyvonne (01:02:06):
Enjoy yeah.
My kid had a school event last night and I got I kid
you not four texts and an emailsaying that the kids had to be
there at 530 to be ready for theprogram at six.
We pull into the parking lot at520 and the doors are still
locked and there's a line ofpeople around the school.
(00:21):
I'm like, and it's, and it'sfreezing cold outside and I have
this weird thing where myfingertips turn white when they
get cold, like my bloodliterally stops circulating in
the tips of my fingers and I'menraged.
William (00:35):
I need to take a vacation to Canada during the
winter.
Welcome to another episode ofthe Cloud Gambit.
We are approaching the end of2024, and this is actually going
(00:59):
to be the last episode goingout for 2024.
I am William your host episodegoing out for 2024.
I am William, your host, butwith me I have someone that is
so serious about the holidaysthat they put Christmas
decorations up before the turkeyis pulled out of the oven on
Thanksgiving.
That is Yvonne Sharp.
How are you doing today, Yvonne?
Eyvonne (01:19):
Hello, and absolutely I fully own the pre-decorating in
my defense.
We were trying to have aholiday party uh, that that
preceded a lot of the otherholiday parties, so the people's
schedules would be more openand could make it.
I'm not sure that it waseffective, though, because
(01:41):
people are busy, everybody'sbusy now.
We had 25 30 people in thehouse that day and we had a
great time.
Um, you were at your kids hockeytournament, so you didn't make
it, which I will try to forgiveyou for that, but uh yeah, we
had all of our decorations uptwo weeks before thanksgiving in
my house that's very, very.
William (02:04):
You know what?
This is one of my first piecesof holiday decor.
I'm just going to wear thisthing.
Eyvonne (02:15):
I should have brought a holiday hat.
I didn't even think about it.
It's not even the regular one.
William (02:19):
It's checkered.
It's like a flannel.
Eyvonne (02:22):
It's perfect for a Kentuckian.
Exactly there we go.
I just need the actual flannel.
I look like a flannel.
William (02:24):
Hey, it's perfect for a Kentuckian.
Eyvonne (02:25):
Exactly there we go.
William (02:26):
I just need the actual flannel.
I look like a lumberjack.
Give me an axe and some wood tochop.
Yeah, that's great Gotcha.
So, anyhow, so this episode.
So Yvonne and I were talkingand we're like you know what.
It's the end of the year andthis has been a crazy year.
A lot has happened, bothpersonally, professionally, you
(02:51):
know, the world markets.
Everything seems to be a littlecrazy.
So we thought, you know what,why not have an episode and just
kind of look back and just takea time to reflect?
Eyvonne (03:02):
you know, it's our retrospective Love it, so that
is what we are time to reflect.
William (03:05):
It's a retrospective.
Love it.
So that is what we are going todo.
Do you want to kick us off,Yvonne?
Eyvonne (03:12):
Yeah, we thought we'd start with a few personal
updates.
So things have been busy in myworld.
I think most of you hopefullyhave noticed I've been trying to
focus more on creating externalpublic facing content.
So part of that is joining theCloud Gambit with William so
(03:34):
that happened mid-year, soexcited to be here and been
reconnecting with some folks onLinkedIn, been hanging out a
little bit more on X and BlueSky.
So, yeah, I've been focusing onthat and I think for me, I've
also been working professionallyat Stretching and Growing as a
(03:55):
Leader, trying to think moreholistically and to also
navigate times of extreme change.
And how do you lead peoplethrough that and provide enough
guidance so that they know whatto do, but not enough, not so
much that you're constrainingand hindering creativity, and so
(04:16):
I've been thinking a lot aboutthose things.
What's been up in your world,william?
William (04:22):
Well, so first of all, before I talk, talk about my
world.
I just want to get yourperspective on something.
So you're talking about likegrowing as a leader and you have
, you know, direct reports andyou're working for, you know, a
big company.
And one thing just myobservations from in the past
you know having like led teamsbut also being in like the
(04:43):
engineer focus, like when you'rean engineer or you're doing any
of the boots on the ground work, you get so focused, of course,
on details.
You get so focused on all thelike very granular maybe tunnel
vision sometimes.
But when you sort of move upthat org chart a little bit,
you've got to like widen thataperture and you have to like
(05:04):
zoom out and you really needboth of those things.
And I remember, just when I wasin that engineering mindset and
I was doing a lot of the work,that used to drive me nuts
because I would always think youknow what these leaders.
They just don't know whatthey're talking about, they
don't understand, they don't seewhat I'm seeing, they don't
know how important this onelittle thing is.
(05:24):
But then as you move up and youget that bigger picture and
that bigger vision and you startseeing all the things you know,
it really changes yourperspective.
But any thoughts on that.
Eyvonne (05:35):
Well, I do think you know, the more your you know
your aperture increases and thewider the breadth of your
responsibility as a leader, itdoes shift your perspective, I
think.
For me it's trying tocontinually rethink.
You know what is important forme to focus on as the leader.
(05:59):
What is it that I need to besure that my folks have the
freedom and, frankly, theaccountability to focus on as
individual contributors.
And then how do I be sure Ihave an effective feedback loop
so I'm getting and understandingwhat are the real boots on the
(06:19):
ground problems?
And then how do I translatethose to the broader
organization to be sure that myindividual contributors can be
effective, right?
And so it's continually tryingto understand what's going on
enough but then also framingthat effectively so that you can
(06:41):
get organizational buy-in toenable your people.
And it's a very hard thing to do.
And I think sometimes it's as Isee we can not really grasp the
luxury that we have to focusdeeply on a thing and completely
deeply understand that thing tothe exclusion of everything
(07:05):
else.
Because when you're a leaderyou don't have that, in a lot of
ways, privilege anymore to justdeeply focus on a thing.
You've got competing prioritiesthat you have to think about
different stakeholders withinthe org.
Who who's who's.
You know who you need tosatisfy um both just
(07:28):
existentially and for thebusiness.
So it's it's.
It's a very different um, it'sjust a very different job yeah,
yeah, yeah, that was.
William (07:40):
you said that perfectly .
I doubt that Better than Iwould have anyway.
So, yeah, personal here, asidefrom traveling for work and
stuff, my son plays travelhockey and they are crushing it
(08:01):
this year.
Eyvonne (08:02):
So he's 10 and under.
William (08:03):
They've been going around and dominating.
They actually won the EasternRegional Silver Sticks
tournament.
It was so funny too, because wewon the final game and I was
walking back to the locker roomand there was someone griping.
They're like they just got beatfrom some team in Kentucky.
Do they even have hockey inKentucky?
That just made me smile.
(08:25):
I'm like putting Kentucky onthe map Ice hockey.
Who would have thought?
Eyvonne (08:29):
You know, there's this great line in the movies, in the
culturally relevant movie SweetHome, alabama, where the main
male lead says honey, justbecause I talk slow it doesn't
mean I think slow and I feellike there are times when I just
have a moment where I'm likeyou know what, don't let the
(08:51):
accent fool you and it's greatwhen you have those moments, and
especially when it's your kid.
How wonderful is that to seeyour kid have an amazing
experience that he probablycan't even realize yet.
What a big deal.
It is right, yeah, that he'llbe able to look back on and be
man.
We did that thing.
That was pretty awesome, wasn'tit.
William (09:13):
Yeah, exactly.
And now it's on to the Canadianteams though.
So we got to travel for theinternational finals up to
Canada in January, and it'sgoing to be a little crazy.
So they're going to be playingsome of the best 10 year olds
out there, and you know you tryand taper, cause I mean, I
thought the regional one Ididn't think we were going to
(09:34):
win that.
You had teams from everywherethat were really good, but these
kids, just they worked so hard.
Yeah, they're all conditionedWell.
They're all conditioned well,they're all taking it, they're
all eating.
Eyvonne (09:52):
good, it's like crazy, uh, like when I was 10 years old
, like there's no way you'd getme to eat on a certain diet.
To prepare, you didn't takeaway your kids halloween candy
for hockey, did you?
William (09:57):
we just very severely limited it, but he wanted to
limit it though.
Eyvonne (10:00):
That's what's cool about it, yeah.
William (10:04):
So there's all that Just travel lined up and I mean
there's been like at theirpractices.
We've had like a different newsagency there, like WHAS, like
Fox, I think, spectrum News,like they've all been like
running little stories on them,but it's like such a big deal
around here anyway, made for TVmovie.
Eyvonne (10:23):
The mayor tweeted out about it.
The governor tweeted out.
William (10:27):
What was that?
Eyvonne (10:28):
Made for TV movie.
Yeah, Like you know, one ofthese great underdogs.
William (10:32):
I should have been documenting this whole time.
You missed it.
Man, I should have beendocumenting, I missed a moment.
Eyvonne (10:38):
Yeah.
William (10:39):
But yeah, Other than that, I have a new.
We have a new addition to thefamily.
Eyvonne (10:49):
She's yes, over three months now.
William (10:49):
Yeah yeah, and let me tell you she's a good baby.
So my, my first two, especiallymy first one, he, oh he,
everything he's really easy now.
He's so good.
But when he was a baby hedidn't want to sleep good, he
didn't want to eat good.
It just nothing was easy,nothing.
And I remember after, after hecame along, I was like I don't
(11:11):
know about having a big family,I think one's good, you know.
But then we had our second oneand she was a little easier.
But this, you know, our newaddition is like, so easy.
She sleeps good.
Like last night she sleptthrough the entire night.
Um, she stocks up during theday.
She just smiles a lot.
She's very patient.
Like she's just such a goodbaby.
(11:33):
Like when I hear all thesefolks in the past talking about
like they're good babies, likeoh, we, I just go to sleep, or
yeah, no, they really don't crythat much.
And I always used to be sojealous, like where is this.
Eyvonne (11:44):
I had one of those yeah .
William (11:46):
Well, you know how it feels then.
Eyvonne (11:48):
Yeah, well, I've had four.
I had one that was, you know,you could lay him down and he
would lay in his crib for likefour hours, like he would sit
there and play for an hour andthen he would lay down, would
startle her awake and, uh, andnap time was over.
(12:10):
So you know it, parenting likeyou get what you get and you
don't throw a fit, that's whatwe tell the, you know the
preschoolers, uh, in theirlittle church classes.
You get what you get when youand you don't throw a fit.
You know the color of popsicleor the candy, and yeah, it's
kind of that way with the kidstoo.
William (12:31):
Yeah, I like that Less control than you realize.
Eyvonne (12:33):
Yeah.
William (12:34):
Venerable, very good.
Eyvonne (12:35):
I may pull that out with my team sometime.
Leadership, yeah, you knowleadership lessons?
William (12:46):
No, I'm probably not, but it's fun to say yes.
Other than that, um, personally, yeah, just busy, and I know
you're busy too.
Eyvonne (12:52):
You've done a lot of conferencing and a lot of you
know public speaking.
You've had some big speakingevents for yourself this year.
Um we did.
We did some travel, so just acombo work, personal trip.
So I've spent some time inLondon and Paris and my husband
and I got to do that, which wasgreat.
We spent a whole year in ournew house.
(13:12):
If you followed the house buildon Twitter, um, a couple of
years ago, so we've been inthere a full year, if you can
believe it.
So, yeah, it's been, uh, it's.
It's, in a lot of ways, areally great year for the Sharps
.
William (13:25):
Yeah, and what a house too, like the way that the
cabinetry and everything is laidout, the shelves, the space
it's yeah, it's very veryoptimized.
Eyvonne (13:36):
I'm really loving my choices, frankly.
Yes, we worked hard and weagonized and we thought a lot
about it and I'm very pleasedwith my choices.
William (13:45):
So yes, it's been good.
If you want ideas, go over toYvonne's place, if you're in the
Louisville area.
Eyvonne (13:51):
ping me and I will show you around.
You'll have to drive an hourand a half to the middle of
nowhere, but I will absolutelyshow you around and cook you a
meal.
When she says middle of nowhere, folks, it is the middle of
nowhere.
William (14:08):
It's very, very much out there over the river and
through the woods that's wherewe are.
Eyvonne (14:11):
Yep the sticks, as they would say.
William (14:11):
Yeah, absolutely beautiful beautiful drive yeah
over the, over the hollerkentucky lingo just that's right
now um.
So what's next?
The uh geopolitical stuff ohman.
Eyvonne (14:24):
So let's talk about what's going on in our world a
little bit, right.
So William and I were kind ofbrainstorming like what are the
big things that happened in 2024?
I wanted to kind of not leaninto like recency bias, but
there's a lot going on in theworld geopolitically whether
you're talking about what'sgoing on in Ukraine and what's
going on with Russia and Chinaand Korea.
(14:47):
And then you look at ournational political scene and it
seems that those things are,have invaded our consciousness,
particularly in the last year,more than they have really for
me in any memory in my adultlife.
They have been front and centermore significantly, as
(15:13):
significantly really, as COVIDwas a few years ago.
So I feel like that's a bigdeal and it's had a big impact
in the last year and willcontinue to have a big impact in
the future.
As we think about what's goingon nationally in the US, with
you know, talking aboutgovernment efficiency, I think
we're going to see downstreamimpacts of that.
(15:35):
And then what's going ongeopolitically?
It's, I think we're going tocontinue to see economic and
real life implications for thatfor us.
What do you think, william?
William (15:47):
Yes, yes, yes and yes.
And I kind of like we weretalking about before, like you
just said, optimization orefficiency and stuff, and we
were kind of just talkingthrough like how much of a you
know our government government'shuge and it's very rigid and
lots of things that have builtup over many, many years.
And of course you know, everyonce in a while I mean even if
(16:10):
you think of, like cleaning yourgarage you let it go a little
too long but you got to go backand you got to like sort stuff
out and organize it.
And you know that's going to beno different for the government
.
It's just it's coming, I think,at a really aggressive pace now
with the new administrationcoming in.
But really at the end of the day, when you go through and you
(16:32):
start undoing things like nomatter, this is not political at
all.
But whoever does it, whoeverstarts digging through and
trying to make changes, you knowit's going to get worse before
it gets better, because you'regoing to break things or it's
going to be a lot of.
You know the whole fenceanalogy.
You know that Yvonne usessometimes.
You know you don't want to teardown that fence until you know
why.
You know, you know why someoneput it up.
(16:53):
So, yeah, I think there's goingto be some major.
I don't want to useconsequences sounds so terrible
because you, you need to go backand you need to make things
more efficient.
Um, you want to make sureyou're using, you know, those
taxpayer dollars.
Eyvonne (17:08):
Maybe unintended outcomes.
There's going to be all kindsof things that could not have
been foreseen.
It is just when you look at ourgovernment and the geopolitical
climate.
It's so complex that there isno way we could quote Gandalf
(17:29):
here and say even the very wisecannot see all ends Like.
That's how I feel about theplace in which we are right now,
regardless of how you know whoyou voted for or what you
believe is right or wrong.
You know, I think we arestaring down a time of, you know
, significant change in manydifferent facets, and it's going
(17:51):
to be a wild ride.
William (17:54):
Yeah, and even like.
So that's thinking completelydomestic.
Well, foreign and domestic, Iguess.
So, like one thing, I I'm goingto have to try and find the
article and dig it out and I'lldefinitely put it in the notes.
But one of these critical, likerare materials that is used in
like chip manufacturing, amongstother things, is gallium.
I did an episode on it like way, way back before I had the
(18:18):
privilege of you joining the pod.
But the problem is a lot ofthat is brought in from China,
so it's imported from China, andyou can see where the conflict
would be here, because you, youknow, I think the goal is to
have less dependency, you know,on other, you know, especially
(18:40):
China, really.
So, like earlier this year Idon't remember, when a huge
deposit of gallium, like reallyhigh um gallium, over like
thousands of acres I think itwas montana or wyoming, it's one
of the two um was found.
So this stuff is bound to besomewhere.
(19:02):
Like the us probably has a ton,of ton of this stuff buried here
and there, of course, but youdon't know where.
Maybe sometimes you stumbleupon it, but it's a the the
deposit they found was extremelyhigh grade, uh, which is really
awesome.
So that'll.
That'll be good, because theless you know, as we're trying
to ramp up chip manufacturinghere in the us, for you know,
(19:25):
more fabs I think I don't knowhow many more were announced,
but there's more in Arizona nowwe're trying to ramp this up and
get on a, you know, sort of aneven level playing field, which
is thinking about the hardwarerequired to do all the things
(19:47):
these days, especially AI.
It's sourcing that hardware andyou know it is an arms race,
like you you had said earlier.
So pretty cool.
Eyvonne (19:55):
We found some cool stuff in the us less dependency
on foreign nations, so yeah yeah, and as, as we talk about, you
know, there's there's thisnational efficiency conversation
going on.
I think we are going to see,you know, continued impacts in
tech based around thatconversation.
I think the notes that we madein advance really were like tech
(20:17):
is growing up in a lot of ways.
You know, we've had 20 years ofsignificant growth in the tech
industry.
And what were these?
You know, scrappy startups thathad created a very new way of
doing things have now they'vechanged the world, they've
become their own economicpowerhouses and we see across
(20:40):
the board you know Google, meta,amazon, I should say Alphabet
you know making.
You know we started seeingefficiencies in headcount in the
last couple years, but I thinkwhat we're going to see is those
organizations making decisionsthat are more driven by the
(21:00):
bottom line than they have beenup until now.
I think we you know that, andto me this isn't even really a
controversial statement, it'sjust an observable fact that
there's going to be, you know,continued focus on how we
measure outcomes and on profit,and I think we're going to see a
focus not on efficiency, notjust, at least, people talking
(21:25):
about it at the federal level,but also talking about it in our
big tech organizations and Ithink you combine that with AI
and I think in a few years techis going to look radically
different than it has looked forthe last decade.
William (21:40):
Good points, and this is such a compounding thing too,
like something.
We didn't talk about this, butsomething I just thought about
was these big companies almosttravel in packs.
You saw it with thework-from-home stuff, amazon.
They came out and they said,hey, everybody's coming back to
the office.
They went through a mountain of, they took the heavy hits as
(22:05):
far as the initial negativepublicity there being the big
elephant in the room.
But then you saw othercompanies coming out and you
know it's like okay, hey, ifamazon's okay with doing this,
and they're doing it and it'sactually working.
Or maybe it's working, maybeit's not, but you know we're
going to try and do it too.
So then you saw other big techcompanies doing it and then even
(22:27):
feeding down into theenterprise space as well.
Eyvonne (22:31):
You know medium to large enterprise, so it's almost
like a cascading effect, whichis kind of interesting, and I
think like there are going to bedownstream consequences from
those changes and it's going totake a while to figure out how
to reoptimize, right.
I mean, I think we're going tosee, you know, less focus,
(22:54):
frankly, on DEI initiatives.
I think we can see thathappening culturally and we also
just see a focus on revenue andwhat that means and how to be
more efficient.
And I think, as we sort of theslope of the curve for tech
(23:20):
growth changes, we're just goingto see a continued focus on
bottom line, on outcomes and onrevenue, and that's going to
have significant impacts for howthose organizations function.
In my mind, that is the naturaltrajectory of organizations
that do what big tech has done.
(23:42):
They've shown up with new ideas, they've changed the world,
they have recalibrated how theworld operates and now they're
going to focus on operationalefficiency.
That's a pretty normaltrajectory and really shouldn't
be a shock, and there will benew industries that come up
behind it that follow a similartrajectory.
(24:04):
We'll have to see how thatplays out.
William (24:09):
We'll have to see how that plays out.
Yeah, it'll be interesting tosee some of the unintended
outcomes.
If you go fully back in theoffice, you're going to lose
some people, and it might besome good people.
So, just of course, mostcompanies, especially the bigger
ones, know there's a lot offolks that are willing to move,
(24:33):
that want those jobs and theywant to fill those seats.
So maybe the impact won't be asfelt as it would be for smaller
companies.
But even then too, like yourcomment about dei, if you think
about a lot of the DEIinitiatives and just social, I
would say social issues ingeneral are much more important
(24:56):
to people today than they wereMaybe not Maybe importance is
not the right word but more outthere in the public and they
mean more especially and they'remore important to the younger
generation as well.
So it'll be interesting to seeany negative outcomes that hit
(25:18):
businesses because of that, asthey're trying to become more
efficient.
Eyvonne (25:22):
I also think we've seen companies have sort of hit the
boundaries of social engagement.
The boundaries of socialengagement and what I mean is
when you have a very broadconstituency, at some point if
you take a strong social standwith a hotly contested topic,
(25:45):
you are alienating a goodportion of your customer base.
And I think we have started tohit some of those boundaries and
started to see the negativeimpacts of that.
And when those things begin tohappen and actually impact the
functioning of the business,you're going to see
organizations say wait a minutelike, do we really want to take
(26:09):
stands that might be divisive,that impact the bottom line?
I think there was a time whenthat was perceived and probably
even demonstrated to be good forthe bottom line, but I don't I
don't know that that's the caseanymore.
It doesn't mean thatorganizations won't have moral
or ethical stands, that theytake, moral or ethical stands
that they take, but I thinkwe're just going to see
(26:36):
everybody back off on choosingsides as much, because that's
played itself out and is turnedis not maybe the benefit that it
was perceived to be at onepoint in time.
William (26:43):
Yeah, and plus, if you like, say you have two different
scenarios or you have twodifferent options or two
different stances you can take.
You know, stance A and stance b.
Well, if you go and you make apublic endorsement of one or the
other, like you were saying,you could lose 30 of your.
You know it could be an anhyzerwas it anhyzer bush or whatever
um beer I think it wasmanufacturer, yeah yeah, they, I
(27:10):
mean they're they.
They lost a ton of.
Yeah, that was just crazy, uh,seeing that happen.
So that's a very real thing.
And yeah, I mean the businessis there to.
I mean, let's be real, you'reyou.
You're beholden to shareholders, you're beholden to a board, or
you know, if you're a privatecompany, like you have all these
(27:30):
different things, and if you'renot growing or at least
optimizing, or if you're doingthings that are risking, you
know, high risk to the, thebottom line and you, you don't
want to end up on the news andpotentially be in the place of
okay, yeah yeah, this companycould literally die.
(27:51):
We could be facing consequencesof doing this for the next 15
years, if we're even stillaround.
And then okay, what was it forthen?
So yeah, remaining completelyneutral or just not having a
public opinion, yeah, I guess itjust makes sense that that
would be the natural progressionat this point.
Eyvonne (28:11):
Yeah, I think that's that's where it's headed is.
Just, you know, organizationsthemselves not taking as many,
you know, stands on culturalissues.
I think that's where we'reheaded.
William (28:26):
And it would be nice to just not see this stuff in the
news anymore, because you have,you know, company A that's like,
ok, we don't like Trump, or wedon't like Biden, or we don't
like conservatives, or you know,we're against X, y and Z, and
then all these people strike andthen you see about it in the
news for a month and then youknow it happens with another
(28:46):
company, but with the alternateside, and you see the same thing
happen and then both sides justcomplain about each other and
that causes discourse.
Eyvonne (28:53):
So it's yeah, I think we're gonna see less of that
yeah, so what is next?
Oh, you know, we can't getthrough a conversation without
talking about ai william oh you,you brought up AI yeah.
Yeah, make me the bad guy.
No, it's.
(29:14):
You know.
I think you know we continue tosee AI conversations increasing
, but I also think that we'restarting to flesh out more where
it fits, where it makes sense.
We've got, you know, ourfriends like John Capobianco out
there talking about how to doAI and networking and how to
(29:35):
really gain efficiency, and Ithink we're beginning to the
world is beginning to understandit a little bit more.
You know there's still a ton ofhype, but we're going to keep
talking about it.
It's it's it's going to beimportant for a long time to
come.
It's, it's, it's, it's.
It's like saying, maybe, thatthere will be a day when we
(29:56):
don't talk about, like, mobilephones anymore.
And it's like you know, they'renever going away.
They're part of our life nowand we're going to be
continually talking about appsand new use cases and how we're
using that technology to changeour lives in the world.
And AI is going to be the samething Like.
If you're hoping that somedaywe're not going to be talking
about it anymore, I'm afraidyou're going to be sorely
(30:19):
disappointed.
William (30:22):
I totally agree, and with that so I I don't know if
you've heard of this company,google, by any chance.
They have a cloud platform Timeor two.
Yeah, google cloud platform, uh, yeah, so they, they developed
so in.
So one thing I wanted to talkabout anyway.
It was like, okay, if I canlook back through 2024, what was
(30:43):
like something that stood outto me as far as like a gigantic
like that stood out to me as faras like a gigantic like wow,
like tech innovation, um, and sothis sort of fits in neatly
with ai as well.
But hold on, I actually copieda summary so I can just read it.
So google has developed a newquantum chip called willow.
(31:05):
I like that reduces errors asit scales up a major
breakthrough in quantum errorcorrection.
Willow also performed acomputation in under five
minutes that would take asupercomputer 10 septillion
years, demonstrating itspotential for solving complex
(31:26):
problems beyond the reach ofclassical computers.
This achievement marks asignificant step towards
building commercially relevantquantum computers that can
revolutionize fields likemedicine, energy and ai.
And this is just such a coolthing, you know, because if you
think of like we were talkingabout, I think before we hit the
(31:50):
record button, but this quantumcomputing thing is really kind
of a crazy paradigm to wrap yourhead around.
So in the traditional sense,like a classical computer,
everything is ones and zeros.
So things are processed in asense through these things
(32:12):
called logic gates withconditional type things.
So these gates basically takeup to two bits you know one or a
zero is like inputs and go intothe machine and then they
return a single bit as an output.
And when you think of likequantum, there's these things
called qubits, like q, quantumbits, and it's I don't know if
(32:35):
the right word is three, threedimension.
I wouldn't be three-dimensional, it doesn't make any sense.
Um, it's like a uh, like atwo-dimensional vector of of
numbers, like real numbers.
So basically it can hold a lotof information, you know, rather
than just a single bit.
But the problem then becomeserroring out, because if you
(32:58):
think of like an individualqubit, it's almost like you know
.
I heard AI described as like agame of probabilities a while
back.
There's a talk on YouTube AgainI'll try and go back and find
it, but it was a really good one.
But this is almost like adistribution of probability, in
a sense, and averages.
(33:18):
So because you don't have aconcrete value and then you have
, you end up getting into athing of like the observer
effect as well, and this is whyin the article they mention like
the multiverse.
There's a little blurb in there.
Eyvonne (33:36):
Can I read that sentence?
William (33:37):
Please do.
Eyvonne (33:38):
It's talking about the size of the numbers and the
speed of the computation, and itsays this mind boggling number
exceeds known time scales inphysics and vastly exceeds the
age of the universe.
It lends credence to the notionthat quantum computation occurs
in many parallel universes, inline with the idea that we live
(33:58):
in a multiverse, a predictionfirst made by David Deutsch.
So this is next level,otherworldly, literally, kind of
discovery.
William (34:14):
That is a bold, crazy sounding statement.
That is a bold, crazy soundingstatement.
Yes, machine, uh, if you will,um, each of the, the, the qubits
, like the, whatever 64 qubits,they can all kind of like talk
(34:36):
and communicate through thisphenomenon, again known as, like
uh, quantum entanglement.
And that's where you see,honestly, like I was watching
this new uh tv show like thatcame out a few months ago called
dark matter really cool show.
But, yeah, it goes into like,it kind of like takes everything
we're talking about and it'sfrom a book, um, first, and then
(35:00):
the book.
I actually went back and readthe book.
The book's so much better thanthe show.
So, if you're listening, readthe book or listen, listen to
the book, great book.
But it kind of goes in andtakes a lot of these principles
and applies it into a reallycompelling and engaging and
awesome uh narrative um, from a,an author named blake crouch, I
(35:20):
think his name is.
Eyvonne (35:22):
So, yeah, very cool the the other thing to think about,
and and we didn't talk aboutthis in our initial discussion,
william but as you think aboutthe rise of quantum computing,
you've got to think about PQC,which is post-quantum
cryptography, and what quantumcomputing would mean for our
(35:43):
current cryptographic systems.
Cryptographic systems,everything that we have right
now is secured by cryptographythat could potentially be broken
by quantum computing, becauseyou suddenly have I'm going to
say infinitely it's notcompletely accurate, but
(36:05):
seemingly infinitely morecomputing capability, and you
think about everything fromBitcoin to SSL transactions.
What does PQC mean for that andwhat are the long-term impacts
and how do we have to evolve oursecurity measures in light of
(36:26):
this new computing power?
It's really it's a lot to thinkabout and and it's going to be
an interesting evolution towatch over the next few years.
As you know, quantum computingcan begins to become a true
reality at a scale beyond.
You know what Google can provein their lab, right.
William (36:46):
Yeah, and like back to that summary, like it was.
It was like it was like takingsteps towards building
commercially relevant quantumcomputers.
Cause right now it's like I seeall this game breaking research
with quantum computers over thepast five years and it's mostly
like nothing.
Nothing burgers.
(37:07):
There are a few, you know, ibmand so forth that have actually
built these things out.
But when you think of likecommercially relevant, you know
it's like okay, this technologycan be used and has an impact on
the rest of us and that's wherethis is going.
This is just one of those stepsin this big staircase.
Eyvonne (37:25):
But it always starts in a lab somewhere right Big
staircase but it always startsin a lab somewhere right, like
there's always an initial proofpoint and there are theories,
and then there's the practicalyou know, the proof point, the
application in a lab, and thenthere's the commercialization of
that technology, and so we'regoing to see that trajectory
play out in the future and it'sgoing to have significant
(37:48):
impacts on computing at scale.
All of it, you know, whetherwe're talking about our phones
or our communications.
William (38:01):
You think about the reliance that the industry has
on encryption today, about whatthat's meant from a regulatory
standpoint and what happens toall of that when we have quantum
computing power at scale canyou imagine if there was just
some scientific medic or justsome crazy scientific
breakthrough where all thepieces fit together and it was
(38:23):
like, okay, we got to get thisstuff out as quick as possible.
Can you imagine the impacts onso many different things that
would have?
Everything?
Would change everything.
This is mainstream yes, it's.
Eyvonne (38:36):
Uh, I mean, I think we'll get there now, whether it
will be slow and incremental orwhether it will be like a flag
day, kind of a you know crowdstrike moment where we all wake
up and nothing works anymore.
Uh, who knows, right?
Um, and actually that's anotherthing that should have made the
2024 list, but I digress.
William (38:54):
Oh yeah, I just wonder.
You know, honestly, the bigquestion here to really ask
yourself is once this does hitmainstream, will IPv6 be
completely adopted or not?
I think that's the big question.
Ipv6 be completely adopted ornot?
I think that's the big question.
Eyvonne (39:10):
No, it's going to be like the year of the Linux
desktop.
We'll talk about it forever.
William (39:19):
Yeah, good times.
I think we have something else,don't we?
Oh, yeah, a couple more.
Eyvonne (39:27):
A couple more.
We would be remiss if we didn'ttalk about the X data center
and what Musk has done to builda pretty incredible data center
in less than a month that doessome pretty amazing things.
William (39:47):
Yeah, no, you're absolutely right.
It's even got a fierce namelike the Titan or it's going to
bother me now Colossus.
I think that's what it isColossus or Titan one of those
two, the Colossus supercomputer.
Eyvonne (40:07):
that's the word you're looking for.
William (40:09):
Yeah, so it's just um, it's crazy.
So when you think about the,the cost of like one, um h100
and they, they have likehundreds of thousands of these
things and and that's basicallybeing used to train um X AI
model, which they call.
Eyvonne (40:29):
ROC.
William (40:32):
And just the time I mean thinking about.
I mean that's just such a shortwindow of time to actually get
the hardware shipped in.
How do you power all of it,which I think they're still
powering a lot of it throughgenerators that they have
running, that they're shippingin a lot of gasoline for?
Eyvonne (40:51):
Racking and stacking, cabling, the logistical bits of
it, right, I mean, and that'sthat's, frankly, like the easy
part, like the, the architecture, the distributed configuration,
all of that, and it's not andit's not something that's been
done before, so it's not likethey're established
(41:12):
architectural patterns that youcan follow it.
William (41:15):
it really is an incredible technological
achievement yeah, even whenjensen's talking about it, like,
yeah, we us at nvidia, wedidn't think this was possible
and yeah, then someone else justcomes and does it.
You know, and part of thereason they did very
unconventional things, that Inever would have thought that
(41:36):
you would do.
So it's either in Finna band orethernet kind of a thing, rdma,
but like they went ahead androlled, I think, their own
infrastructure, I think they umas they were building it out
like and they're trying to scaleout to all these, these GPUs,
they actually moved an offloadedTCP and into hardware, which is
(41:59):
a very interesting, you know,on the surface, but when you
think about the commoditizationof some of this hardware and
like trying to reach that pointof, okay, we can scale as we
need to, we can just keep adding, adding racks, adding racks,
adding racks and really getting,you know, ramping up, like that
Um, like, if you go through andyou actually read, there was a
(42:22):
presentation from the um, the umhot chips 2024, in which I
think I don't know if it's teslaor whoever um presented at just
as far as like how they viewsome of the things they're
coming out with and I don't knowlike when you I know that the
(42:43):
offloading of tcpip or tcp to Iknow that they did that for
Tesla what they use to trainTesla's um, all these names, not
, it's not claw dojo, they callit, they call it dojo.
So what they do there may bedifferent than how they've
scaled up the um, the datacenter in Tennessee.
(43:04):
I'm actually kind of curiousabout that.
I'll try and find something.
But just these are like verymuch off, very unorthodox
approaches and ideas thatapparently nobody else thought
of, or they would have beendoing it.
Somebody else would havealready done this and nvidia
would have probably said, hey,you should do this, because hey
(43:26):
you what?
We're going to sell more GPUsif you can scale to a higher
number, you know, so it's a veryinteresting space.
Eyvonne (43:33):
The thing that we see with Musk is he has a very
different risk profile than mostorganizations.
He is concerned more aboutaccomplishing the thing than he
is about the profitability ofthat, which, interestingly
enough, has been a prettyprofitable approach for him,
(43:56):
right, apparently yeah.
But he, you know, hey, I want todo this, we're going to do this
.
What's it take to make ithappen?
Screw all the objectiverealities we think we know, like
, throw those out the window.
Let's start at the beginning.
Let's say we're going to dothis, now what do we need to do
(44:17):
to get it done?
And I think that approach hasbeen incredibly effective.
Now there's collateral damagewhen you take that approach that
we've all seen.
That can be true, and at thesame time, this is an incredible
technological feat.
There's no other way to see it.
William (44:38):
Two things can be true at the same time Always, yep.
Eyvonne (44:43):
And in that vein, really the last observation that
we have for 2024 is just thecontinued rise of a new hardware
arms race.
Right, we have seen NVIDIA isstill on top from an earnings
standpoint.
There's a great conversationwith Jensen, the CEO of NVIDIA,
(45:05):
basically saying you know, wehave a situation in which, even
when our competitors make theirproduct free, the cost of
ownership of ours is still lowerbecause it is so efficient and
effective.
And I think you know NVIDIAunlocked a ton of potential and
capability there.
And so we see this new hardwarearms race.
(45:26):
We see NVIDIA doing well, wesee AMD doing well, we see
Broadcom doing well and we seeIntel struggling.
Any comments on the hardwarearms race before I talk a little
bit more about Broadcom,William.
William (45:44):
No, it is an arms race.
It is an arms race, I mean,when you think of the
manufacturing, the, the amountof just land, the amount of
engineering, the amount of, youknow how complicated the supply
chain is.
And I mean even you knowbroadcom.
You mentioned broadcom they'remaking don't they make gpus now.
(46:04):
I think they make gpus now,which I didn't know.
I think I read that.
I'll have to double check, butyeah, it is definitely an arms
race and I think, the ability tohave more efficient, faster,
more computing power.
If you think of AI as like anation-changing technology.
(46:27):
Ai is like a nation-changingtechnology.
Whoever has the most powerfuland innovative infrastructure
that runs it is going toobviously have some sort of
advantage on the global stage.
Eyvonne (46:40):
And that has a lot of impacts.
It makes hardware cool again.
It makes power incrediblyimportant.
It makes power incrediblyimportant.
This wasn't officially on ourlist, but we see the rise of
(47:06):
hyperscalers, investing innuclear and new you know, small
form nuclear and new ways topower, where we were talking
about the effectiveness of GPUsto run AI workloads, and one of
the metrics that we monitor isnot just the utilization of that
GPU but its temperature,because if the temperature is
(47:29):
higher on the chip, we know thatmeans it's working harder and
it's more fully utilized, and sothere's actually things you can
learn about monitoring theperformance of that chip.
It is, how hot is it runningand are we getting the most out
(47:57):
of it that we could.
So it's a whole new world ofunderstanding efficiency, how
best to utilize that hardwareand also how to keep all those
GPUs in sync, because it's notlike packet delivery right,
where you send a bunch of themand they get there when they get
(48:18):
there and then it getsdecomposed on the other side.
That's not how GPU processingworks.
And the other observation I'llmake in about 2024 is with
Broadcom.
In many ways it was the best oftimes and it was the worst of
times the day we're recording.
(48:38):
We saw a huge jump in Broadcomstock price because of their
recent earnings report and, atthe same time, we have customers
reeling from the acquisitionBroadcom has done of VMware and
how they are going to plan forthe future of their VMware
infrastructure.
And I think Broadcom is makingsome efforts to roll back some
(49:01):
of those changes and to be alittle bit more customer
friendly.
I think the thing that we'regoing to find out next year is
whether it's too little, toolate.
Do they do irrecoverable damageto the VMware business?
I think it's different, as weit will be a different
organization than we've everknown it to be.
(49:21):
But what's what's?
What's the future of VMwaregoing to look like coming into
2025?
And where it's going to looklike coming into 2025.
And so Broadcom, for me, is oneof the companies that probably
generates the most cognitivedissonance, because we're seeing
it's just puzzling to see whatthey're doing, both on the
(49:44):
hardware side and on thesoftware side with their
acquisitions and try to puzzleout exactly what the future
looks like with thatorganization yeah, it's wild,
like, yeah, I mean even like, Ithink one I mean vmware was like
channel, channel first, rightum, so just they started they
(50:06):
opened up the wholevirtualization.
you talk about innovation.
Like you know, virtualization,you talk about innovation.
Virtualization really didn'texist at any kind of scale until
VMware, and so, in a lot ofways, everything that we're
doing in our hyperscalersabsolutely is building on their
shoulders.
William (50:24):
Yeah.
Eyvonne (50:25):
Conceptually.
William (50:26):
I remember so I was in tech.
That was actually when I firststarted getting into tech Well,
it was a little after but thewhole physical, the virtual
migration thing was gigantic.
I remember sitting on a CTOtown hall and it was just like
the coolest thing, where theywere like, yeah, we have this
much physical servers today,this time next year, this
(50:50):
percentage of that is going tobe completely virtualized.
So you had this whole numbersgame.
But then you started seeing,like you started seeing it
happening.
That's how I got into vmwareand it was just a huge, a huge
game changer and you think ofhow go ahead?
Eyvonne (51:05):
well, I was going to say it was an incredibly
necessary technology when ithappened in the early 2000s,
because organizations had builtout data centers, they had
expanded to the max capacity oftheir floor space.
Either they couldn't get morepower, which is very similar to
what we're seeing, or they justdidn't have the floor space to
(51:28):
put more gear of what we'reseeing, or they just didn't have
the floor space to put moregear.
And I mean I remember workingfor an organization that was
like we have reduced our entiredata center footprint down to 12
floor tiles right byvirtualization and by, you know,
implementing this consolidation, and the financial benefit for
(51:50):
that was incredible, I think,you know, and that kind of
innovation isn't over.
It's just going to happendifferently, but it was.
I mean, it was huge.
You just can't hardly quantifywhat a big deal that was.
(52:14):
No-transcript Across the wholeworld and I just checked,
broadcom's stock is up 24% todayWow, despite all of the
craziness with VMware.
So you know, regardless of howyou feel about Hawk, his
(52:35):
business from a numbersperspective is through the roof.
William (52:42):
Yeah, it's interesting times interesting times and I
think if they dial back, I meanif they're showing, I mean of
course they're gonna makechanges if it's affecting their
bottom line.
But you know, ripping out thechannel game and then saying,
hey, we're gonna direct sell allthis stuff, definitely, I mean
I, anybody, most people thatwork in this industry could have
said, hey, that's not a goodidea.
But I think they're trying toreverse that now and they're
(53:06):
trying to do some reversals ofsome of the bigger oopsies.
Because, hey, vmware has greatcompetition out there, the
technology is out there.
You've got Proxmox, which is afantastic hypervisor and all
that it takes is a lot of thosealienated partners to wrap up
and scale up their businessmodel to start, and then it
(53:28):
turns into a rising tide, raisesall ships thing because the
partners are winning.
You have some sort of things onthe Proxmox side.
More money is going to go intoProxmox, making it a better
product, licensing and such, andit kind of would almost become
the new VMware in a sense.
Eyvonne (53:48):
So yeah, and I'm biased here as a former VMware person,
but I will still argue that foryou know, enterprise
on-premises virtualizationtechnology, there are
competitors out there, but asfar as a holistic solution I
(54:10):
don't think there's anybody whocan still hold a candle to
VMware today.
But they are gaining and I thinkthat's going to continue to
happen today, but they aregaining and I think that's going
to continue to happen and, withpublic clouds, very competitive
.
That's a different paradigm.
Sure, you can run VMware inpublic cloud.
Lots of folks do it.
(54:32):
I talk a lot about it.
Google has a great solution forit, for example.
But we're also just seeing ahuge paradigm shift from the
center of gravity not beingOS-based virtualization hosts,
it's container-based workloadsor it's other higher order you
(54:53):
know, platform services, and soyou know I think that shift is
as the industry matures, assoftware continues to be
developed.
That's the direction theindustry is moving Full stop.
You know there are still goingto be VM workloads for decades,
just like there are stillmainframes out there doing their
(55:13):
thing.
I mean, I work ininfrastructure and so a ton of
my career is still centered onthose kinds of workloads and
they're incredibly important.
At the same time, you know thetime stops for no man or woman
and things are going to continueto march forward.
William (55:31):
Yeah, I just I hope mainframes don't go away in my
lifetime.
They're so pretty, the zerodown times from IBM.
They're beautiful inside.
It's like looking into OptimusPrime, like the Matrix, like
those things are so cool lookingbut I would never want to run
them anymore.
Eyvonne (55:49):
Okay, all right, you appreciate them for their
aesthetic beauty.
William (55:55):
And they do crunch some major numbers.
They do.
Eyvonne (55:58):
Well, and you know we're back to the what's the
business justification forreplacing this thing, right?
And I know of more than oneorganization in which lots of
people had resume generatingevents because they planned to
(56:19):
replace a mainframe and itdidn't work out.
They spent a ton of money,found out that they couldn't
really do it and then they endedup, you know, staying on one,
and I'm not saying that that'salways the right choice, because
it's not.
Sometimes you absolutely haveto modernize, but I think it's
smart to say, okay, what's thisreally going to cost, what's
(56:40):
it's really going to take, anddoes it really make sense for us
to put all of our efforts here,or should we put that
innovation effort somewhere else?
And it's a smart leader,technical leader, who does that
for their business and their usecase, because there's no one
right answer.
I know organizations that arerebuilding everything to replace
(57:02):
mainframes that are beingwildly successful both
technically and in themarketplace, and I know
organizations that are keepingtheir mainframes that are being
wildly successful and innovative, both technically and in the
marketplace, right?
So there's no one answer forthat question.
And I think we're going to seethe same thing play out for VM
workloads.
We're going to see likeeventually there's probably
(57:25):
going to be a delivery mechanismother than a container.
I know I'm insulted.
William (57:35):
No, just kidding.
No, you're right, though youknow it's just going to keep on.
I mean it's funny becausecontainers are kind of like re.
You know they're modernizedjails of old.
You know they're kind of like akind of the same technology.
You know it's the whole.
You know rule 11 thingrepurposed, reoptimized for a
(57:56):
new world and new workloads andnew application designs and
things.
Eyvonne (58:01):
It's a different, more contained abstraction that can
be more managed and deployed incode, and it provides more
consistency and less dependenceon the underlying infrastructure
, including the operating system.
Right, um, and and that's agreat thing like we love
(58:24):
abstractions right, we just wantto be sure we're using the
right ones we.
William (58:29):
You know what we ought to do.
I know we probably need to tieup, but where we need to get a
cobalt developer.
We need to get a cobaltdeveloper on here to kind of
discuss, or somebody that's veryin tune with mainframe
architecture and the tie-ins toinfrastructure and all the like.
What, what are the?
What would it actually requireto get rid of mainframes?
(58:51):
Because they're still in thedna of a lot of our core
infrastructure and cobaltdevelopers are aging out?
Eyvonne (59:00):
I would also love to see somebody do a.
A.
There's a word that I'mthinking, that I'm that I want
to use, that I can't pullmagical musical chairs?
Yeah, no no, you're not helpingshoe manufacturer.
No, you're not helping.
Shoe manufacturer, I would loveto do a discovery of all of the
(59:25):
abstractions throughout thehistory of computing, like on a
timeline, right, you've gotbinary, you've got assembly, and
then on top of assembly you'vegot your programming languages,
and then on top of yourprogramming languages, you've
got your compilers andinterpreters, and then there's a
code level abstraction, andthen you've got like the
(59:47):
hardware abstractions, right.
I think it would be fascinating.
William (59:52):
Because then when my stuff moves to software, they,
you know, offload certain piecesof it to hardware again and
then like it's yeah, stuff, thatwould be a great conversation.
I'm down, yeah, if, uh, theworld is listening, if you know
anybody that wants to have thatconversation and is an expert it
takes about like five peopleand then you can talk about
networking abstractions.
Eyvonne (01:00:12):
Right, a whole nother kind of abstraction.
You start with the bits on thewire and then you talk about
protocols and encapsulation andthen encryption and then, yeah,
the osi model like it's.
There's so much fun stuff youcan talk about yeah, that sounds
like a nice round table that'sa project for 2025, william
(01:00:32):
sounds good.
William (01:00:33):
Well, we better wrap up .
We went over the normal time,but this was a lot of fun, a lot
of unplanned, just off the cuffconversation, and I just want
to extend, you know, happyholidays to all of you.
You know, no matter what you'redoing, try to find time to
relax.
Um, if you're in cold climates,find a good warm tea or hot
(01:00:54):
chocolate, uh, whatever, relax,read a book, you know.
Get off the screens for alittle while, maybe do something
, play some board games.
Eyvonne (01:01:04):
Watch a Christmas movie .
William (01:01:06):
There you go.
A Christmas Story is myall-time favorite.
I love that movie.
Eyvonne (01:01:11):
It's a Wonderful Life is one of those nostalgic
favorites, and I might be asucker for a Hallmark Christmas
movie or two.
I can't watch too many of them,but every now and then there's
something comforting about thatand you can cringe all you want.
It's fine, go watch people blowstuff up.
William (01:01:32):
Yeah, I mean those Hallmark movies are great,
though, like if I'm reallyhaving a hard time sleeping I
can just crank one on.
Eyvonne (01:01:40):
I have a few of those too.
Yeah, hey, there's nothingwrong with that nice, nothing
wrong with a predictable storywhen you know what it is and you
need just to feel good for aminute.
William (01:01:51):
Yeah, yeah, all right whatever you got to do to
naturally generate that dopamineright there you go, exactly, go
, exactly, all right.
So, yeah, thanks for the time,merry Christmas.
Eyvonne (01:02:04):
Happy Hanukkah.
William (01:02:05):
Yeah, enjoy, enjoy, enjoy.
Eyvonne (01:02:06):
Enjoy yeah.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com