July 16, 2025 • 35 mins
The tech industry's shifting strategies and priorities take center stage in this episode as we examine several major developments reshaping the landscape. We dive into Ingram Micro's recent ransomware attack by the SafePay group, analyzing both the immediate impacts and potential long-term consequences of the four-day operational disruption. The incident raises important questions about security protocols, attack vectors, and the possibility of data exfiltration that could have lingering effects.
Next, we explore CoreWeave's massive $9 billion acquisition of CoreScientific, representing a fascinating pivot from cryptocurrency mining infrastructure to AI computing resources. This transaction epitomizes the tech industry's current trajectory – moving away from speculative cryptocurrency ventures toward equally speculative but potentially more productive AI applications. We discuss the diminishing returns of crypto mining versus the still-unprofitable but product-focused world of AI, questioning where this astronomical investment might lead.
The networking world sees significant realignment as Arista acquires VeloCloud from Broadcom, bolstering their WAN portfolio after VMware's SD-WAN solution changed hands multiple times in recent years. We compare potential integration approaches with Cisco's historical challenges integrating Viptela, speculating on how Arista might handle this transition while maintaining their commitment to a unified operating system.
Perhaps most concerning is Microsoft's continued workforce reduction – cutting 4% of employees while pledging billions toward AI development. This "robbing Peter to pay Paul" approach appears throughout the tech industry as companies bet heavily on AI's future promise while sacrificing present human capital. We contrast these widespread layoffs with contradictory reports claiming robust tech hiring, questioning the true state of industry employment.
Join us biweekly for future episodes as we transition to a fortnightly schedule focusing on quality over quantity. Share your thoughts about our schedule change or any topics discussed – we're especially interested in hearing from VeloCloud customers who've experienced the product's journey through multiple corporate owners.
Purchase Chris and Tim's new book on AWS Cloud Networking: https://www.amazon.com/Certified-Advanced-Networking-Certification-certification/dp/1835080839/
Check out the Fortnightly Cloud Networking News
https://docs.google.com/document/d/1fkBWCGwXDUX9OfZ9_MvSVup8tJJzJeqrauaE6VPT2b0/
Visit our website and subscribe: https://www.cables2clouds.com/
Follow us on BlueSky: https://bsky.app/profile/cables2clouds.com
Follow us on YouTube: https://www.youtube.com/@cables2clouds/
Follow us on TikTok: https://www.tiktok.com/@cables2clouds
Merch Store: https://store.cables2clouds.com/
Join the Discord Study group: https://artofneteng.com/iaatj
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Chris Miles (00:14):
Hello and welcome back to another episode of the
Cables to Clouds News Update,where Tim and I are going to go
through some of the networkingand cloud-adjacent tech news
articles that have come out overthe last couple of weeks and
give our opinions and commentsalike.
It is a very dark and earlyTuesday morning here in Sydney,
(00:35):
Australia, and, I'm assumingtemperament and beautiful
afternoon Monday afternoon inRaleigh, North Carolina.
But before we get into the news, I do want to take care of a
couple of housekeeping items.
Um, first of all, Tim, how areyou doing?
Tim McConnaughy (00:52):
What are we doing this now?
Chris Miles (00:54):
Well, I just realized now that we don't work
together anymore.
I probably check in with you alittle less than I, than I
normally would as a as acoworker.
So, yeah, what's new man?
You man, you just went to japanyeah, yeah, I just got back.
Tim McConnaughy (01:06):
Actually, I got a ton of uh pictures and stuff.
Uh, I'm putting together somekind of post because everybody's
asking me like, hey, how was it?
And it's like I have so manypictures, I have so many stories
, like I'm just not gonna typeor say that a hundred times.
So, yeah, I'll put somethingtogether and I'll post it.
Uh, it was good though.
I mean, I took the test, uh,took the jlpt n3.
I was amazed by how many peoplewere there, like that it was a
(01:29):
convention hall filled withdesks.
I could not believe how manypeople were there from different
countries, um, so, likejapanese, uh, nationals, uh
natives rather, cannot take thejlpt like the provisions test
for language.
Uh, only foreigners basically.
So it was all foreigners.
Chris Miles (01:45):
So that's even crazier, that it was that well
attended.
Yeah, non-nationals yeah, yeah.
Tim McConnaughy (01:50):
So I mean, if I did fail, it's probably because
the listening section, becausethe listening section had a
decent amount of echo because ofthe size of the of the room and
everything I was.
I I knew it was gonna be aproblem.
I just well, it is what it is,we'll see how it went.
Um, I'm hopeful, but yeah, I, Iwill know if I pass.
I probably just barely barelypassed, which a pass is a pass.
(02:11):
But uh, if I want to move on tothe next level, then it's not
gonna work anymore.
They definitely make it a lottighter per level.
So we'll see, man, we'll see.
But yeah, it's, it was good, itwas nice being in the same time
zone for a little bit of time,about 10 days or eight days or
so.
Chris Miles (02:26):
Yeah, yeah, that's true.
Very rare occasion that we'reon the same sleep schedule.
So that was nice.
Yeah, well, that's good.
Good luck, I'm sure you'll passwith fine colors.
We'll find out in a couple ofmonths, I guess.
But yeah, so before we get intoit, another update from us here
(02:46):
on the podcast.
So Tim and I, just with ourkind of ever busy schedules you
know me starting a new job andTim obviously has a lot of irons
in the fire as well We'vedecided to push the show back to
a fortnightly release for theoverall show.
So this includes our regularepisodes as well as our news
(03:07):
episodes.
So you'll be getting a Cablesto Clouds episode every two
weeks versus every week thatyou're seeing now.
So that will.
I think we want to take a shiftmore towards focusing on
quality of episodes rather thanquantity of episodes, and with a
weekly cadence it was startingto feel like we were rushing to
(03:27):
make sure we were getting peoplein to record episodes and
things like that, and I feellike the content was potentially
suffering a little bit.
So we want to shift back andmake sure we can kind of have a
little bit more thoughtfulconversations and things like
that.
So we're going to be releasingon a bi-weekly or fortnightly
cadence starting from this week.
So 16th of July will be thestart of that.
(03:50):
So you'll get another episodetwo weeks after that.
So just a little heads up andwe'll let everybody know on
social media as well.
And if you have any comments orfeedback on that, please reach
out.
I'm happy to hear from you guysas much as possible.
So look forward to that.
And with that out of the way,let's hop into the news.
Let's see All right.
(04:13):
So first up, we have an articlehere that was first brought to
us by Bleeping Computer.
So Ingram Micro, who is one ofthe world's largest IT
distributors we actually doquite a bit of work here in
Australia with them as wellsuffered a ransomware attack by
a ransomware group calledSafePay.
(04:34):
So basically, back on the 5thof July they've kind of reported
that there was a cybersecurityincident going on and that some
of their internal systems hadbeen compromised, and that's
some of their internal systemshad been compromised.
There was the attack lastedover it sounds like about four,
(05:03):
four days or so before theyactually were able to resume
their businesses or thebusinessgram systems.
It's kind of confusing thetimeline here because they came
out and said initially that theythought someone had gotten in
through compromised credentialsover their global to protect VPN
platform, which is kind of aVPN onboarding platform provided
(05:26):
by Palo Alto Networks.
And then Palo Alto Networkskind of commented that they were
looking at the incident withIngram as well, and then later
there was a comment saying thatthe VPN gateways were not
compromised using these passwordspray attacks that were called
out in the attack as well.
So I don't know exactly if thatmeans that's not how the
(05:47):
intruder got in or if that'sjust it wasn't a password spray
attack and maybe it was justsomething that was a little bit
more contained than that, butnonetheless it seems like this
was a pretty serious attack.
They were actually able to, orforced to, kind of shut down a
(06:08):
certain sector of theiroperations.
They were telling people to gohome and work from home,
although if the remote VPN wasthe platform that was
compromised, I'm sure they shutthat down too.
So if you were potentially anemployee of Ingram Micro, you
might have been sitting at hometwiddling your thumbs for a
little while.
So this seems like it's prettyserious.
(06:28):
Wish the best to all involved.
It doesn't seem like there'sbeen any money that's exchanged
hands here with regard to theransomware attacks.
So hopefully this is the lastof it.
But you know, sometimes whenyou hear that you know there
hasn't been any exchange offunds, sometimes there's, you
know, exploitation that comesout later due to data that was
(06:51):
was compromised during the event.
So yeah, so pretty rough onehere.
I'm glad to see them back innormal operations, though.
Any comments from you, tim?
Tim McConnaughy (07:01):
No, I agree.
Any comments from you, tim?
No, I agree.
The timeline on the articlesthat we were reading as kind of
published by Ingram and thencollected in these articles was
a little strange and it wasn'tclear.
I still, even now, they haven'tbeen clear on exactly how the
attack was successful.
They did initially believe itwas compromised credentials or
password spray.
(07:21):
Now Paolo specifically came outand said our global protect VPN
was not compromised or VPNgateways were not compromised.
But as a vendor, saying thatyour gateways were not
compromised, that doesn't saythat just means like the road
was not compromised, but likethe car could have been, you
know, yeah for sure it couldstill have been and probably was
(07:44):
.
Ultimately, you know, leaked orphished credentials, potentially
that that gave them entranceinto it, although you kind of
(08:06):
wonder, you know, from a zerotrust perspective, how leaked
credentials could lead to adecent amount of especially
apparently important enoughcomputers to end up being, you
know, ransomed.
So maybe there was like a lackof zero trust in the and this
would have been, I assume, on onIngram side cause.
Global protect VPN really justprovides the access, you know,
and then and then the policybased on how the company would
have configured it Right.
So yeah, it's.
It's very unclear from thearticle, but I would assume
(08:26):
Occam's Razor would suggest that, yeah, compromise, spearfish,
potentially credentials, andthen maybe a lack of true zero
trust enforcement that wouldallow lateral movement or
something like that that wouldlet an attacker compromise and
ransom multiple systems.
I do agree with you thatnowhere in here does it say
(08:47):
anything about them paying theransom.
In fact, because they were downfor something more like four
days or something, it's far morelikely that what they did was
contain and then replace thecompromised devices.
But yeah, good question Wasthere exfiltration that happened
during that time that's goingto have data sold on the dark
(09:09):
web for recouping some of thatmoney for the attack?
Or are there other backdoors orsomething potentially that have
been put in Because that couldhave happened?
Maybe they picked some devicesthat they didn't ransom.
Maybe they've kept them asbackup root kits or something
like that where they could comein back later.
(09:29):
So I think time will tell.
And this is the problemultimately, whenever you're
attacked by, whenever you suffera cyber attack, is that it is
truly difficult to really knowhow deep it goes.
Right, like you know someonewho has access to your systems.
You know, the ransomware partmight just be one method of
(09:49):
trying to extract value from theattack, right?
So yeah, I guess we'll see.
Yeah, best of luck to Ingram,and hopefully they actually
rooted it out and won't sufferany problems in the future.
Chris Miles (10:03):
Yeah, for sure.
I mean it seems like they did.
I mean we've talked about cyberattacks on this show quite a
bit and the thing is like thesethings are going to happen, you
know, from time to time, and Imean it looks like they had a
relatively good approach for,you know, instituting controls
to stop some of that.
(10:23):
Um, you know, instituting umcontrols to to to stop some of
that.
I think they did like a globalreset of every employee's MFA
across the entire company, whichis, you know, probably not fun
for the employees, but but uh,but it's good that they were
able to do that Right, um, andthat's uh, that's a way to at
least kind of um stop thebleeding for a bit.
So, um you know yeah, all thebest.
Tim McConnaughy (10:44):
Hopefully it works out okay.
Okay, so the next story comesto us from Network World.
Coreweave is acquiring CoreScientific.
No relation I know that theyboth I guess they're now related
, but previously no relation for$9 billion with a B, to quote
unquote power their AIinfrastructure push.
Billion with a B dollars, toquote unquote power their AI
infrastructure push.
This is interesting.
(11:05):
So CoreWeave is actually apublicly traded company that is
an AI cloud, an AI backed cloud.
Nvidia backs CoreWeave, likethey're invested or whatnot in
CoreWeave and they provide AIservices, right, just as a
on-demand AI services.
So that's CoreWeave.
Now CoreWeave is buying acompany called CoreScience
(11:27):
CoreScientific, rather.
Corescientific is actually acryptocurrency mining operation.
So you can already, you know,from a technology perspective,
you can obviously see the valueof an AI company acquiring a
farm full of GPUs.
So I think we talked about thisa while ago, like how
(11:52):
cryptocurrency farming and AItechnology, if you will, the
underpinning technology poweringAI with GPUs are.
First of all, they're bothextremely power hungry.
They're just like sucking theplanet dry of resources but also
how complimentary, you know, wewere kind of talking about like
who's going to win the gpu wars?
Right?
Like are we going to end upwith with crypto miners as the
(12:15):
winners, or is ai going to bethe winners?
I think we're seeing adefinitive move towards ai.
Um, partially because, first ofall, crypto has just been in the
game a lot longer, and thething with crypto is, the longer
a coin exists, the harder it isto continue to mine that coin.
So it's a what do they call it?
A diminishing return.
(12:35):
The whole thing is adiminishing return, so over time
it has to get harder and theROI, the return on investment
there is worse.
So the Network World articlepoints out the like OK, so
Quorum Scientific might havebeen struggling.
And then this is a good exitstrategy for them as a company
to just like cash out,essentially just get out of the
(12:56):
game and I couldn't agree more.
Honestly, if somebody's got $9billion to throw at it, I
couldn't imagine a better exitfor that company.
And then to just sell all yourshit to ai and the overpowered,
over, overblown, overpriced, uhtime that you can sell it right,
like well, it's as expensive asprobably ever you know ever
gonna be, and then and then exit.
(13:17):
So yeah, we'll, we'll see.
I mean, there's not much to seehere.
They sold it, so I guess we'llsee um if these guys get their
return on investment.
That's the real question.
Chris Miles (13:27):
Yeah, I mean, I guess, if you have somewhat of a
cynic point of view, it's funnyto see one technology that was
new on the scene and promised tomake people a lot of money now
is getting bought by another newtechnology that promises to
make a lot of people, a lot ofmoney that still is not
profitable, um, in roughly anycapacity.
So, um, some people are probablysaying, oh well, what's the
(13:50):
next thing?
What's going to buy core weaveand you know what.
It's five, six years orsomething like that.
Um, I mean not to be mean, butlike it's I'm, I'm kind of glad
to see some of the crypto miningand crypto adjacent stuff
starting to be on the way out.
Like it's I'm, I'm kind of gladto see some of the crypto
mining and crypto adjacent stuffstarting to be on the way out.
Like it sounds like even, um,some of the blockchain
technology stuff is is getting,yeah, um, overpowered by the um,
(14:14):
by the quantum computingconversations nowadays, right,
so it sounds like all of that iskind of being swept back under
the rug.
I don't know if it's going todie a fiery death or just slowly
wither away, but yeah, it'sjust funny to see this happening
, but yeah.
Tim McConnaughy (14:32):
Yeah, it's like pure speculation buying pure
speculation Like there'sactually, at least there's an
actual demonstrable product onthe AI side of it Right, at
least there's an actualdemonstrable product on the AI
side of it.
You know, like versus Bitcoinor or cryptocurrency, which is
truly like digging a hole inyour backyard and just picking
up rocks and telling everybodythey're worth a lot of money.
(14:55):
You know, like I don't, I don'tknow how else to you know,
there truly is no actual productassociated with blockchain
besides the blockchain itself,right, which is, as we've seen,
there's a lot of challenges,like proof of work and all this
other stuff that makes actualconsumption contracts and all of
(15:16):
that blockchain technologyactually not make a lot of sense
for the most part, right.
So, again, it's a product, it'spure speculation and at least
AI has, even if it's notprofitable, at least there's a
demonstrable value propositionof product to be produced.
But, like you pointed out, likethere's, it's still so far from
(15:37):
profitability.
I think they're still castingwide nets to try to figure out.
Okay, well, how do we, asuseful as all this is, how do we
make it profitable versus thecost of production of bringing
it to the market?
So I think we're going to stillbe searching for that golden
ticket for quite a while.
So, yeah, maybe something willcome, Maybe the next thing will
(15:57):
come along and buy that.
So I mean, there's a planetleft.
It's interesting.
Chris Miles (16:05):
Yeah, we'll see if this kind of stuff and the
stargate project and all thatkind of there's just so much
money being pumped into thisshit, man, it's like, yeah, it's
like amounts of money thatwe've never even had to like
fathom in any capacity and andwhere's it coming from?
Tim McConnaughy (16:19):
like truly, where is it?
Chris Miles (16:20):
all coming from?
Yeah, because the ai is notmaking any money yet.
So like what?
Yeah?
Yeah, it's all speculation.
So maybe we're due for anotherdepression or something We'll
see.
Tim McConnaughy (16:30):
It is the 20s.
Chris Miles (16:32):
Yeah, I know, right , it could happen, all right,
moving on, so we've alreadytalked about this a little bit
on one of our past shows, but itwas kind of just a rumored
thing and now it seems it's uh,it's actually come to uh, come
to fruition.
But arista um has basicallyannounced that they have um
(16:53):
acquired, or starting to acquireuh velo cloud from broadcom um,
which was uh previously, priorto being uh vmware being bought,
um was a v, a VMware product aswell, right?
So VelaCloud is a Vela orVMware.
It was VMware's SD-WAN offeringthat they had on the market was
, you know, relatively wideadopted for a while, which is, I
(17:16):
think, a good thing for Arista.
I don't think they had a strongSD-WAN portfolio prior to this,
so this will be something tokind of help round out their
kind of WAN.
I don't know if people wouldlike people always have their
unique names internally for whattheir products thing.
So this is going to fit intothat kind of WAN category and I
(17:36):
think Arista has been very knownfor being prominent in the data
center and maybe here and therefrom, you know, private clouds
or even public clouds, et cetera.
So I think this will be good tokind of expand their portfolio
with the the way in as well,which is good.
I got to say, man, it's like ifyou were, if you were, an OG
(17:57):
VeloCloud customer, you.
I'm curious to hear opinionsfrom them on what this, what
this ride has been like, like asif you came on being a VMware
customer.
I mean, let's be honest, likepeople that like VMware, that
doesn't necessarily mean they'regoing to like anything offered
by Broadcom or even Arista.
Those are kind of the companiesthat have their very you know
(18:20):
opinionated fans in certaincapacities and ones that don't.
So I'm sure they've been on awild ride, uh, over the over the
course of the past couple ofyears.
Um, but yeah, so I think thisis good.
Um, I think it does leave alittle bit of uncertainty on the
table for, um, things like,apparently, broadcom and
(18:41):
VeloCloud brought a SASEsolution to the market as well,
combining the SD-WAN and theSymantec offering from Broadcom,
which was also under theirumbrella.
So I don't know if that's justgoing to kind of get the axe.
I don't think it had an amazingamount of the market cap in the
(19:01):
SASE world, but you know, Idon't know what the future of
that is going to look like.
Maybe we'll have to look intothat.
And kind of on the back of this,they've also Arista have
announced that they've announcedor sorry, they've announced
that they've hired a new COO inTodd Nightingale, who's coming
over from Fastly.
(19:21):
It's where he's been the CEOfor the past three years.
So I'll say he kind of comesacross and say he's a networking
person, he's excited to getback into kind of the networking
world and things like that,which is good.
I think he kind of worked hisway up through the ranks at
Cisco for many years, so he'skind of coming back into that
(19:44):
camp, although probably workingfor someone that Cisco has not
been too fond of for quite awhile so that's kind of funny.
But you know, here we are.
I mean, I will say, with thisVeloCloud acquisition, things
like that, I think he's got hiswork cut out for him.
He's got a lot of, um, you know, strategic execution that's
(20:05):
going to need to, um, uh, kindof come to fruition and play out
for arista in this part, but,um, yeah, overall I think it's a
relatively good move for themand, uh, we'll have to see where
it goes.
But if you are, if you are anog velo cloud customer, when
they were arista and now you've,you've, you've come in, uh,
vmware when they were Arista andnow you've come you mean VMware
(20:26):
, sorry, vmware, yeah, sorry.
VMware VeloCloud customer, andnow you've made it through the
ranks over to Arista.
I'd be curious to hear opinionson what that transition has
been like, so reach out to us.
Tim McConnaughy (20:39):
Well, and more importantly, importantly, are
you still like, because when vm,when broadcom bought, you know,
vmware, um did they come outwith a hey, your velo cloud is
now 10 billion dollars.
You know, like, did that happen?
Chris Miles (20:52):
or yeah, maybe the maybe the bills will go.
Maybe they went up 10x andmaybe they'll come back down to,
like you know, 2x or somethingit's so interesting because I
think everybody I mean, how manytimes have we talked about this
?
Tim McConnaughy (21:06):
broadcom runs like a, like a uh, what's the
word?
A private equity firm?
Right, yeah, broadcom workslike a private equity firm.
That's how they make theirmoney.
They act like a private equityfirm.
So them spinning off bellowcloud is not surprising in the
least.
It probably was not a hugemoney maker for um, it was
probably one of those thingswhere VMware had it, because it
(21:26):
made loyal customers, kept intheir ecosystem and provided a
solution for them.
But if you come in and you'relike Broadcom, the PE company,
you're looking at it and you'resaying, well, this is a business
unit we're paying for.
That isn't making.
You know, we can't charge abillion, million billion dollars
for it, so let's just sell thebusiness unit and focus on the
parts that we can overchargecustomers for.
(21:48):
So that's not a surprise?
Actually I would not.
I guess it makes sense.
It's one of those things thatafter the fact you're like, yeah
, that makes sense.
But I wouldn't have thought atthe time about Arista buying
Velo, Because doesn't Aristahave some weird like off the
shelf, like something they builtthemselves for, like a STYN?
Chris Miles (22:06):
Yeah, I think that's.
The thing is Arista, like Isaid, being such a prominent
vendor in the data center world,it's like I've never really
heard too much about them havinga.
They have something.
I know they do.
They definitely did havesomething.
I remember when this wasrumored, I started looking up
and they had something.
But the thing is the way it waslisted it was like was vague
enough to where I just wouldn'tunderstand if it was an SD-WAN
(22:27):
product or not.
Tim McConnaughy (22:28):
Yeah, you know what I mean Pathfinder,
pathfinder, it was calledPathfinder.
Yeah, it was part of the CloudVision thing.
We did look this up when wefirst saw the rumors.
I remember now that was a whileback now, but yeah, pathfinder,
they called it.
Chris Miles (22:42):
I mean now they obviously have a very strong
portfolio in the SD-WAN campversus something that we weren't
even sure was SD-WAN right.
I mean SD-WAN has gotten to thepoint where it is table stakes.
Tim McConnaughy (22:53):
It really is Right.
Chris Miles (23:17):
So it's like you know, they talked about it.
It really is like India, so tosay a ton of MPLS still, but
yeah, so I don't think itnecessarily killed that, but
it's definitely like you got tohave your WAN has to have some
sort of intelligence built intoit now.
So we're definitely at thatpoint.
So I think they, rather thanyou know, kind of being behind
the eight ball on that.
I think it's probably good toget something with as much
(23:38):
recognition as VeloCloud.
Tim McConnaughy (23:40):
Yeah, I wonder if they got it cheap as well.
I mean relatively cheap, right?
Relatively cheap, because I'mpretty sure Broadcom was trying
to get rid of it.
Like I said, they'll probablyspin off other business units
that aren't profit.
They can't charge insane profitmargins on anyway, because
that's just their thing.
So, yeah, maybe, maybe aristawas like just opportunistic
(24:03):
about it, like, hey, ourpathfinder is okay, but we'd
much rather have a I.
So now it comes down tointegration.
Let's see, arista, whatarista's integration strategy is
and how they fix it and put itinto their cloud vision or their
, you know, their, theirplatform that they're using for
all the other, uh, smart stuffthat they're doing.
Chris Miles (24:22):
I mean sadly.
I wonder if this goes the exactsame route or different route
than Cisco, cisco, like, like.
Arista is very yeah, causeArista is very big on the
unified operating system, right.
Tim McConnaughy (24:35):
Like EOS.
Eos is their is like their oneis their one thing that runs
pretty much most of theirplatform.
Chris Miles (24:42):
Veloclide's obviously not going to run that,
so it's like do you merge thetwo.
Do you cause?
That headache for people, orcan you do it in a much better
fashion than Cisco did?
Are they going to learn thelesson?
Yeah, I don't know.
Tim McConnaughy (24:54):
Will they learn the lesson, because that really
hurt that.
That really hurt sdn adoptionuh, waiting, waiting to backport
it to uh ios xc.
Now, ultimately, I think it didactually do well for the
platform, like, I think, sdn,cisco, sdn is doing better, but
at the time, with vitela beingsuch a market leader, uh, the
stakes are different.
(25:15):
Now.
Vitela is not a like velo cloudwhatever, like sd-WAN doesn't
have this race that it hadeither.
Yeah, true, so like now, theycould probably take their time.
You know, they could justsupport existing customers and,
in the background, with nopressure whatsoever, work on
that integration.
I think that's probably whatwe'll see.
Chris Miles (25:35):
Yeah, I mean it feels like we're kind of
poo-pooing Cisco.
I think it was needed.
They needed to do it.
I think there was just some.
I mean hindsight's always 20-20, I guess.
Tim McConnaughy (25:44):
Oh, of course.
But, and knowing how thebusiness units at Cisco are
generally operated, I was notsurprised.
You know, I was at Cisco at thetime.
I was one of the you know, oneof the people you know we had a
that was working on it when itfirst launched, you know, within
(26:06):
like a few months of theacquisition and, yeah, I mean it
needed to happen.
I think a big complaint ofcustomers, and even internally
at the time, was that and thiswas this is all water under the
bridge, right?
Was that at the time, Vivitellahad a lot of things planned
that were new and exciting, andremember this is what six years
ago now or something, almostfive or six years now that it's
(26:26):
been and a lot of new featuresat the time that were brand new
features that weremarket-leading, going to be huge
features that would reallycement Vivitella as the leader
at that time.
Um, that really got put on theshelf, you know, and you know so
that this could all be movedover to ios xc and so you had
like a whole year where all thecompetition then was able to
(26:47):
catch up and and and, like velocloud being one of them,
actually a great example of oneof them I think that was when
the acquisition happened.
Chris Miles (26:55):
Right yeah, that's what.
Tim McConnaughy (26:56):
I'm saying it's not the same world.
Now my guess is that Aristawill probably take their time
and just support the existingcustomer base while they figure
out how to do.
But I do agree with you on theEOS thing.
That is the kind of their Oneof their big, huge things is
like hey, we've got ES one one,one operating system everywhere.
(27:17):
So yeah, man, I think we willsee it, but I don't.
I don't think they have thepressure to deliver as quickly.
Chris Miles (27:23):
I agree.
Yeah Well, we'll have to seewhat they do.
I think, I think this is.
I mean, this is just the painsof an acquisition, right, this
is what you have to go through,yeah.
Tim McConnaughy (27:38):
So we'll, we'll keep our ear to the ground and
we'll see what they end up doing.
Speaking of the pains oftechnology, so our next article
comes from Reuters and this isback in July 2nd.
So this actually, I believe,has already happened.
I've seen it on LinkedIn, a fewpeople already saying like hey,
I've been let go from Microsoft.
So Microsoft was announcingthey were going to lay off 4% of
the workforce.
Another cut Like Microsoft isjust like chopping thousands and
(28:01):
thousands of employees.
Chris Miles (28:06):
Now they had said they were going to put like
$9,000 per year to pay for theAI.
Tim McConnaughy (28:09):
I think this did come out to be about $9,000.
Yeah, yeah, and I think that'sall, yeah, something like that.
So, yeah, so they had pledged$80 billion in capital spending
and then part of that was to payfor that was going to be
cutting workforce.
So this was and we reported onthis when they announced, when
they first announced it monthsago and this is just like
fruition.
This is the next round,essentially, of something that
(28:32):
we're already working on.
So this is so interesting.
A lot of tech companies aredoing this right, robbing Peter
to pay Paul.
Basically, they're cuttingtheir headcount to pay for their
expansion into AI.
From the AI is going to be morethan what they're going to lose
in productivity and, and youknow, whatever outcomes that
(29:01):
they would from having humancapital.
Um, I, it's.
I wish I knew or couldunderstand what their thinking
was here, because it'scompletely speculative, right,
like there's been no actualthere.
There's been, there's still, tothis day, still now, there has
been no indicator that ai isgoing to be profitable to the to
(29:24):
the point of, like, what you'repaying at laying out for it now
, will it eventually becomeprofitable?
Maybe there has to be somemicronization of technology.
That's going to have to happen.
You know, whatever, whatever,but it's still very much in the
we've got a pitchfork and we'rethrowing money on the fire phase
.
Except in this case, it's notmoney we're throwing on the fire
, it's essentially people,people's livelihoods anyway.
(29:46):
But yeah, they're betting onthis, and so is Meta and Google
and Amazon.
All the big tech giants aregoing through this motion to pay
to basically keep up with theJoneses.
So it's not sustainable.
But yeah, but before we getinto that just I wanted to
there's one more article that Ithink is hilarious and we'll
(30:07):
cover, you know, we'll talkabout them together, and it's
one that, according to NetworkWorld and I forget which.
Let me just look real quick theI forget what it was.
Chris Miles (30:16):
It was CompTIA.
Tim McConnaughy (30:16):
CompTIA, that, and I forget which.
Let me just look real quick theI forget what it was.
It was CompTIA.
Comptia, that's right.
It was a CompTIA study sayingthat tech hiring exceeds
expectations in June.
So, like, literally, you havethese two articles right next to
each other saying that the techunemployment rate fell slightly
in June as companies listed455,000 job listings for tech
positions.
(30:37):
455,000 job listings for techpositions according to this tech
jobs report.
It really buries the lead onthis, saying that, like, tech
unemployment is low and it'slower than expected.
And then it goes on to pointout that, like, almost all of
the new job postings are aroundAI, like people that know AI,
people that are involved with AI, data science, et cetera, et
cetera.
So, and it also points out thatthose 455,000 listings are just
(31:00):
listings.
It's not really saying that,like, more people are being
hired than fired or whatnot.
It's just saying that, you know, exceeds the expectations,
which 2.8% unemployment asopposed to 3.4% expected.
I think both of those numbersare completely wrong, by the way
, like completely doctorednumbers.
You know they do that thingwhere when people stop, when
(31:23):
people go off of unemployment,they don't count them as
unemployed anymore, like youknow what I mean Like that kind
of like the net, like the waythe federal government does it.
So I really I look at thisarticle, I look at this, this
quote unquote study, and I'mjust like sounds like complete
bullshit to me, according toeverything I've seen everywhere
else.
So, anyway, what do you have toadd there?
I know you got some thoughtsabout it.
Chris Miles (31:45):
I mean it's back to the Microsoft thing.
It's, I mean, we're talkingabout Microsoft here, right,
it's one of the one of thebiggest kind of global employers
that exists on the planet,right?
So when they talk about doinglayoffs and purely to invest in
AI, I think there's theirjustification.
When they come to the table isprobably gonna be around, not
(32:07):
just like, hey, we need thiscapital just to fund AI.
I think they're probably also.
I mean, they're probably alsodoing the thing where they're
like, oh well, we can actuallyget rid of this job and just let
AI do it as well, right, soit's, I think, once we get into
the conversation about tokensand things like that, that's
consumed, maybe, maybe it doespay someone's salary, but
(32:30):
nonetheless, yeah, man, it'slike the tech, tech hiring,
exceeding expectations articleis also like you said.
I'm a little bit skeptical onthat because, like, there's no
way to know if it was justcompanies like shuffling deck
chairs around, things like that.
Like it's.
I mean, there's a lot ofjurisdiction around whether or
not you know when, when you'regoing to hire someone for a new
(32:53):
position, you know the job posthas to be made public for a
certain amount of days, etc.
Um, and we don't know like howmany times have you been on the
outside?
You see a job post and you'relike, oh, that looks amazing,
I'm gonna apply for that.
And then you, you like you're aperfect fit, you apply and you
hear crickets and the thing wasit was just earmarked for
someone internally that younever got to see, etc.
(33:15):
So I don't, I don't know, Iwonder if, like if we are
getting to that point where AIis, quote unquote, taking jobs,
like I wonder if they're findingpositions for people internally
to do something else andobviously they have to make a
posting for the position thatthey're going to move that
person into, while their youknow kind of rudimentary job was
replaced by AI.
(33:36):
But I mean, I don't know.
I mean, obviously there's onlyso many people in the world, but
tech is a growing employmentsector, so it's like it's hard
to not be skeptical about thesenumbers half the time when you
see so many layoffs on LinkedInand things like that.
So, yeah, not much, you know,kind of uh, good commentary to
(33:59):
add, but I think I'm with you,tim.
I'm very skeptical, uh, aboutthe um, uh, the reality of what
this means.
Tim McConnaughy (34:06):
It's uh well, and then you heard about.
You hear about ghost jobs allthe time, where companies are
posting jobs that they neverintend to fill just to look like
they're growing, so that youknow, know, wall street will,
wall street will react favorablyand like stuff like that
there's, it's out, there, it'sreal.
So somebody something I waslooking at I don't forget what
it was was estimating the actualunemployment rate is probably
(34:28):
closer to actually like 20,which is nuts, which is nuts if
you think about it.
So, but but then you keepseeing like here's a thousand
here, 9,000 people getting laidoff from this company, here's
5,000 from this company.
We're in the hundreds andhundreds and hundreds of
thousands of laid off peopleuntil the last year now.
So, yeah, actually, yeah, I cankind of see it.
(34:48):
You know, because all thosepeople did not go get another
job.
You still see people lookingfor jobs all over the place.
Chris Miles (34:55):
So yeah, maybe these jobs are being posted, for
you know hackers in Korea.
Tim McConnaughy (35:00):
That's right to come in and hack the network.
Chris Miles (35:05):
Yeah, exactly, all right With that, we'll go ahead
and wrap up the show.
So thank you so much.
If you stuck it out to the end,you must have enjoyed something
along the way, so please sharewith a friend.
We would love to expand theshow and our listener base there
, so we will go ahead and wrapit up and, like I said, we are
(35:26):
moving this to a fortnightlycadence for the overall show, so
you won't get another newsupdate for about a month from
today, and in two weeks we'llsee you back here with a, an
episode with kind of ourstandard episode with a guest
and things like that, where wetalk about an interesting topic.
So with that, we'll take itaway and we'll see you in a
couple weeks.
Goodbye.
Hello and welcome back to another episode of the
Cables to Clouds News Update,where Tim and I are going to go
through some of the networkingand cloud-adjacent tech news
articles that have come out overthe last couple of weeks and
give our opinions and commentsalike.
It is a very dark and earlyTuesday morning here in Sydney,
(00:35):
Australia, and, I'm assumingtemperament and beautiful
afternoon Monday afternoon inRaleigh, North Carolina.
But before we get into the news, I do want to take care of a
couple of housekeeping items.
Um, first of all, Tim, how areyou doing?
Tim McConnaughy (00:52):
What are we doing this now?
Chris Miles (00:54):
Well, I just realized now that we don't work
together anymore.
I probably check in with you alittle less than I, than I
normally would as a as acoworker.
So, yeah, what's new man?
You man, you just went to japanyeah, yeah, I just got back.
Tim McConnaughy (01:06):
Actually, I got a ton of uh pictures and stuff.
Uh, I'm putting together somekind of post because everybody's
asking me like, hey, how was it?
And it's like I have so manypictures, I have so many stories
, like I'm just not gonna typeor say that a hundred times.
So, yeah, I'll put somethingtogether and I'll post it.
Uh, it was good though.
I mean, I took the test, uh,took the jlpt n3.
I was amazed by how many peoplewere there, like that it was a
(01:29):
convention hall filled withdesks.
I could not believe how manypeople were there from different
countries, um, so, likejapanese, uh, nationals, uh
natives rather, cannot take thejlpt like the provisions test
for language.
Uh, only foreigners basically.
So it was all foreigners.
Chris Miles (01:45):
So that's even crazier, that it was that well
attended.
Yeah, non-nationals yeah, yeah.
Tim McConnaughy (01:50):
So I mean, if I did fail, it's probably because
the listening section, becausethe listening section had a
decent amount of echo because ofthe size of the of the room and
everything I was.
I I knew it was gonna be aproblem.
I just well, it is what it is,we'll see how it went.
Um, I'm hopeful, but yeah, I, Iwill know if I pass.
I probably just barely barelypassed, which a pass is a pass.
(02:11):
But uh, if I want to move on tothe next level, then it's not
gonna work anymore.
They definitely make it a lottighter per level.
So we'll see, man, we'll see.
But yeah, it's, it was good, itwas nice being in the same time
zone for a little bit of time,about 10 days or eight days or
so.
Chris Miles (02:26):
Yeah, yeah, that's true.
Very rare occasion that we'reon the same sleep schedule.
So that was nice.
Yeah, well, that's good.
Good luck, I'm sure you'll passwith fine colors.
We'll find out in a couple ofmonths, I guess.
But yeah, so before we get intoit, another update from us here
(02:46):
on the podcast.
So Tim and I, just with ourkind of ever busy schedules you
know me starting a new job andTim obviously has a lot of irons
in the fire as well We'vedecided to push the show back to
a fortnightly release for theoverall show.
So this includes our regularepisodes as well as our news
(03:07):
episodes.
So you'll be getting a Cablesto Clouds episode every two
weeks versus every week thatyou're seeing now.
So that will.
I think we want to take a shiftmore towards focusing on
quality of episodes rather thanquantity of episodes, and with a
weekly cadence it was startingto feel like we were rushing to
(03:27):
make sure we were getting peoplein to record episodes and
things like that, and I feellike the content was potentially
suffering a little bit.
So we want to shift back andmake sure we can kind of have a
little bit more thoughtfulconversations and things like
that.
So we're going to be releasingon a bi-weekly or fortnightly
cadence starting from this week.
So 16th of July will be thestart of that.
(03:50):
So you'll get another episodetwo weeks after that.
So just a little heads up andwe'll let everybody know on
social media as well.
And if you have any comments orfeedback on that, please reach
out.
I'm happy to hear from you guysas much as possible.
So look forward to that.
And with that out of the way,let's hop into the news.
Let's see All right.
(04:13):
So first up, we have an articlehere that was first brought to
us by Bleeping Computer.
So Ingram Micro, who is one ofthe world's largest IT
distributors we actually doquite a bit of work here in
Australia with them as wellsuffered a ransomware attack by
a ransomware group calledSafePay.
(04:34):
So basically, back on the 5thof July they've kind of reported
that there was a cybersecurityincident going on and that some
of their internal systems hadbeen compromised, and that's
some of their internal systemshad been compromised.
There was the attack lastedover it sounds like about four,
(05:03):
four days or so before theyactually were able to resume
their businesses or thebusinessgram systems.
It's kind of confusing thetimeline here because they came
out and said initially that theythought someone had gotten in
through compromised credentialsover their global to protect VPN
platform, which is kind of aVPN onboarding platform provided
(05:26):
by Palo Alto Networks.
And then Palo Alto Networkskind of commented that they were
looking at the incident withIngram as well, and then later
there was a comment saying thatthe VPN gateways were not
compromised using these passwordspray attacks that were called
out in the attack as well.
So I don't know exactly if thatmeans that's not how the
(05:47):
intruder got in or if that'sjust it wasn't a password spray
attack and maybe it was justsomething that was a little bit
more contained than that, butnonetheless it seems like this
was a pretty serious attack.
They were actually able to, orforced to, kind of shut down a
(06:08):
certain sector of theiroperations.
They were telling people to gohome and work from home,
although if the remote VPN wasthe platform that was
compromised, I'm sure they shutthat down too.
So if you were potentially anemployee of Ingram Micro, you
might have been sitting at hometwiddling your thumbs for a
little while.
So this seems like it's prettyserious.
(06:28):
Wish the best to all involved.
It doesn't seem like there'sbeen any money that's exchanged
hands here with regard to theransomware attacks.
So hopefully this is the lastof it.
But you know, sometimes whenyou hear that you know there
hasn't been any exchange offunds, sometimes there's, you
know, exploitation that comesout later due to data that was
(06:51):
was compromised during the event.
So yeah, so pretty rough onehere.
I'm glad to see them back innormal operations, though.
Any comments from you, tim?
Tim McConnaughy (07:01):
No, I agree.
Any comments from you, tim?
No, I agree.
The timeline on the articlesthat we were reading as kind of
published by Ingram and thencollected in these articles was
a little strange and it wasn'tclear.
I still, even now, they haven'tbeen clear on exactly how the
attack was successful.
They did initially believe itwas compromised credentials or
password spray.
(07:21):
Now Paolo specifically came outand said our global protect VPN
was not compromised or VPNgateways were not compromised.
But as a vendor, saying thatyour gateways were not
compromised, that doesn't saythat just means like the road
was not compromised, but likethe car could have been, you
know, yeah for sure it couldstill have been and probably was
(07:44):
.
Ultimately, you know, leaked orphished credentials, potentially
that that gave them entranceinto it, although you kind of
(08:06):
wonder, you know, from a zerotrust perspective, how leaked
credentials could lead to adecent amount of especially
apparently important enoughcomputers to end up being, you
know, ransomed.
So maybe there was like a lackof zero trust in the and this
would have been, I assume, on onIngram side cause.
Global protect VPN really justprovides the access, you know,
and then and then the policybased on how the company would
have configured it Right.
So yeah, it's.
It's very unclear from thearticle, but I would assume
(08:26):
Occam's Razor would suggest that, yeah, compromise, spearfish,
potentially credentials, andthen maybe a lack of true zero
trust enforcement that wouldallow lateral movement or
something like that that wouldlet an attacker compromise and
ransom multiple systems.
I do agree with you thatnowhere in here does it say
(08:47):
anything about them paying theransom.
In fact, because they were downfor something more like four
days or something, it's far morelikely that what they did was
contain and then replace thecompromised devices.
But yeah, good question Wasthere exfiltration that happened
during that time that's goingto have data sold on the dark
(09:09):
web for recouping some of thatmoney for the attack?
Or are there other backdoors orsomething potentially that have
been put in Because that couldhave happened?
Maybe they picked some devicesthat they didn't ransom.
Maybe they've kept them asbackup root kits or something
like that where they could comein back later.
(09:29):
So I think time will tell.
And this is the problemultimately, whenever you're
attacked by, whenever you suffera cyber attack, is that it is
truly difficult to really knowhow deep it goes.
Right, like you know someonewho has access to your systems.
You know, the ransomware partmight just be one method of
(09:49):
trying to extract value from theattack, right?
So yeah, I guess we'll see.
Yeah, best of luck to Ingram,and hopefully they actually
rooted it out and won't sufferany problems in the future.
Chris Miles (10:03):
Yeah, for sure.
I mean it seems like they did.
I mean we've talked about cyberattacks on this show quite a
bit and the thing is like thesethings are going to happen, you
know, from time to time, and Imean it looks like they had a
relatively good approach for,you know, instituting controls
to stop some of that.
(10:23):
Um, you know, instituting umcontrols to to to stop some of
that.
I think they did like a globalreset of every employee's MFA
across the entire company, whichis, you know, probably not fun
for the employees, but but uh,but it's good that they were
able to do that Right, um, andthat's uh, that's a way to at
least kind of um stop thebleeding for a bit.
So, um you know yeah, all thebest.
Tim McConnaughy (10:44):
Hopefully it works out okay.
Okay, so the next story comesto us from Network World.
Coreweave is acquiring CoreScientific.
No relation I know that theyboth I guess they're now related
, but previously no relation for$9 billion with a B, to quote
unquote power their AIinfrastructure push.
Billion with a B dollars, toquote unquote power their AI
infrastructure push.
This is interesting.
(11:05):
So CoreWeave is actually apublicly traded company that is
an AI cloud, an AI backed cloud.
Nvidia backs CoreWeave, likethey're invested or whatnot in
CoreWeave and they provide AIservices, right, just as a
on-demand AI services.
So that's CoreWeave.
Now CoreWeave is buying acompany called CoreScience
(11:27):
CoreScientific, rather.
Corescientific is actually acryptocurrency mining operation.
So you can already, you know,from a technology perspective,
you can obviously see the valueof an AI company acquiring a
farm full of GPUs.
So I think we talked about thisa while ago, like how
(11:52):
cryptocurrency farming and AItechnology, if you will, the
underpinning technology poweringAI with GPUs are.
First of all, they're bothextremely power hungry.
They're just like sucking theplanet dry of resources but also
how complimentary, you know, wewere kind of talking about like
who's going to win the gpu wars?
Right?
Like are we going to end upwith with crypto miners as the
(12:15):
winners, or is ai going to bethe winners?
I think we're seeing adefinitive move towards ai.
Um, partially because, first ofall, crypto has just been in the
game a lot longer, and thething with crypto is, the longer
a coin exists, the harder it isto continue to mine that coin.
So it's a what do they call it?
A diminishing return.
(12:35):
The whole thing is adiminishing return, so over time
it has to get harder and theROI, the return on investment
there is worse.
So the Network World articlepoints out the like OK, so
Quorum Scientific might havebeen struggling.
And then this is a good exitstrategy for them as a company
to just like cash out,essentially just get out of the
(12:56):
game and I couldn't agree more.
Honestly, if somebody's got $9billion to throw at it, I
couldn't imagine a better exitfor that company.
And then to just sell all yourshit to ai and the overpowered,
over, overblown, overpriced, uhtime that you can sell it right,
like well, it's as expensive asprobably ever you know ever
gonna be, and then and then exit.
(13:17):
So yeah, we'll, we'll see.
I mean, there's not much to seehere.
They sold it, so I guess we'llsee um if these guys get their
return on investment.
That's the real question.
Chris Miles (13:27):
Yeah, I mean, I guess, if you have somewhat of a
cynic point of view, it's funnyto see one technology that was
new on the scene and promised tomake people a lot of money now
is getting bought by another newtechnology that promises to
make a lot of people, a lot ofmoney that still is not
profitable, um, in roughly anycapacity.
So, um, some people are probablysaying, oh well, what's the
(13:50):
next thing?
What's going to buy core weaveand you know what.
It's five, six years orsomething like that.
Um, I mean not to be mean, butlike it's I'm, I'm kind of glad
to see some of the crypto miningand crypto adjacent stuff
starting to be on the way out.
Like it's I'm, I'm kind of gladto see some of the crypto
mining and crypto adjacent stuffstarting to be on the way out.
Like it sounds like even, um,some of the blockchain
technology stuff is is getting,yeah, um, overpowered by the um,
(14:14):
by the quantum computingconversations nowadays, right,
so it sounds like all of that iskind of being swept back under
the rug.
I don't know if it's going todie a fiery death or just slowly
wither away, but yeah, it'sjust funny to see this happening
, but yeah.
Tim McConnaughy (14:32):
Yeah, it's like pure speculation buying pure
speculation Like there'sactually, at least there's an
actual demonstrable product onthe AI side of it Right, at
least there's an actualdemonstrable product on the AI
side of it.
You know, like versus Bitcoinor or cryptocurrency, which is
truly like digging a hole inyour backyard and just picking
up rocks and telling everybodythey're worth a lot of money.
(14:55):
You know, like I don't, I don'tknow how else to you know,
there truly is no actual productassociated with blockchain
besides the blockchain itself,right, which is, as we've seen,
there's a lot of challenges,like proof of work and all this
other stuff that makes actualconsumption contracts and all of
(15:16):
that blockchain technologyactually not make a lot of sense
for the most part, right.
So, again, it's a product, it'spure speculation and at least
AI has, even if it's notprofitable, at least there's a
demonstrable value propositionof product to be produced.
But, like you pointed out, likethere's, it's still so far from
(15:37):
profitability.
I think they're still castingwide nets to try to figure out.
Okay, well, how do we, asuseful as all this is, how do we
make it profitable versus thecost of production of bringing
it to the market?
So I think we're going to stillbe searching for that golden
ticket for quite a while.
So, yeah, maybe something willcome, Maybe the next thing will
(15:57):
come along and buy that.
So I mean, there's a planetleft.
It's interesting.
Chris Miles (16:05):
Yeah, we'll see if this kind of stuff and the
stargate project and all thatkind of there's just so much
money being pumped into thisshit, man, it's like, yeah, it's
like amounts of money thatwe've never even had to like
fathom in any capacity and andwhere's it coming from?
Tim McConnaughy (16:19):
like truly, where is it?
Chris Miles (16:20):
all coming from?
Yeah, because the ai is notmaking any money yet.
So like what?
Yeah?
Yeah, it's all speculation.
So maybe we're due for anotherdepression or something We'll
see.
Tim McConnaughy (16:30):
It is the 20s.
Chris Miles (16:32):
Yeah, I know, right , it could happen, all right,
moving on, so we've alreadytalked about this a little bit
on one of our past shows, but itwas kind of just a rumored
thing and now it seems it's uh,it's actually come to uh, come
to fruition.
But arista um has basicallyannounced that they have um
(16:53):
acquired, or starting to acquireuh velo cloud from broadcom um,
which was uh previously, priorto being uh vmware being bought,
um was a v, a VMware product aswell, right?
So VelaCloud is a Vela orVMware.
It was VMware's SD-WAN offeringthat they had on the market was
, you know, relatively wideadopted for a while, which is, I
(17:16):
think, a good thing for Arista.
I don't think they had a strongSD-WAN portfolio prior to this,
so this will be something tokind of help round out their
kind of WAN.
I don't know if people wouldlike people always have their
unique names internally for whattheir products thing.
So this is going to fit intothat kind of WAN category and I
(17:36):
think Arista has been very knownfor being prominent in the data
center and maybe here and therefrom, you know, private clouds
or even public clouds, et cetera.
So I think this will be good tokind of expand their portfolio
with the the way in as well,which is good.
I got to say, man, it's like ifyou were, if you were, an OG
(17:57):
VeloCloud customer, you.
I'm curious to hear opinionsfrom them on what this, what
this ride has been like, like asif you came on being a VMware
customer.
I mean, let's be honest, likepeople that like VMware, that
doesn't necessarily mean they'regoing to like anything offered
by Broadcom or even Arista.
Those are kind of the companiesthat have their very you know
(18:20):
opinionated fans in certaincapacities and ones that don't.
So I'm sure they've been on awild ride, uh, over the over the
course of the past couple ofyears.
Um, but yeah, so I think thisis good.
Um, I think it does leave alittle bit of uncertainty on the
table for, um, things like,apparently, broadcom and
(18:41):
VeloCloud brought a SASEsolution to the market as well,
combining the SD-WAN and theSymantec offering from Broadcom,
which was also under theirumbrella.
So I don't know if that's justgoing to kind of get the axe.
I don't think it had an amazingamount of the market cap in the
(19:01):
SASE world, but you know, Idon't know what the future of
that is going to look like.
Maybe we'll have to look intothat.
And kind of on the back of this,they've also Arista have
announced that they've announcedor sorry, they've announced
that they've hired a new COO inTodd Nightingale, who's coming
over from Fastly.
(19:21):
It's where he's been the CEOfor the past three years.
So I'll say he kind of comesacross and say he's a networking
person, he's excited to getback into kind of the networking
world and things like that,which is good.
I think he kind of worked hisway up through the ranks at
Cisco for many years, so he'skind of coming back into that
(19:44):
camp, although probably workingfor someone that Cisco has not
been too fond of for quite awhile so that's kind of funny.
But you know, here we are.
I mean, I will say, with thisVeloCloud acquisition, things
like that, I think he's got hiswork cut out for him.
He's got a lot of, um, you know, strategic execution that's
(20:05):
going to need to, um, uh, kindof come to fruition and play out
for arista in this part, but,um, yeah, overall I think it's a
relatively good move for themand, uh, we'll have to see where
it goes.
But if you are, if you are anog velo cloud customer, when
they were arista and now you've,you've, you've come in, uh,
vmware when they were Arista andnow you've come you mean VMware
(20:26):
, sorry, vmware, yeah, sorry.
VMware VeloCloud customer, andnow you've made it through the
ranks over to Arista.
I'd be curious to hear opinionson what that transition has
been like, so reach out to us.
Tim McConnaughy (20:39):
Well, and more importantly, importantly, are
you still like, because when vm,when broadcom bought, you know,
vmware, um did they come outwith a hey, your velo cloud is
now 10 billion dollars.
You know, like, did that happen?
Chris Miles (20:52):
or yeah, maybe the maybe the bills will go.
Maybe they went up 10x andmaybe they'll come back down to,
like you know, 2x or somethingit's so interesting because I
think everybody I mean, how manytimes have we talked about this
?
Tim McConnaughy (21:06):
broadcom runs like a, like a uh, what's the
word?
A private equity firm?
Right, yeah, broadcom workslike a private equity firm.
That's how they make theirmoney.
They act like a private equityfirm.
So them spinning off bellowcloud is not surprising in the
least.
It probably was not a hugemoney maker for um, it was
probably one of those thingswhere VMware had it, because it
(21:26):
made loyal customers, kept intheir ecosystem and provided a
solution for them.
But if you come in and you'relike Broadcom, the PE company,
you're looking at it and you'resaying, well, this is a business
unit we're paying for.
That isn't making.
You know, we can't charge abillion, million billion dollars
for it, so let's just sell thebusiness unit and focus on the
parts that we can overchargecustomers for.
(21:48):
So that's not a surprise?
Actually I would not.
I guess it makes sense.
It's one of those things thatafter the fact you're like, yeah
, that makes sense.
But I wouldn't have thought atthe time about Arista buying
Velo, Because doesn't Aristahave some weird like off the
shelf, like something they builtthemselves for, like a STYN?
Chris Miles (22:06):
Yeah, I think that's.
The thing is Arista, like Isaid, being such a prominent
vendor in the data center world,it's like I've never really
heard too much about them havinga.
They have something.
I know they do.
They definitely did havesomething.
I remember when this wasrumored, I started looking up
and they had something.
But the thing is the way it waslisted it was like was vague
enough to where I just wouldn'tunderstand if it was an SD-WAN
(22:27):
product or not.
Tim McConnaughy (22:28):
Yeah, you know what I mean Pathfinder,
pathfinder, it was calledPathfinder.
Yeah, it was part of the CloudVision thing.
We did look this up when wefirst saw the rumors.
I remember now that was a whileback now, but yeah, pathfinder,
they called it.
Chris Miles (22:42):
I mean now they obviously have a very strong
portfolio in the SD-WAN campversus something that we weren't
even sure was SD-WAN right.
I mean SD-WAN has gotten to thepoint where it is table stakes.
Tim McConnaughy (22:53):
It really is Right.
Chris Miles (23:17):
So it's like you know, they talked about it.
It really is like India, so tosay a ton of MPLS still, but
yeah, so I don't think itnecessarily killed that, but
it's definitely like you got tohave your WAN has to have some
sort of intelligence built intoit now.
So we're definitely at thatpoint.
So I think they, rather thanyou know, kind of being behind
the eight ball on that.
I think it's probably good toget something with as much
(23:38):
recognition as VeloCloud.
Tim McConnaughy (23:40):
Yeah, I wonder if they got it cheap as well.
I mean relatively cheap, right?
Relatively cheap, because I'mpretty sure Broadcom was trying
to get rid of it.
Like I said, they'll probablyspin off other business units
that aren't profit.
They can't charge insane profitmargins on anyway, because
that's just their thing.
So, yeah, maybe, maybe aristawas like just opportunistic
(24:03):
about it, like, hey, ourpathfinder is okay, but we'd
much rather have a I.
So now it comes down tointegration.
Let's see, arista, whatarista's integration strategy is
and how they fix it and put itinto their cloud vision or their
, you know, their, theirplatform that they're using for
all the other, uh, smart stuffthat they're doing.
Chris Miles (24:22):
I mean sadly.
I wonder if this goes the exactsame route or different route
than Cisco, cisco, like, like.
Arista is very yeah, causeArista is very big on the
unified operating system, right.
Tim McConnaughy (24:35):
Like EOS.
Eos is their is like their oneis their one thing that runs
pretty much most of theirplatform.
Chris Miles (24:42):
Veloclide's obviously not going to run that,
so it's like do you merge thetwo.
Do you cause?
That headache for people, orcan you do it in a much better
fashion than Cisco did?
Are they going to learn thelesson?
Yeah, I don't know.
Tim McConnaughy (24:54):
Will they learn the lesson, because that really
hurt that.
That really hurt sdn adoptionuh, waiting, waiting to backport
it to uh ios xc.
Now, ultimately, I think it didactually do well for the
platform, like, I think, sdn,cisco, sdn is doing better, but
at the time, with vitela beingsuch a market leader, uh, the
stakes are different.
(25:15):
Now.
Vitela is not a like velo cloudwhatever, like sd-WAN doesn't
have this race that it hadeither.
Yeah, true, so like now, theycould probably take their time.
You know, they could justsupport existing customers and,
in the background, with nopressure whatsoever, work on
that integration.
I think that's probably whatwe'll see.
Chris Miles (25:35):
Yeah, I mean it feels like we're kind of
poo-pooing Cisco.
I think it was needed.
They needed to do it.
I think there was just some.
I mean hindsight's always 20-20, I guess.
Tim McConnaughy (25:44):
Oh, of course.
But, and knowing how thebusiness units at Cisco are
generally operated, I was notsurprised.
You know, I was at Cisco at thetime.
I was one of the you know, oneof the people you know we had a
that was working on it when itfirst launched, you know, within
(26:06):
like a few months of theacquisition and, yeah, I mean it
needed to happen.
I think a big complaint ofcustomers, and even internally
at the time, was that and thiswas this is all water under the
bridge, right?
Was that at the time, Vivitellahad a lot of things planned
that were new and exciting, andremember this is what six years
ago now or something, almostfive or six years now that it's
(26:26):
been and a lot of new featuresat the time that were brand new
features that weremarket-leading, going to be huge
features that would reallycement Vivitella as the leader
at that time.
Um, that really got put on theshelf, you know, and you know so
that this could all be movedover to ios xc and so you had
like a whole year where all thecompetition then was able to
(26:47):
catch up and and and, like velocloud being one of them,
actually a great example of oneof them I think that was when
the acquisition happened.
Chris Miles (26:55):
Right yeah, that's what.
Tim McConnaughy (26:56):
I'm saying it's not the same world.
Now my guess is that Aristawill probably take their time
and just support the existingcustomer base while they figure
out how to do.
But I do agree with you on theEOS thing.
That is the kind of their Oneof their big, huge things is
like hey, we've got ES one one,one operating system everywhere.
(27:17):
So yeah, man, I think we willsee it, but I don't.
I don't think they have thepressure to deliver as quickly.
Chris Miles (27:23):
I agree.
Yeah Well, we'll have to seewhat they do.
I think, I think this is.
I mean, this is just the painsof an acquisition, right, this
is what you have to go through,yeah.
Tim McConnaughy (27:38):
So we'll, we'll keep our ear to the ground and
we'll see what they end up doing.
Speaking of the pains oftechnology, so our next article
comes from Reuters and this isback in July 2nd.
So this actually, I believe,has already happened.
I've seen it on LinkedIn, a fewpeople already saying like hey,
I've been let go from Microsoft.
So Microsoft was announcingthey were going to lay off 4% of
the workforce.
Another cut Like Microsoft isjust like chopping thousands and
(28:01):
thousands of employees.
Chris Miles (28:06):
Now they had said they were going to put like
$9,000 per year to pay for theAI.
Tim McConnaughy (28:09):
I think this did come out to be about $9,000.
Yeah, yeah, and I think that'sall, yeah, something like that.
So, yeah, so they had pledged$80 billion in capital spending
and then part of that was to payfor that was going to be
cutting workforce.
So this was and we reported onthis when they announced, when
they first announced it monthsago and this is just like
fruition.
This is the next round,essentially, of something that
(28:32):
we're already working on.
So this is so interesting.
A lot of tech companies aredoing this right, robbing Peter
to pay Paul.
Basically, they're cuttingtheir headcount to pay for their
expansion into AI.
From the AI is going to be morethan what they're going to lose
in productivity and, and youknow, whatever outcomes that
(29:01):
they would from having humancapital.
Um, I, it's.
I wish I knew or couldunderstand what their thinking
was here, because it'scompletely speculative, right,
like there's been no actualthere.
There's been, there's still, tothis day, still now, there has
been no indicator that ai isgoing to be profitable to the to
(29:24):
the point of, like, what you'repaying at laying out for it now
, will it eventually becomeprofitable?
Maybe there has to be somemicronization of technology.
That's going to have to happen.
You know, whatever, whatever,but it's still very much in the
we've got a pitchfork and we'rethrowing money on the fire phase
.
Except in this case, it's notmoney we're throwing on the fire
, it's essentially people,people's livelihoods anyway.
(29:46):
But yeah, they're betting onthis, and so is Meta and Google
and Amazon.
All the big tech giants aregoing through this motion to pay
to basically keep up with theJoneses.
So it's not sustainable.
But yeah, but before we getinto that just I wanted to
there's one more article that Ithink is hilarious and we'll
(30:07):
cover, you know, we'll talkabout them together, and it's
one that, according to NetworkWorld and I forget which.
Let me just look real quick theI forget what it was.
Chris Miles (30:16):
It was CompTIA.
Tim McConnaughy (30:16):
CompTIA, that, and I forget which.
Let me just look real quick theI forget what it was.
It was CompTIA.
Comptia, that's right.
It was a CompTIA study sayingthat tech hiring exceeds
expectations in June.
So, like, literally, you havethese two articles right next to
each other saying that the techunemployment rate fell slightly
in June as companies listed455,000 job listings for tech
positions.
(30:37):
455,000 job listings for techpositions according to this tech
jobs report.
It really buries the lead onthis, saying that, like, tech
unemployment is low and it'slower than expected.
And then it goes on to pointout that, like, almost all of
the new job postings are aroundAI, like people that know AI,
people that are involved with AI, data science, et cetera, et
cetera.
So, and it also points out thatthose 455,000 listings are just
(31:00):
listings.
It's not really saying that,like, more people are being
hired than fired or whatnot.
It's just saying that, you know, exceeds the expectations,
which 2.8% unemployment asopposed to 3.4% expected.
I think both of those numbersare completely wrong, by the way
, like completely doctorednumbers.
You know they do that thingwhere when people stop, when
(31:23):
people go off of unemployment,they don't count them as
unemployed anymore, like youknow what I mean Like that kind
of like the net, like the waythe federal government does it.
So I really I look at thisarticle, I look at this, this
quote unquote study, and I'mjust like sounds like complete
bullshit to me, according toeverything I've seen everywhere
else.
So, anyway, what do you have toadd there?
I know you got some thoughtsabout it.
Chris Miles (31:45):
I mean it's back to the Microsoft thing.
It's, I mean, we're talkingabout Microsoft here, right,
it's one of the one of thebiggest kind of global employers
that exists on the planet,right?
So when they talk about doinglayoffs and purely to invest in
AI, I think there's theirjustification.
When they come to the table isprobably gonna be around, not
(32:07):
just like, hey, we need thiscapital just to fund AI.
I think they're probably also.
I mean, they're probably alsodoing the thing where they're
like, oh well, we can actuallyget rid of this job and just let
AI do it as well, right, soit's, I think, once we get into
the conversation about tokensand things like that, that's
consumed, maybe, maybe it doespay someone's salary, but
(32:30):
nonetheless, yeah, man, it'slike the tech, tech hiring,
exceeding expectations articleis also like you said.
I'm a little bit skeptical onthat because, like, there's no
way to know if it was justcompanies like shuffling deck
chairs around, things like that.
Like it's.
I mean, there's a lot ofjurisdiction around whether or
not you know when, when you'regoing to hire someone for a new
(32:53):
position, you know the job posthas to be made public for a
certain amount of days, etc.
Um, and we don't know like howmany times have you been on the
outside?
You see a job post and you'relike, oh, that looks amazing,
I'm gonna apply for that.
And then you, you like you're aperfect fit, you apply and you
hear crickets and the thing wasit was just earmarked for
someone internally that younever got to see, etc.
(33:15):
So I don't, I don't know, Iwonder if, like if we are
getting to that point where AIis, quote unquote, taking jobs,
like I wonder if they're findingpositions for people internally
to do something else andobviously they have to make a
posting for the position thatthey're going to move that
person into, while their youknow kind of rudimentary job was
replaced by AI.
(33:36):
But I mean, I don't know.
I mean, obviously there's onlyso many people in the world, but
tech is a growing employmentsector, so it's like it's hard
to not be skeptical about thesenumbers half the time when you
see so many layoffs on LinkedInand things like that.
So, yeah, not much, you know,kind of uh, good commentary to
(33:59):
add, but I think I'm with you,tim.
I'm very skeptical, uh, aboutthe um, uh, the reality of what
this means.
Tim McConnaughy (34:06):
It's uh well, and then you heard about.
You hear about ghost jobs allthe time, where companies are
posting jobs that they neverintend to fill just to look like
they're growing, so that youknow, know, wall street will,
wall street will react favorablyand like stuff like that
there's, it's out, there, it'sreal.
So somebody something I waslooking at I don't forget what
it was was estimating the actualunemployment rate is probably
(34:28):
closer to actually like 20,which is nuts, which is nuts if
you think about it.
So, but but then you keepseeing like here's a thousand
here, 9,000 people getting laidoff from this company, here's
5,000 from this company.
We're in the hundreds andhundreds and hundreds of
thousands of laid off peopleuntil the last year now.
So, yeah, actually, yeah, I cankind of see it.
(34:48):
You know, because all thosepeople did not go get another
job.
You still see people lookingfor jobs all over the place.
Chris Miles (34:55):
So yeah, maybe these jobs are being posted, for
you know hackers in Korea.
Tim McConnaughy (35:00):
That's right to come in and hack the network.
Chris Miles (35:05):
Yeah, exactly, all right With that, we'll go ahead
and wrap up the show.
So thank you so much.
If you stuck it out to the end,you must have enjoyed something
along the way, so please sharewith a friend.
We would love to expand theshow and our listener base there
, so we will go ahead and wrapit up and, like I said, we are
(35:26):
moving this to a fortnightlycadence for the overall show, so
you won't get another newsupdate for about a month from
today, and in two weeks we'llsee you back here with a, an
episode with kind of ourstandard episode with a guest
and things like that, where wetalk about an interesting topic.
So with that, we'll take itaway and we'll see you in a
couple weeks.
Goodbye.
Popular Podcasts
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
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.