中級 美國腔 31 分類 收藏
開始影片後,點擊或框選字幕可以立即查詢單字
字庫載入中…
回報字幕錯誤
>> Announcer: Live, from
Las Vegas, it's theCUBE,

covering AWS re:Invent 2017.
Presented by AWS, Intel, and
our ecosystem of partners.

>> Welcome back everyone,
live here in Las Vegas

it's AWS re:Invent, it's
theCUBE exclusive coverage

of re:Invent 2017, our
5th year covering Amazon,

watching the explosive growth.
I'm John Furrier, the co-founder
of SiliconANGLE Media,

with Justin Warren my co-host
here for this segment.

Our next guest is Raejeanne
Skillern, Vice President of the

Data Center Group and also General Manager
of the Cloud Service
Provider, part of Intel.

Welcome to theCUBE.
>> Thank you very much,
it's nice to be here again.

>> So, Intel inside the
Cloud is big growth for you,

what's the numbers look
like, your earnings

look pretty good, what's
the business update?

>> You know, it is
growing, and it's growing

in so many different angles,
it's coming from multiple fronts.
Part of it is just these killer workloads
that are driving the need for change.
Artificial Intelligence, immersive media,
network transformation
to handle all the data.

It is causing this real spur in growth,
from both the very largest
of Cloud service providers

across the ecosystem, and
one of the things when

I look at the growth, I
try to say, now where is it

really coming from?
Yes, Amazon is just
insanely growing fast and

the big guys are doing well,
their global expansion,

But, also, the next wave
of Cloud service providers,

and regions, and different countries,
are doing really well too,
they're actually growing faster,
so the whole ecosystem
is growing with demand.

>> Yeah, it's certainly not slowing down.
I can't see any way in which
that it would slow down

at all, if anything it's
just gonna get faster.

>> I think so, and when I track
trends like CapEx spending

and analyst predictions on the future,
consumer demand, everything indicates
we're gonna see another strong 5-10 years.
>> Justin: Yep.
>> So how do the Cloud service providers
shake out because if you look at this
creativity of software renaissance,
the interactivity, all
the amazing use cases,

from Thorn, finding
missing exploited children,

to, you know, high end HPC scientists,
genome sequencing stuff.
Massive use cases.
>> Raejeanne: Right.
>> So that's going to be a
tsunami of software developers,

Changing.
How does that shape the Cloud providers,
does it segment them into tiers,
how are you guys looking at that business?
>> Well I think it breeds
a lot of new opportunity.

So, yes, you see more
and more new services

and capabilities to
enable developers to more

easily develop code, to more quickly,
to get better utilization
out of their code.

But I also think just the
concept of a mocking performance

and ease, means new
insights, new services,

new capabilities, things that
we just couldn't, a developer,

we could not have done before.
So it's that, just like electricity,
when you make it cheaper
and easier to consume,

we use more of it.
I think Cloud computing is the same way.
So we're seeing kind of
a natural tam expansion,

market expansion, around
these things you just

couldn't do without the Cloud.
>> And new workloads, too.
>> And new workloads.
>> And all of those
workloads, Intel is basically

completely repositioning itself from just
being a chip company, it's like,
you're a data company now.
So, what are some of the things
that Intel's doing to help people
understand how they can make use of these
new software techniques,
and these new tools,

and the capabilities of the Intel chips
they're dependent on.
So what are some of the things
that Intel's doing there?

>> So obviously it starts at the bottom,
with the best silicon, not just compute,
but compute, network, and
storage, and accelerators,

for all different workloads.
We move up to the
platforms, we do a lot of

hardware engineering with the ecosystem,
with our top CSPs, actually many top CSPs,
we do direct engineering
work to get better

systems of architecture.
We have a host of libraries
we're creating that

ease of use, and mocking
performance out of it.

Reference architectures,
co-partners, and solutions.

So for us when we talk
about being a data company,

we can't do it from
just even being a chip,

we have to be a solutions partner,
with Cloud service providers, enterprise,
IOT edge solutions, we try to be there.
>> You guys always enabled some
very cool demos in the day,

even back on the PC you
always had interactivity,

you pioneered multimedia.
You always had that eye
on the applications.

Okay, on stage today all
this greatness is out there:

NFL demos, all this cool stuff.
That's really powering your business,
so talk about your relationship
with Amazon web services,

what's it like, how do
you guys engage with them,

what's the relationship,
'cause you guys are power

engine for AWS.
>> Yeah, well we try to be.
We wanna be the best performance
into their data center.

We've had a many, many, many
year deep relationship with Amazon.
It started from simple
co-development and engineering,

and is extended much more pervasively
across their environment and it poured
sometimes into the services.
We just, we want to one,
make sure that what we're delivering today
has been already optimized
specifically for their unique environment,
and that means we have to
start a year or two before,

if not earlier, to really
understand where they're going,

and get their feedback,
so that when we either

optimize a product or technology
across compute network

or storage, or create
something potentially

custom for them, right, it
takes a number of years of work,

so our partnership has a
lot of in-depth engineering,

it has a lot of future
and near-term enabling,

and then we hope to see an expansion.
What we really want to
do is use our technology

to differentiate Amazon.
We want their services
better because of the

technologies or capabilities we put in,
so wherever they wanna align in terms of
strategic investment and growth markets,
we wanna make sure the
silicon can enable it.

>> Intel always marched to the cadence
of Moore's law, and you guys have always
been a rapid machine, execution.
Amazon looks good, I
mean, they're executing.

>> Raejeanne: They're fast.
>> I mean, what's it like?
Share some story, I mean,
they're years ahead,

what's it like working with Amazon?
>> Well I think, I mean, they are fast.
It's amazing how quickly
they can move and innovate,

how rapidly those innovations
roll out to the market.

I will be honest, there are times
where we miss windows because we are slow,
and they just look at us
like, "Well we told you

"this four weeks ago, here it is, right?"
(laughing)
>> Can't design a chip in four weeks.
We don't have kubernetes,
no containers for chips.

>> I mean they are
just, they have it down.

So what we're trying to do is part of this
transition from being a client company,
to a Cloud and data center company,
and IOT company, means everything
we have to do is faster.
So we have to design
our chips more quickly,

we have to put in more modularity
for faster derivatives,
and we have to move at Cloud speed,
not classic Intel speed.
>> Right, yeah.
So what are some of the lessons
that other companies can take from Intel,
I mean it's a hardware company,
or it was originally a hardware company,
and now you've transitioned
to being a Cloud company,

and you're being pushed by Amazon
to move faster and faster.
So what are some of the lessons
that you can share with other companies
who are trying to start
moving at Cloud speed?

>> You know, I think, I
love Jeff Bezos' approach,

customer-obsessed.
If you aren't understanding
how the end customers,

starting with Amazon's customers,
but also then my customer, Amazon,
how they're using and
consuming technology,

we can't really create good technology.
I would say a lot of companies create
a great thing and then try
to go sell it at markets,

>> Yep.
>> Versus starting with the market,
and creating the specific thing.
The other thing we've learned,
I mean, Intel is a very
data driven company,

both in our decision making,
as well as our company growth,

this is, and we talk about it from
a developer envirogroup, but
it's the same iterate fast.

Fail quickly and move on.
You don't need perfect.
This is one of our learnings, right?
Don't wait six months for perfect,
move fast, get 70-80% of the way there,
I've heard governments
say they get 40-50 way

percent of there, make decisions,
because they have to move that quickly.
For military or other exercises.
So what we're trying to do is match that
type of speed as well.
>> It's a world of compute now,
I mean, I was at Alibaba Conference,
they had their Cloud
coachings hearing it here,

the same message: more compute.
They're not saying I
need more little chips,

they're saying I want more horsepower.
>> Raejeanne: Right.
>> And you guys just
announced the C5 instance,

recently, a couple weeks.
>> And the budget.
>> What is that gonna do?
I mean, it's fairly new.
>> Raejeanne: Yes.
>> What does that mean?
Is that gonna be an IOT edge opportunity?
Is it all workloads?
That's gonna be like, a pretty big deal.
What's your take on it?
>> It is, their C instant line has always
been for high performances
growing workloads.

We're seeing like, for HPC workloads,
anywhere from two to
four-and-a-half X the performance

moving from C4 to C5, right?
This is an instance
that can handle the most

demanding workloads from
high performance compute

to artificial Intelligence to others.
So, you know, we have
our latest and greatest

Intel ZN scalable process in there,
a very high performant
one, that we customized

specifically for their environment,
but then they do all this amazing software
work and efficiency work around it,
to really unlock.
I was really glad to see
when they talked about

those C5 instances launching,
25% on workload's price performance
up to 50% price performance improvements
on some others.
So, I mean, once again, when you can
take more compute and make
it more cost effective,

it's just a lot more things
people can do in the industry,

so we're very excited about that instance.
>> What's the biggest thing
in the past five years

that jumped out at you
in the massive change

in the industry?
Application, startups, business growth?
What amazes you?
>> Yeah, there's so much, I would kind of
combine it under what the industry calls
Digital Transformation.
You know, when I look
at it, one of the points

that always sticks in my mind,
is CEOs, 70% of them have
digital transformation

at the heart because the data suggests
that by the end of 2018, the top 40%
of the top 20, 40% of
the top 20 industries

are gonna be disrupted.
So that to me, that amount
of disruption happening,

and the company's trying
to disrupt themselves

whether it's healthcare,
retail, manufacturing,

oil and glass,
the use and pervasiveness of where
technology and Cloud can fit
has really kind of astonished me
and I love, once again, I love that
they're making what they do today better.
But the new things that they're doing,
I mean, in healthcare, right?
It's just amazing.
>> I mean, we used to use
the word back in the day

when I broke into the biz in the '80s,
Data Processing Department.
I mean, the Cloud is
one big data processor.

>> It is.
It is a compute power.
We call it the brain, right?
If Cloud is now ubiquitous, right?
It is, from public Cloud
to private to Edge,

and everything in between,
it's that brain, right?
>> John: It's the brain.
>> That's enabling the
compute, so we have to--

>> You people have
always been on the inside

of everything, so
congratulations on your success,

on the Cloud growth.
General manager of the
Cloud service provider

and Vice President of Data Center Group,
Raejeanne Skillern, here
inside theCUBE with Intel,

I'm Jeff Furrier with Justin Warren,
back with more live
coverage, here in Las Vegas,

for Amazon web services 2017 re:Invent.
More after this short break.
(techno music)
提示:點選文章或是影片下面的字幕單字,可以直接快速翻譯喔!

載入中…

Raejeanne Skillern, Intel | AWS re:Invent大會 (Raejeanne Skillern, Intel | AWS re:Invent)

31 分類 收藏
alex 發佈於 2018 年 1 月 6 日
看更多推薦影片

影片討論

載入中…
  1. 1. 單字查詢

    在字幕上選取單字即可即時查詢單字喔!

  2. 2. 單句重複播放

    可重複聽取一句單句,加強聽力!

  3. 3. 使用快速鍵

    使用影片快速鍵,讓學習更有效率!

  4. 4. 關閉語言字幕

    進階版練習可關閉字幕純聽英文哦!

  5. 5. 內嵌播放器

    可以將英文字幕學習播放器內嵌到部落格等地方喔

  6. 6. 展開播放器

    可隱藏右方全文及字典欄位,觀看影片更舒適!

  1. 英文聽力測驗

    挑戰字幕英文聽力測驗!

  1. 點擊展開筆記本讓你看的更舒服

  1. UrbanDictionary 俚語字典整合查詢。一般字典查詢不到你滿意的解譯,不妨使用「俚語字典」,或許會讓你有滿意的答案喔