Stage 2: Datacenter

Intel’s Datacenter Group SVP, Navin Shenoy, also took to the stage at CES in order to discuss some new products in Intel’s portfolio, as well as to deliver updates on ones that were disclosed last year. Back in 2018, Intel held its Datacenter Summit in August, where it lifted the lid on Cascade Lake, Cooper Lake, and 10nm Ice Lake. Along with this, we saw new instruction support for AI and security as the top two areas of discussion.

Cascade Lake: Get Yours Today

Intel’s first generation of Xeon Scalable processors, Skylake-SP, was launched over 18 months ago. We’ve been hearing about the update to that family, Cascade Lake-SP, for a while now, along with its brother Cascade Lake-AP and how it will tackle the market. The announcement today from Intel is that the company is now shipping Cascade Lake for revenue.

This means, to be crystal clear, that select customers are now purchasing production-quality processors. What this doesn’t mean is retail availability. These select customers are part of Intel’s early sampling program, and have likely been working with engineering samples for several months. These customers are likely the big cloud providers, the AWS / Google / Azure / Baidus of the world.

It’s worth pointing out that at Intel’s Datacenter Summit, they said that half of all of its Xeons sold were ‘custom’ processor configurations that were not sold though its distributors – these parts are often described as ‘off roadmap’. It is likely that when Intel says Cascade Lake-SP is shipping for revenue to select customers that they are likely to be purchasing these off-roadmap processors. They might be running at a higher TDP than Intel expects for the commercial parts, or have different core/cache/frequency/memory configurations as and when they are needed.

One of the big draws for Cascade Lake is Intel’s Optane DC Persistent Memory support, which will enable several terabytes of memory per socket, but also the hardware security patches for Spectre v2. Businesses who want to be sure their hardware is patched can guarantee security if it's in the hardware, rather than relying on a firmware/software stack. So this might be part of why Intel’s demand for 14nm CPUs is at an all-time high and outstripping supply – if a company wants to be 100% sure it is protected, they need the hardware with baked-in security.

The full retail launch of Cascade Lake is expected in 2019. Based on what we saw at Supercomputing in November, given by a rolling slide deck at the booth of one of Intel’s OEM partners, that time frame looks to be somewhere from March to May.

Nervana for Inference: NNP-I coming in 2019

To date, when Intel has discussed the Nervana family of processors, we have only known about them in the context of large-scale neural network acceleration. The idea is that these big pieces of silicon are designed to accelerate the types of compute commonly found in neural network training, at performance and power efficiency levels above and beyond what CPUs and GPUs can do. It has been disclosed that Intel is working on that family of parts, NNP-L, for a while now, and we are still waiting on a formal launch. But in the meantime, Intel is announcing today that it is working on a part that's optimized for inference as well.

There are two parts to implementing machine learning with neural networks: making the network learn (training), and then using the trained network on new information to do its job (inference). The algorithms are often designed such that the more you can train a network, the more accurate it is and sometimes the less computationally intensive it is to apply it to an external problem. The more resources you put into training, the better. But the scale of compute between training and inference is several orders of magnitude: you need a big processor for training, but don’t need a big processor for inference. This is where Intel’s announcement comes in.

The NNP-I is set to be a smaller version of the NNP-L and built specifically for inference, with Intel stating that it will be coming in 2019. Exact details are not being disclosed at this time, so we don't have any information on the interface (likely PCIe), power consumption, die size, architecture, etc. However, we can draw some parallels from Intel’s competition. NVIDIA has big Tesla V100 GPUs with HBM2 for training that can draw 300-350W each, with up to eight of them in a system at once. However for inference it has the Tesla P4, which is a small chip below 75W, and we’ve seen systems designed to hold 20 of NVIDIA's various inference processors at once. It is likely that this new NNP-I design is along the same lines.

Snow Ridge on 10nm: An SoC for Networking and 5G (Next-Gen Xeon-D?)

The Data Center Group will be making two specific announcements around 10nm. The first is disclosing the Snow Ridge family of processors, focused on networking and specifically targeting the wide array of 5G deployments coming up over the next decade. The purpose of Snow Ridge is to enable wireless access base stations and deployments, as well as functions required at the edge of the network, such as compute, virtualization, and potentially things like artificial intelligence.

Intel gave no other details, however going back in my mind, I realise that we’ve heard this before with Intel. They already have processors on their roadmap focused specifically on networking, with 40 GbE support and features like QuickAssist Technology to accelerate networking cryptography: the Xeon-D line of processors. This makes me believe that Snow Ridge will be the name for the next generation of Xeon D, either the Xeon D-2500 or Xeon D-3100, depending on the power envelope Intel is going for.

Given this assumption, and the fact that Intel has said that this is a 10nm processor, I suspect we’re looking at a multi-core Sunny Cove enterprise design with integrated networking MACs and support for lots of storage and lots of ECC memory. There’s an outside chance that it might support Optane, allowing for bigger memory deployments, although I wouldn’t put money on it at this stage.

Ice Lake Xeon Scalable on 10nm

To finish up Intel’s announcements, Nevin also talked about Ice Lake Xeon Scalable. At Intel’s Architecture Day, a processor was shown at the event that was described as Ice Lake Xeon, so this is just Intel repeating the fact that they now have working silicon in the labs. There is still no word as to how Intel is progressing here, with question marks over the yields of the smaller dies, let alone the larger Xeon ones. Working silicon in this case is just a functional test to make sure it works – what comes now is the tuning for frequency, power, performance, and optimizing the silicon layout to get all three. I’m hoping that Intel keeps us apprised of its progress here.

 

 

What Happened at CES 2018, and why CES 2019 is Different

A memory that will stick in my mind is Intel’s CES 2018 announcements. At the heart of the show, we wanted to know about the state of Intel’s 10nm process, and details were not readily available. 10nm wasn’t mentioned in the keynote, and when I tried to ask then-CEO Brian Krzanich about it, another Intel employee hastily cut in to the conversation saying that nothing more would be said. In the end we got a single sentence from Gregory Bryant at an early morning presentation the day after the keynote, and that sentence was only after 10 minutes of saying how well Intel was executing. That single sentence was to say that Intel was shipping 10nm parts in 2017, although so far only two consumer products (in limited quantities, and region specific) have ever been seen.

This year, coming off the back of Intel’s Architecture Day last month, shows that Intel is becoming more open to discussing future products and roadmaps. A lot of us in the press and analyst community are actively encouraging this trend to continue, and the contrast between CES 2018 and CES 2019 is clear to see. Companies tend to hide or obfuscate details when product execution isn’t going to plan; now that Intel is starting to open up with details, the outlook is clearly returning to one with more optimism.

Stage 1: Consumer
POST A COMMENT

61 Comments

View All Comments

  • KateH - Monday, January 07, 2019 - link

    dude, what?
    first off, what on earth does this have to do with "china and arabs" vs "america"?

    TSMC = based in Taiwan (it's even in the name)
    GlobalFoundaries = based in California

    second off, .... uhh nevermind. <facepalm>
    Reply
  • TheJian - Tuesday, January 08, 2019 - link

    America First pal ;) I'd rather have INTEL win than ANYONE from Korea or controlled by Arabs who spend a ton of money on terrorism, more than any other group in the world. Even allies like Saudi’s/UAE do it.
    https://www.washingtonpost.com/news/global-opinion...
    https://www.forbes.com/sites/forbesinternational/2...

    Global Foundries is ATIC...LOL. Face palm. You don't read past the product label do you? Just google ATIC fab global foundries. WE don't own it in USA (who owns you matters, not where you park the building), as AMD sold it ages ago.

    If another president takes over before trump breaks the China economy more, they will RULE many things if they keep stealing us blind and we keep giving up jobs. At some point we might be speaking Chinese in 20-50yrs (their plan is just that!). Trump is reversing that though, we had 16 steel factories before, then down to 4 up to Obama (nafta, China WTO entrance etc did this 70,000 factories close over this stuff). Back to 10 now already and growing. IF the wall is made of steel, more will re-open. Go 100% block, and they go bankrupt quick.
    https://www.scmp.com/economy/china-economy/article...
    If another pres gets in wall street will just tell them to stop China trade war (politicians are usually easily bought), and go back to fleecing of America. A few years down the road they take over NK, Taiwan, Japan, Philippines, etc with a president that acts like Obama when Russia took Crimea (gave them tents...LOL, instead of GUNS/AMMO).

    OH, and MU (a stock I own) bought a company in Taiwan, and they had traitors there working for China to STEAL tech.
    https://www.justice.gov/opa/pr/prc-state-owned-com...
    See the lawsuit FTC/DOJ just nailed them on. So, even NOT conquered, you still can't trust stuff that is CHINA related in ANY way, including Taiwan. You should be VERY careful with anyone owned or operated by Chinese. They are buying up Silicon Valley to steal everything too. IE, startups being bought right and left, to send it all home to China after we build it all.

    https://www.idahostatesman.com/news/business/artic...
    Traitors. Did you notice its Taiwan employees STEALING tech for CHINA (we’re talking Executive level here)? Nice local rundown of everything in the case. Currently China is blocked from buying AMAT or Lam Research tools that would allow them to compete with USA (can’t make high end stuff). But a loser president could change all that. A fighter like trump will stop it, at least for 6 more years if he builds that wall but if not, likely not elected again as that was the MAJOR campaign promise. Oh, and Limbaugh, Colter, Bongino etc will attack constantly as Rush already did it for 2 days on the wall saying trump was caving, trump let him know he wasn’t. Clearly both of these and more will attack, and trump will go down. Together, a few pundits can end trump if he doesn’t build the wall with the size of their audiences. A LOT of people will hear the message and vote him out. He was worried enough to comment on Rush’s two day bashing ;) Good, he should be worried. Fortunately the MU tech stolen was OLD anyway.

    https://www.bloomberg.com/news/articles/2019-01-05...
    Taiwan president and people afraid, and China thinks they OWN Taiwan (80% thinks China can piss off, 20% side with China though…ouch). Do you read news? Building Military islands to take over the trade routes etc. Google Xi Jinping’s proposal for unification. Uh, sounds like they want it all ;)
    https://www.scmp.com/comment/insight-opinion/artic...
    The article title says it all doesn’t it?
    “The day a US Navy vessel arrives in Kaohsiung is the day that our People’s Liberation Army unifies Taiwan with military force,” he told mainland media. OK, so they are actually saying they’re looking for an excuse to take them over…LOL.

    Ignorance is bliss I guess. Google China 2050 plan (must get Taiwan back by then). Nuff said? Stop watching fake news and READ more. :)
    Reply
  • wut - Tuesday, January 08, 2019 - link

    Why in the world are you posting all that here? It's not even directly related to the topic. You are wasting your time... I'm trying to help you, by the way. Post all that in a different topic, like under comments section of a political article and/or an article talking about intellectual property.

    Otherwise, you'd likely to be ignored. Why waste your breath.
    Reply
  • cheshirster - Wednesday, January 09, 2019 - link

    It is actually China and Korea vs America Reply
  • Santoval - Tuesday, January 08, 2019 - link

    As Lord Polonius had said, "Brevity is the soul of wit". Reply
  • TheJian - Tuesday, January 08, 2019 - link

    So I should take advice from a pompous idiot character from a Shakespeare book? Nah...
    https://en.wikipedia.org/wiki/File:Graham%27s_Hier...
    Next time say something worthwhile. There's a pic explaining EXACTLY how you do that. Which you didn't even get on the chart IMHO, but I guess someone could think you made the 1st or 2nd level...LOL.
    Reply
  • Retycint - Tuesday, January 08, 2019 - link

    Using ALL CAPS on words does not make your point stronger, it just exposes your vitriol Reply
  • bji - Tuesday, January 08, 2019 - link

    Used for single words or small phrases just adds emphasis to those words. That was not an ALL CAPS post, so get off your high horse. Reply
  • TheJian - Tuesday, January 08, 2019 - link

    You consider it shouting, I don't. That is YOUR problem, not mine. You assume vitriol, but you don't even know me...LOL. I don't think it makes my point stronger, I think it gets your attention. It did, or you wouldn't be whining about it RIGHT? ;)

    So you just did the same thing...Whatever. That must be vitriol, you mad bro? ROFLMAO. What exactly was cruel or bitter about my criticism of a site CLEARLY (see what I did there?) lacking in data that users of these cards would WANT to know? Anandtech hasn't wronged me personally in some way (I know they lie a lot & mislead the readers even more - but it doesn't hurt my purchases etc, I ignore the fake news...LOL), they are misleading their entire audience for ages as I noted clear back to 660ti and before. I just finally got figured people should be aware of the REAL data, and posted for 660ti and many times since. I guess all the LOL/ROFL, grins, winks etc are lost on some of you people :) It is typical of libtards to tell you to shut up, or "you can't say that", etc instead of following this:
    https://en.wikipedia.org/wiki/File:Graham%27s_Hier...
    ...So you can actually form a REAL argument to debate someone. As Jordan Peterson said, you can't discuss things without risking offending people. I don't care if you're offended by my caps nor what you perceive as "vitriol"...ROFL. Crawl back under your rock until you can figure out how to form an argument please. You're wasting my time ;)
    Reply
  • HStewart - Tuesday, January 08, 2019 - link

    I am surprise how people are so blind and believe that everything revolves around the desktops like in the 90's. Also it is important that we compare apples to apples - not Intel 14mm vs 7nm - we should comparing Intel 10nm vs 7nm. It also important one company 7nm vs 10nm does mean it better - it is important to note that nm rating is process and can vary per vendor. It also important to note architexture changes - and it likely Intel's new Sunny Cove could beat AMD even on 14nm.

    10nm is coming this year and high volume by holiday season. I think you are confusing 2020 date with new GPU from Intel.

    It would be foolish to believe that Intel is not coming back strong after AMD efforts with Ryzen. 2019 is going to be exciting year even with desktops on 10nm. Keep in mind that with CPU's using less and less power, it likely one day - desktop chips will be same as mobile chips.

    Please no CAPs - it just reminds me to much of Wccftech and is not reliable.
    Reply

Log in

Don't have an account? Sign up now