8 Comments

Summary:

A report says Google is building its own ARM-based custom server chips. This rumor has surfaced a few times in the last few years, but now the economics and technology are right for it to happen.

Inside a Google data center. Image courtesy of Google
photo: Google

A thinly sourced report from Bloomberg Thursday speculated that Google is building its own custom server chips using the ARM architecture. This isn’t a new rumor, but something that has popped up every now and again since Google purchased ARM-based server chip design firm Agnilux back in 2010. I’ve tried to chase it down at varying intervals over the years.

After all, Google makes its own switches (although it did buy third-party chips for the boxes) and isn’t afraid to develop its own gear to optimize on performance and cost. When you make money delivering websites, it only makes sense that you are going to make the process as efficient as possible, and building servers designed to handle the relatively small workload of serving web pages would be an excellent use case for the lower-power ARM-based chips. Google has several other large-scale workloads where a custom processor could also make sense.

What’s changed in the last few years is the economics and power of ARM processors. As more companies are building out ARM-based systems designed for servers, the cost of building custom-systems on a chip has dropped. ARM is a modular architecture as opposed to one where aspects like networking are integrated into the chip. Thus building a custom chip using ARM and other elements doesn’t cost as much as it once did.

I even said this would come to pass in a story written this summer, when I quoted AMD’s Andrew Feldman discussing how large webscale companies would soon be building their own custom chips. From that story:

Andrew Feldman, GM and corporate VP at AMD, explained this idea in a series of conversations with me over the last few weeks in which he estimated that one could build an entirely custom chip using the ARM architecture in about 18 months for about $30 million. He compared this to the three or four-year time frame and $300 million to $400 million in development costs required to build an x86-based server chip….

… in the ARM world things are different. Because any number of players can license an ARM core, each one is looking for points of differentiation outside the core (some with architecture licenses are looking to tweak the core itself) and can make chips with better I/O or specific workload accelerators. An example here is Calxeda, which is using an ARM core in highly dense servers but has also built a custom interconnect to send information rapidly between its hundreds of ARM cores.

So when the mega data centers look at the opportunities presented by ARM, it’s not as simple as buying a piece of silicon from Marvell or Applied Micro, or a Calxeda box from HP. According to Feldman, web giants are looking at co-developing ARM-based chips that will take advantage of the greater levels of customization offered outside of the CPU so they can optimize for their own applications’ needs.

And as ARM has evolved its cores to meet the needs of the server market with 64-bit capabilities and an emerging server software ecosystem, webscale companies are evaluating the instruction set for everything from storage to data processing. Heck, AMD has even taken an ARM license.

So if Google is indeed considering custom ARM-based chips for certain workloads, that makes sense. For an investment in the few tens of millions it might be able to optimize workloads that could help it speed up its service or lower the cost of providing it. With a more modular and licensable IP core, if ARM can do the job, why not take a look at using it? When reached via email today, Feldman said he still believe that for Google it’s not a question of if, but of when, although he suspects we’d see chips coming out in the next three to five years. However, to make that possibility it would have to start designing today.

This might be a blow to Intel, which currently counts Google as its fifth largest customer according to Bloomberg, but my hunch is that Google would still use some x86-based chips in its hardware where that makes sense. This isn’t the religious war between ARM and Intel so much as its rationalization of the cost of computing.

When you get a large enough number of low-cost machines all doing the same thing — as you see in cloud computing or these large webscale players — placing all your bets on general-purpose computing is like depending on a Major League pitcher to also be a damn fine cleanup hitter.

  1. The problem with this is that this is low volume and we got a bunch of ARM players building custom cores and that’s far more costly.
    Sure Google today can take a bunch of A57 cores and make a chip but in a couple of years there might be a few better ARM cores than what ARM offers (we know at least a few companies making custom ARM cores and there might be a few that we don’t know about – like AMD ,makes no sense for them not to go there). Also not everything is about the core and not so sure how fit Google is for everything else on the chip.
    It’s early on,the market is yet to mature but that won’t last so in the end maybe Google can go for semi-custom chips made by one of the bigger ARM players.
    Sure if Google wants to get more serious about making it’s own SoCs , they could but they would need to target a lot more than just some of their own servers to justify a far more serious investment. They might be better off targeting very low power for wearables since in phones the race is for performance and everybody went rather crazy on power consumption. Just Glass could justify the costs if Google stops messing it up and the product doesn’t remain vaporware.

    Share
  2. Google is building systems. If it is buying 0.XB now it will buy 3XB in three years. It is evident that future computers at Google server-farms will be heavily weighted toward Graphic Processing Units (16 GPUs per CPU) to handle the video/videoconferencing loads Google is planning. Intel has said that it will provide a foundry for future chips. That probably means that conceptual design of server-farms is proceeding at Google, with simulated performance of both ARM and Intel chip designs. Fast vector processing will require up to a Bbytes of on chip memory for each GPU. Google will not accept a we know what is best for you chip. If Intel won’t produce a near optimal chip at a competitive price Google will go elsewhere.

    Optimization by vector space methods and AI computations rely heavily on GPUs to serve the web, which by some estimates can be 80% dependant on Google server farms.

    Share
    1. AMD VP said that everything they now do is driven by graphics… AMD will end up being the dominant chip supplier to the cloud in a few years.

      Only Nvidia has competitive graphics, and they are way, way behind in integrated graphics, and lack the server experience of AMD.

      Share
  3. ARMdevices.net Friday, December 13, 2013

    Google surely has been experimenting with ARM servers for years now, and is probably ready to implement ARM servers in large volume, I bet Google already is testing significantly large volums of ARM servers, and this is mainly about cost. Not only cost to buy the hardware/performance, but most importantly about giant cost savings in power consumption thus in cooling. Google can perhaps design some fanless ARM server solution, low power means low heat means Google can save perhaps something like $100 million or more per datacenter per year in cooling and stuff like that. Using ARM Is the absolute only way Google can possibly keep up with the giantly increasing demand for more online services, the giant increase of website traffic, increase in search, increase in youtube video streams, increase in advertising usage, increase in Google Drive storage needs, all this increase can only be possible by using ARM technology. I think Google will design their own and also use several of the ones being developed by AMD, Calxeda, TI, Applied Micro and others, each ARM Server solution will be used in optimized fashion by each of Google many different optimized server service requirements, Intel should design ARM Processors and Intel should offer to Fab Google’s ARM Server chip also, that’s what Intel should do.

    Share
    1. Anonymous Coward Monday, January 20, 2014

      Cheaper cooling: sure. But also computing power density. Putting a few hundred ARM cores in the same space that at most 32 classical x64 cores would take up means data centers can provide a lot more computing power in the same physical space. That’s another cost reduction – a significant one, both in terms of initial investment and running costs.

      Share
  4. >>>” According to Feldman, web giants are looking at co-developing ARM-based chips that will take advantage of the greater levels of customization offered outside of the CPU so they can optimize for their own applications’ needs.”<<<

    Feldman's comment about 3 years from NOW, if Google starts today, are an indication that IF Google wants its own semi-custom chips sooner, they should partner with AMD.

    AMD’s 64-bit “Seattle” ARM processor brings best of breed hardware and software to the data center
    http://community.amd.com/community/amd-blogs/amd-business/blog/2013/12/12/amd-s-64-bit-seattle-arm-processor-brings-best-of-breed-hardware-and-software-to-the-datacenter

    Why do you suppose that code name is "Seattle"?

    What huge innovative cloud company is in Seattle? AMAZON!

    Amazon's James Hamilton: Why Innovation Wins
    AMD SeaMicro AMD SeaMicro

    Redmond is very close to Seattle… Does MSFT like that Intel has a monopoly on server chips?

    Share
  5. Make sense for Google to migrate to AMD X86-64 – APU – ARM – HUMA . Mmm might buy some AMD stock ?!.

    Share
  6. Google should just acquire AMD

    Share

Comments have been disabled for this post