SAN FRANCISCO—ARM Ltd. executives sketched the broad outlines of an ambitious but potentially risky move into high-performance computing Wednesday (Sept. 8), disclosing some details of the 2.5-GHz Cortex A15 processor, codenamed Eagle, at an event here.
The long-rumored move sets the stage for potential head-to-head competition against Intel in server and other high-performance computing markets. Intel’s relentless drive for performance has yielded increasingly power-hungry processors and prompted concern among OEMs and consumers about everything from power dissipation to cost to environmental impact.
“I’m very excited to see someone who knows power come up the performance curve,” said Mitch Markow, strategic processor technologist with Dell who participated in the event as a panelist.
Markow wouldn’t say whether Dell was committed to using the eight-way multicore capable A15 in a future Dell server product, but he said the company is looking at the device “very closely.”
There were precious few technical details of the Cortex A15 Eagle during the event; company executives said more details would be forthcoming late this year. But it is possible to piece together enough information to get some idea of the scope of the processor and to infer what ARM is planning. The bottom line is that the Eagle is not your smartphone’s ARM core.
A technology departure
To start at the fundamentals, the A15 core has been designed in conjunction with the development of the 32nm and 28nm process technologies that IBM, GlobalFoundries, and Samsung are taking into production.
This in itself is something of a departure for ARM — there was no mention of a TSMC version of the core. One presumes that market pressure will demand one, but the absence was striking, and perhaps is a signal of the market presence that GlobalFoundries and Samsung now command in the foundry space.
On this process roadmap, ARM has constructed a 2.5 GHz-capable superscalar core. These figures in themselves depict high single-thread performance, which ARM did not quantify beyond saying the single-thread performance of the A15 would be about five times that of the previous top-of-the-range Cortex A9.
There has also been exacting attention to energy conservation, including the ability to power-reduce individual stages of the execution pipelines on the fly and the ability to get the entire CPU into sleep mode in 10 microseconds, allowing a far more aggressive use of sleep than is possible with existing large-scale CPUs. Thus used either as a single CPU or in multicore configurations, the A15 should hit a very aggressive energy/operation point.
ARM in the cloud
From here on up, the Eagle is clearly aimed not at advanced smartphones or mobile media devices, but at the network infrastructure, server, and cloud-computing space. To begin with, the A15 architecture is designed for big-machine problems. The total address space is 1 terabyte, far in excess of the needs of mobile or embedded computing.
This space is augmented by hardware virtualization support. ARM did not give details, other than to say the hardware virtualization features would be supported by existing virtualization middleware products such as VMWare (seeARMv7 gets 40-bit, virtualization support)
Virtualization hardware will work hand-in-hand with ARM’s TrustZone security system to provide both the task mobility and the intertask security necessary for cloud-like computing environments and massive server farms.
The CPU core also is intended for multiprocessing. ARM showed illustrations of the core deployed in clusters of four CPUs sharing an integral L2 cache, and two of these clusters linked by the previously-announced AMBA-4 bus architecture.
Such a configuration, ARM said, would be fully coherent, easing virtualization and allowing threads to be distributed widely across multiple cores. Again, these are server-class features, not the stuff of embedded processing.
ARM CEO Warren East describes the Cortex A15 CPU
A further hint at thinking about very large systems is ARM’s statement that the core includes soft-error detection and recovery — presumably on the caches, rather than the logic circuitry — so that even as the total amount of memory in the system becomes enormous, up-time can remain reasonable.
The three ARM silicon partners working on the first implementations of the A15 — Samsung, ST-Ericsson, and Texas Instruments -- focused their presentations on high-end mobile devices where single- or dual-core implementations could move a hand-held computer out of the smartphone category and into the realm of serious local computing.
But the overall feature set of the Eagle architecture — virtualization support, the enlarged address space, hardware-supported coherent eight-way multiprocessing, and soft-error contro l— suggest another vision entirely.
That said, Eric Klein, vice president of technical planning for Nokia, said his handset company is keenly interested in the increased performance the A15 brings.
“When you talk about 5x (performance increase), we’re going to use every bit of that,” he said.
A lesson from history?
For all the buzz leading up to the event, some in the audience were left unimpressed.
“It’s predictable. It’s the next logical thing to do,” said Jeff Bier, president and founder of the technical consultancy BDTI. “They’ve got a chance (to make inroads in the server market)…but they were very careful about” not making aggressive claims about their server-market opportunities, he added.
Several observers compared ARM’s potential charge into servers with Intel’s 1990s-era attempts to own the communications-silicon space, which turned into a strategic retreat after several years.
Bier said what would have been more interesting would be to hear ARM CEO Warren East and Mike Inglis, general manager of the company’s processor division, describe real potential for innovative applications that take advantage of low-power / high-performance cores such as the A15.
ARM is holding in early November in Santa Clara, Calif. ARM Technology Conference during which ARM is expected to disclose more details.
Server Support Company provides the most flexible and cost effective server support services that can be custom tailored to fit your company’s needs. Basically, you've tried the rest now try the best server support company.
Kiran - I think you are 100% correct. ARM has to demonstrate its viability in these classes of applications to be believed. That has to come from server specific devices and more importantly systems based on those chips entering the market. Not sure what you mean by "A11" - if you mean ARM11, then the A15 is radically different. Ability to cluster more than 4 CPUs together, much deeper pipelines to support processor frequencies in the 2.5GHz etc, new on-chip interconnect bus etc.
A chinese company, Nufront, announced today a dual core Cortex-A9 device running at 2GHz. Marvell has discussed a quadcore device addressing similar frequencies being available this year, so the frequencies are starting to get interesting for a range of non-mobile applications. And getting past the 4GB physical address memory limitation which the A15 does, I would argue, further increases the applicability of ARM.
ARM's approach to enter into the high performance computing market like servers and networking equipment is not really aggressive. Cortex A15 doesn't really look anything different from their current highly used processor architectures such as A9 and A11. Anyways there will be a huge demand for the low power and high performance processors in the server segment since companies could save huge money. We need to wait and see few successful designs with A15 powered processors to know more details.
Weatherhead - you are absolutely correct. Cortex-A15 is a 32-bit processor. Server workloads tend to be quite different by application. Just like SeaMicro, a start-up OEM in the Bay Area, who is pursuing certain server applications with Atom, ARM and its partners see an opportunity to address applications that need modest processing performance. Of course, we are new to the space. You have every right to be skeptical until you see proof. That's what me and my team are working on every day.
I'm assuming from recent context that this is still a 32-bit processor (being a member of the ARMv7 Cortex family) with 40-bit physical addressing (i.e. 1 TB). How can it possibly complete in the "server space" like everyone is implying without a true 64-bit architecture? If this is v7 with physical addressing extensions you are still going to be running 32-bit processes. May be viable for a supercharged smart phone but I don't see it being competitive against existing Intel offerings in the server or even laptop space as those are all capable of running full 64-bit applications and operating systems as far as I know. Now if they extend the instruction set architecture to full 64 bit addressing and integer performance in like a "Cortex" ARMv8 architecture that would be something to talk about. It is interesting though that one of the advantages of ARMv7 is the ability to mix 16-bit ARM thumb 2 opcodes and full 32-bit ARM opcodes, how does that translate to a true 64-bit processor?
Disclosure: I work for ARM
Just to add to Gary's comment above
1) Due to the volumes and cost sensitivity of the phone market, a company will absolutely not use a processor that is over functional for the design. We still have companies licensing ARM11 for this space to address the needs of less functional handsets
2) Even once Cortex-A15 based silicon becomes available, there is quite a period time that elapses before handsets based on the devices are purchasable
3) These transitions take a long period of time. To illustrate, for 1Q2010, Cortex shipments represented 6% of the total ARM shipments. We absolutely expect the performance driven end of phone- and indeed other applications to transition over to the Cortex-A15, but it will take a long period of time
We have been hearing about ARM going into the HPC market for years now, but nothing has materialised yet. ARM is being very conservative for understandable reasons, but I wish them well in the high end uprocessor market segment as we need more competition there.
Three ARM licencees made presentations centered on mobile application of the A15. See the article.
The A15 provides more processing power than the A8/A9, which you will need to get into higher-end applications. The A8 and A9 are in presently shipping devices, while A15 designs probably won't appear until next spring, if not next summer.
For these types of applications, Marvell typically design their own ARM compatible CPUs. Server workloads vary quite substantially from application to application. Some with modest CPU performance requirements...Others that absolutely demand the maximum available CPU processing power available. A range of ARM Powered devices will appear; some Cortex-A9 based, some Cortex-A15 based, etc
Something missing in this picture?. If A15 is less than twice the area of A9, and 5 times performance, why would this not get into all cell phone, even running at half the speed of A9.
ARM is in troube...they have to make sure Apple, Qualcomm, Marvell, Broadcom Samsung, TI suceed with A9/A8 combo, yet need to introduce A15.
Intel always give true claims, however the x86 architecture is screwed up, the claims are never wrong......