Background — High Performance Attached Processors Handicapped By Architecture
The application of high-performance accelerators, notably GPUs, GPGPUs (APUs in AMD terminology) to a variety of computing problems has blossomed over the last decade, resulting in ever more affordable compute power for both horizon and mundane problems, along with growing revenue streams for a growing industry ecosystem. Adding heat to an already active mix, Intel’s Xeon Phi accelerators, the most recent addition to the GPU ecosystem, have the potential to speed adoption even further due to hoped-for synergies generated by the immense universe of x86 code that could potentially run on the Xeon Phi cores.
However, despite any potential synergies, GPUs (I will use this term generically to refer to all forms of these attached accelerators as they currently exist in the market) suffer from a fundamental architectural problem — they are very distant, in terms of latency, from the main scalar system memory and are not part of the coherent memory domain. This in turn has major impacts on performance, cost, design of the GPUs, and the structure of the algorithms:
Performance — The latency for memory accesses generally dictated by PCIe latencies, which while much improved over previous generations, are a factor of 100 or more longer than latency from coherent cache or local scalar CPU memory. While clever design and programming, such as overlapping and buffering multiple transfers can hide the latency in a series of transfers, it is difficult to hide the latency for an initial block of data. Even AMD’s integrated APUs, in which the GPU elements are on a common die, do not share a common memory space, and explicit transfers are made in and out of the APU memory.
HP today announced the Moonshot 1500 server, their first official volume product in the Project Moonshot server product family (the initial Redstone, a Calxeda ARM-based server, was only available in limited quantities as a development system), and it represents both a significant product today and a major stake in the ground for future products, both from HP and eventually from competitors. It’s initial attractions – an extreme density low power x86 server platform for a variety of low-to-midrange CPU workloads – hides the fact that it is probably a blueprint for both a family of future products from HP as well as similar products from other vendors.
Geek Stuff – What was Announced
The Moonshot 1500 is a 4.3U enclosure that can contain up to 45 plug-in server cartridges, each one a complete server node with a dual-core Intel Atom 1200 CPU, up to 8 GB of memory and a single disk or SSD device, up to 1 TB, and the servers share common power supplies and cooling. But beyond the density, the real attraction of the MS1500 is its scalable fabric and CPU-agnostic architecture. Embedded in the chassis are multiple fabrics for storage, management and network giving the MS1500 (my acronym, not an official HP label) some of the advantages of a blade server without the advanced management capabilities. At initial shipment, only the network and management fabric will be enabled by the system firmware, with each chassis having up two Gb Ethernet switches (technically they can be configured with one, but nobody will do so), allowing the 45 servers to share uplinks to the enterprise network.
Earlier this week Dell joined arch-competitor HP in endorsing ARM as a potential platform for scale-out workloads by announcing “Copper,” an ARM-based version of its PowerEdge-C dense server product line. Dell’s announcement and positioning, while a little less high-profile than HP’s February announcement, is intended to serve the same purpose — to enable an ARM ecosystem by providing a platform for exploring ARM workloads and to gain a visible presence in the event that it begins to take off.
Dell’s platform is based on a four-core Marvell ARM V7 SOC implementation, which it claims is somewhat higher performance than the Calxeda part, although drawing more power, at 15W per node (including RAM and local disk). The server uses the PowerEdge-C form factor of 12 vertically mounted server modules in a 3U enclosure, each with four server nodes on them for a total of 48 servers/192 cores in a 3U enclosure. In a departure from other PowerEdge-C products, the Copper server has integrated L2 network connectivity spanning all servers, so that the unit will be able to serve as a low-cost test bed for clustered applications without external switches.
Dell is offering this server to selected customers, not as a GA product, along with open source versions of the LAMP stack, Crowbar, and Hadoop. Currently Cannonical is supplying Ubuntu for ARM servers, and Dell is actively working with other partners. Dell expects to see OpenStack available for demos in May, and there is an active Fedora project underway as well.