Subject: Processors | November 8, 2017 - 02:03 PM | Ryan Shrout
Tagged: qualcomm, centriq 2400, centriq, arm
At an event in San Jose on Wednesday, Qualcomm and partners officially announced that its Centriq 2400 server processor based on the Arm-architecture was shipping to commercial clients. This launch is of note as it becomes the highest-profile and most partner-lauded Arm-based server CPU and platform to be released after years of buildup and excitement around several similar products. The Centriq is built specifically for enterprise cloud workloads with an emphasis on high core count and high throughput and will compete against Intel’s Xeon Scalable and AMD’s new EPYC platforms.
Paul Jacobs shows Qualcomm Centriq to press and analysts
Built on the same 10nm process technology from Samsung that gave rise to the Snapdragon 835, the Centriq 2400 becomes the first server processor in that particular node. While Qualcomm and Samsung tout that as a significant selling point, on its own it doesn’t hold much value. Where it does come into play and impact the product position with the resulting power efficiency it brings to the table. Qualcomm claims that the Centriq 2400 will “offer exceptional performance-per-watt and performance-per dollar” compared to the competition server options.
The raw specifications and capabilities of the Centriq 2400 are impressive.
|Centriq 2460||Centriq 2452||Centriq 2434|
|Process Tech||10nm (Samsung)||10nm (Samsung)||10nm (Samsung)|
|Base Clock||2.2 GHz||2.2 GHz||2.3 GHz|
|Max Clock||2.6 GHz||2.6 GHz||2.5 GHz|
|Memory Speeds||2667 MHz
|Cache||24MB L2, split
|23MB L2, split
|20MB L2, split
|PCIe||32 lanes PCIe 3.0||32 lanes PCIe 3.0||32 lanes PCIe 3.0|
Built on 18 billion transistors a die area of just 398mm2, the SoC holds 48 high-performance 64-bit cores running at frequencies as high as 2.6 GHz. (Interestingly, this appears to be about the same peak clock rate of all the Snapdragon processor cores we have seen on consumer products.) The cores are interconnected by a bi-directional ring bus that is reminiscent of the integration Intel used on its Core processor family up until Skylake-SP was brought to market. The bus supports 250 GB/s of aggregate bandwidth and Qualcomm claims that this will alleviate any concern over congestion bottlenecks, even with the CPU cores under full load.
The caching system provides 512KB of L2 cache for every pair of CPU cores, essentially organizing them into dual-core blocks. 60MB of L3 cache provides core-to-core communications and the cache is physically divided around the die for on-average faster access. A 6-channel DDR4 memory systems, with unknown peak frequency, supports a total of 768GB of capacity.
Connectivity is supplied with 32 lanes of PCIe 3.0 and up to 6 PCIe devices.
As you should expect, the Centriq 2400 supports the ARM TrustZone secure operating environment and hypervisors for virtualized environments. With this many cores on a single chip, it seems likely one of the key use cases for the server CPU.
Maybe most impressive is the power requirements of the Centriq 2400. It can offer this level of performance and connectivity with just 120 watts of power.
With a price of $1995 for the Centriq 2460, Qualcomm claims that it can offer “4X better performance per dollar and up to 45% better performance per watt versus Intel’s highest performance Skylake processor, the Intel Xeon Platinum 8180.” That’s no small claim. The 8180 is a 28-core/56-thread CPU with a peak frequency of 3.8 GHz and a TDP of 205 watts and a cost of $10,000 (not a typo).
Qualcomm had performance metrics from industry standard SPECint measurements, in both raw single thread configurations as well as performance per dollar and per watt. I will have more on the performance story of Centriq later this week.
More important than simply showing hardware, Qualcomm and several partners on hand at the press event as well as many statements from important vendors like Alibaba, HPE, Google, Microsoft, and Samsung. Present to showcase applications running on the Arm-based server platforms was an impressive list of the key cloud services providers: Alibaba, LinkedIn, Cloudflare, American Megatrends Inc., Arm, Cadence Design Systems, Canonical, Chelsio Communications, Excelero, Hewlett Packard Enterprise, Illumina, MariaDB, Mellanox, Microsoft Azure, MongoDB, Netronome, Packet, Red Hat, ScyllaDB, 6WIND, Samsung, Solarflare, Smartcore, SUSE, Uber, and Xilinx.
The Centriq 2400 series of SoC isn’t perfect for all general-purpose workloads and that is something we have understood from the outset of this venture by Arm and its partners to bring this architecture to the enterprise markets. Qualcomm states that its parts are designed for “highly threaded cloud native applications that are developed as micro-services and deployed for scale-out.” The result is a set of workloads that covers a lot of ground:
- Web front end with HipHop Virtual Machine
- NoSQL databases including MongoDB, Varnish, Scylladb
- Cloud orchestration and automation including Kubernetes, Docker, metal-as-a-service
- Data analytics including Apache Spark
- Deep learning inference
- Network function virtualization
- Video and image processing acceleration
- Multi-core electronic design automation
- High throughput compute bioinformatics
- Neural class networks
- OpenStack Platform
- Scaleout Server SAN with NVMe
- Server-based network offload
I will be diving more into the architecture, system designs, and partner announcements later this week as I think the Qualcomm Centriq 2400 family will have a significant impact on the future of the enterprise server markets.
Subject: General Tech, Systems | March 8, 2017 - 12:20 PM | Jeremy Hellstrom
Tagged: qualcomm, OCP, microsoft, falkor, centriq 2400, azure, arm, 10nm
Last December Qualcomm announced plans to launch their Centriq 2400 series of platforms for data centres, demonstrating Apache Spark and Hadoop on Linux as well as a Java demo. They announced a 48 Core design based on ARM v8 and fabbed with on Samsung's 10nm process, which will compete against Intel's current offerings for the server room.
Today marks the official release of the Qualcomm Falkor CPU and Centriq 2400 series of products, as well as the existence of a partnership with Microsoft which may see these products offered to Azure customers. Microsoft has successfully configured a version of Windows Server to run on these new chips, which is rather big news for customers looking for low powered hosting solutions running a familiar OS. The Centriq 2400 family is compliant with Microsoft's Project Olympus architecture, used by the Open Compute Project Foundation to offer standardized building blocks upon which you can design a data centre from scratch or use as an expansion plan.
Enough of the background, we are here for the specifications of the new platform and what can be loaded onto a Centriq 2400. The reference motherboard supports SOCs of up to 48 cores, with both single and dual socket designs announced. Each SOC can support up to six channels of DDR4 in either single or dual channel configurations with a maximum of 768GB installed. Falkor will offer 32 lanes of PCIe 3.0, eight SATA ports and a GbE ethernet port as well as USB and a standard 50Gb/s NIC. NVMe is supported, one design offers 20 NVMe drives with a PCIe 16x slot but you can design the platform to match your requirements. Unfortunately they did not discuss performance during their call, nor any suggested usage scenarios. We expect to hear more about that during the 2017 Open Compute Platform US Summit, which starts today.
The submission of the design to Open Compute Project ensures a focus on compatibility and modularity and allows a wide variety of designs to be requested and networked together. If you have a need for HPC performance you can request a board with an HPC GPU such as a FirePro or Tesla, or even drop in your own optimized FPGA. Instead of opting for an impressive but expensive NVME storage solution, you can modify the design to accommodate 16 SATA HDDs for affordable storage.
Qualcomm have already announced Windows 10 support on their Snapdragon, but the fact that Microsoft are internally running Windows Server on an ARM v8 based processor is much more impressive. Intel and AMD have long held reign in the server room and have rightfully shrugged of the many times in which companies have announced ARM based servers which will offer more power efficient alternatives. Intel have made huge advances at creating low power chips for the server room; AMD's recently announced Naples shows their intentions to hold their market share as well.
If the submission to the OPC succeeds then we may see the first mainstream ARM based servers appear on the market. Even if the Windows Server instances remain internal to Microsoft, the Centriq series will support Red Hat, CentOS, Canonical and Ubuntu as well as both GCC and LLVM compilers.
(click to seriously embiggen)
ARM may finally have reached the server market after all these years and it will be interesting to see how they fare. AMD and Intel have both had to vastly reduce the power consumption of their chips and embrace a diametrically opposite design philosophy; instead of a small number of powerful chips, servers of the future will consist of arrays of less powerful chips working in tandem. ARM has had to do the opposite, they are the uncontested rulers of low powered chips but have had to change their designs to increase the processing capabilities of their chips in order to produce an effective product for the server room.
Could Qualcomm successful enter the server room; or will their ARMs not have the necessary reach?
Subject: General Tech | December 7, 2016 - 01:25 PM | Jeremy Hellstrom
Tagged: qualcomm, centriq, centriq 2400, server
The days when AMD and Intel were the two choices to build a server with are long gone. The ARM architecture has been making serious inroads as various vendors have begun to offer various solutions utilizing ARM designs, up to and including AMD for that matter. Today, Qualcomm have joined these ranks, announcing their first processor family designed to power a server. The Centriq 2400 series is based on a 10nm process node, with up to 48 cores. As The Inquirer points out, this is a rather impressive shot across Intel's bow as Qualcomm will ship a 10nm FinFET before Intel does.
"The Qualcomm Centriq 2400 series, the first in the Centriq product family that Qualcomm has been working on for four years, has up to 48 ARMv8-compliant cores targeting compute-intensive data centre applications that require power efficiency and is built on the 10nm FinFET manufacturing processor."
Here is some more Tech News from around the web:
- Broadcom quietly dismantles its 'Vulcan' ARM server chip project @ The Register
- Sony kills off secret backdoor in 80 internet-connected CCTV models @ The Register
- Mikko Hypponen On The Death Of Antivirus @ Tech ARP
- Christmas 2016 Mega Worldwide Giveaway @ NikKTech