Showing posts with label ARM. Show all posts
Showing posts with label ARM. Show all posts

Monday, May 27, 2019

Arm unveils next gen Cortex CPU for 5G

Arm unveiled its latest Cortex-A77 CPU, promising a 20% performance improvement over current generation Cortex-A76 devices.

The new Cortex-A77, which targets smartphones, laptops, and other mobile devices, is positioned as Arm's Cortex-A77 5G-ready processor. It will be implemented in 7nm. The company notes that the past two generations of Cortex-A7x series processors (Cortex-A76 and Cortex-A77) have increased overall ML performance by 35x.

Arm is also introducing its Mali-G77 GPU, which leverages its new Valhall architecture to deliver a nearly 40% performance improvement over the previous Mali-G76 devices. Mali-G77 also boasts key microarchitecture enhancements including engine, texture pipes, and load store caches, which achieve 30% better energy efficiency and 30% more performance density. 

In addition, Cadence Design Systems announced that its full-flow digital and signoff tools support the new Arm Cortex-A77 CPU. Cadence delivered a complete 7nm Rapid Adoption Kit (RAK) that utilizes Arm 7nm POP™ IP libraries.

https://www.arm.com/company/news/2019/05/delivering-next-generation-ai-experiences-for-the-5g-world

Tuesday, March 12, 2019

First edge infrastructure dev platform for 7nm Arm Neoverse

Arm, Cadence Design Systems, and Xilinx introduced a development platform cloud-to-edge infrastructure based on the new Arm Neoverse N1.

The Neoverse N1 System Development Platform (SDP) is based on TSMC’s 7nm FinFET process technology and is also the industry’s first 7nm infrastructure development platform enabling asymmetrical compute acceleration via the CCIX interconnect architecture/

The joint solution is available to hardware and software developers for hardware prototyping, software development, system validation, and performance profiling/tuning. It includes Cadence IP for CCIX, PCI Express (PCIe) Gen 4 and DDR4 PHY IP.

The SDP includes a Neoverse N1-based SoC with an operating frequency of up to 3GHz, full-sized caches and generous amounts of memory bandwidth with the latest optimized system IP. The robustness of the SDP is ideal for development, debug, performance optimization and workload analysis on a wide range of applications including those for machine learning (ML), artificial intelligence (AI) and data analytics.

“The new Neoverse platforms deliver the performance and efficiency required to enable the cloud-to-edge infrastructure for a world with a trillion connected devices. Our ongoing SDP collaboration with Cadence, TSMC, and Xilinx truly enables developers with the system development tools necessary to innovate and deliver optimized Neoverse-based designs,” stated Drew Henry, senior vice president and general manager, Infrastructure Line of Business, Arm.

Sunday, January 6, 2019

Huawei unveils 7nm, 64-core ARM-based CPU

Huawei introduced Kunpeng 920, a 7nm ARM-based server CPU independently designed by Huawei based on ARMv8 architecture license.

The Kunpeng 920 integrates 64 cores at a frequency of 2.6 GHz. This chipset integrates 8-channel DDR4, and memory bandwidth exceeds incumbent offerings by 46%. System integration is also increased significantly through the two 100G RoCE ports. Kunpeng 920 supports PCIe 4.0 and CCIX interfaces, and provides 640 Gbps total bandwidth. In addition, the single-slot speed is twice that of the incumbent offering, effectively improving the performance of storage and various accelerators.

Huawei says its design significantly improves processor performance by optimizing branch prediction algorithms, increasing the number of OP units, and improving the memory subsystem architecture.

Huawei also introduced its TaiShan series servers powered by Kunpeng 920, including three models: one with a focus on storage, another on high density, and a third focused on balancing both requirements. The TaiShan servers are built for big data, distributed storage, and ARM-native application scenarios. The ARM architecture is best suited for these scenarios with advantages in many-core and performance per watt.

"With Kirin 980, Huawei has taken smartphones to a new level of intelligence. With products and services (e.g., Huawei Cloud) designed based on Ascend 310, Huawei enables inclusive AI for industries," William Xu noted. "Today, with Kunpeng 920, we are entering an era of diversified computing embodied by multiple cores and heterogeneity. Huawei has invested patiently and intensively in computing innovation to continuously make breakthroughs. We will work with our customers and partners to build a fully connected, intelligent world."

Huawei said it will continue to work with Intel to develop servers.

Thursday, January 3, 2019

Arm intros image signal processors

Arm introduced two image signal processors (ISPs) designed for everyday devices including drones, smart home assistants and security, and internet protocol (IP) cameras.

The Mali-C52 and Mali-C32 devices, which leverage Arm’s Iridix technology and other industry-leading algorithms for noise and color management, apply over twenty-five processing steps to each pixel, of which three critical ones deliver key differentiation in terms of image output quality. These include high-dynamic range (HDR), noise reduction and color management. Arms said its Mali-C52 and Mali-C32 ISPs efficiently deliver all three at high resolution and in real-time (e.g. 4k resolution at 60fps).

The Mali-C52 can be configured for two different optimizations - image quality or area. This flexibility allows our silicon partners to use the same IP and software across a range of products and use cases. The Mali-C32 is optimized specifically for area in lower-power, cost-sensitive embedded vision devices such as entry-level access control or hobby drones.


Thursday, October 18, 2018

NEC adopts Arm's Platform Security, develops facial recognition accelerator

NEC announced a strategic collaboration with Arm to drive the development of secure Internet of Things (IoT) solutions for Smart Cities using artificial intelligence (AI).

Under the alliance, NEC will adopt the Arm Platform Security Architecture, which is an industry framework for building securely connected devices. This framework includes the concept of threat modeling, architectural guidelines and implementation resources that reduce the cost, complexity, and risk associated with IoT security.

NEC is also developing a box-type accelerator device featuring real-time facial recognition that will be securely managed and provisioned by Arm Pelion Device Management. The accelerator will be equipped with the Arm Cortex-A53 processor. It will accelerate the device's facial recognition process and enable faster analytics by offloading the analytics workload carried out on servers or the cloud, thereby helping to reduce costs for networks and systems.

NEC says its facial recognition technology is the world's most accurate and fastest as measured by the U.S. National Institute of Standards and Technology (NIST) benchmark tests.

"We expect the new accelerator to promote further adoption of face recognition throughout many areas, including the public safety, entertainment and transportation fields, as it contributes to the security and convenience of consumers worldwide," said Naoki Hashitani, SVP, NEC Corporation.

Tuesday, September 18, 2018

Ampere releases its ARM-based CPU for data center servers

Ampere, a start-up based in Santa Clara, California, released its first generation Armv8-A 64-bit processors for data centers.

Key specs:
  • 32 Ampere-designed Armv8-A cores running up to 3.3 GHz Turbo
  • Eight DDR4-2667 memory controllers
  • 42 lanes PCIe 3.0 for high bandwidth I/O
  • 125W TDP for maximum power efficiency
  • TSMC 16nm FinFET process

Pricing:

32 cores @ up to 3.3 Ghz Turbo $850
16 cores @ up to 3.3 Ghz Turbo $550.

Ampere said it has been selected by Lenovo as well as several ODMs.

“We have made tremendous progress since our launch eight months ago, continuing to execute on our first and second generation products. More importantly, we are ahead of schedule on building out a robust, multi-product roadmap that meets the performance and features demanded by the cloud computing ecosystem,” said Renee James, chairman and CEO of Ampere. “We are partnering with world-class OEMs like Lenovo and several ODMs to address the unique design requirements for our cloud customers and meet their total cost and performance targets.”

Ampere emerges from stealth with 64-bit ARM server designs

Ampere, a start-up based in Santa Clara, California, emerged from stealth and revealed its plans for 64-bit ARM-based server processors aimed at hyperscale cloud applications and next-generation data centers.

Ampere Computing is headed by Renee James, the former president of Intel until 2016. Its team also includes three other Intel veterans: Atiq Bajwa, Chief Architect, and foremerly VP and GM of product architecture at Intel; Rohit Avinash Vidwans, Executive Vice President of Engineering, with 25 years experience at Intel including work on Xeon microprocessors for data center and enterprise servers; and Greg Favor, Senior Fellow, and 25 years experience at Intel including over 60 patents. Ampere is backed by The Carlyle Group.

Ampere said its processors will offer a high performance, custom core Armv8-A 64-bit server operating at up to 3.3 GHz, 1TB of memory at a power envelope of 125 watts. It will also offer mixed signal I/O features including PCIE Gen 3, SATA Gen 3, USB and workload accelerators, as well as the high-performance on-chip fabric. The processors are sampling now and will be in production in the second half of the year.

In October, The Carlyle Group acquired the compute business of AppliedMicro from MACOM, which earlier in 2017 acquired Applied Micro Circuits Corporation (AppliedMicro") in a deal was valued at approximately $770 million on the day it was announced.

In March 2017, AppliedMicro announced the sampling of its third generation 16-nanometer FinFET Server-on-a-Chip (SoC) solution, X-Gene 3.  The device is an ARMv8-A compatible processor that matches comparable x86 processors in CPU throughput, per-thread performance, and power efficiency while offering advantages in memory bandwidth and total cost of ownership. It features 32 ARMv8-A 64-bit cores operating at speeds up to 3.0 GHz, eight DDR4-2667 memory channels with ECC and RAS supporting up to 16 DIMMs and addressing up to 1TB of memory and 42 PCIe Gen 3 lanes with eight controllers.

Tuesday, August 28, 2018

ARM and Facebook join Yocto Project for Linux in embedded devices

The Yocto Project, the open source collaboration project that launched in 2011 to help developers create custom Linux-based systems for embedded products, announced ARM and Facebook as new platinum members, joining its 20 other member companies.

The Yocto Project provides a flexible set of tools and a space where embedded developers worldwide can share technologies, software stacks, configurations, and best practices to create tailored Linux images for embedded and Internet of Things (IOT) devices. An upcoming release is expected this fall.


“The next release will demonstrate Yocto Project’s ability to efficiently build and importantly, test complete Linux software stacks which are reproducible, easily audited and totally customizable in a maintainable way,” said Richard Purdie, Project Architect of the Yocto Project.

Tuesday, June 19, 2018

Mellanox supplies Infiniband for Sandia's ARM supercomputer

Mellanox Technologies will supply an InfiniBand solution to accelerate the world’s top Arm-based supercomputer to be deployed in Sandia National Laboratory in the second half of 2018.

The Astra supercomputer will include nearly 2600 nodes, and will leverage InfiniBand In-Network Computing acceleration engines. Astra is the first system in a series of the Vanguard program of advanced architecture platforms, supporting the US Department of Energy’s National Nuclear Security Administration (NNSA) missions.

“InfiniBand smart In-Network Computing acceleration engines will enable the highest performance and productivity for Astra, the first large scale Arm-based supercomputer,” said Gilad Shainer, vice president of marketing at Mellanox Technologies. “InfiniBand is the world-leading interconnect technology that supports the full range of compute and storage infrastructures, including X86, Power, Arm, GPUs, FPGAs and more. We are happy to support the Department of Energy’s efforts to explore new supercomputing platforms for their future needs.”

http://www.mellanox.com

Monday, February 5, 2018

Ampere emerges from stealth with 64-bit ARM server designs

Ampere, a start-up based in Santa Clara, California, emerged from stealth and revealed its plans for 64-bit ARM-based server processors aimed at hyperscale cloud applications and next-generation data centers.

Ampere Computing is headed by Renee James, the former president of Intel until 2016. Its team also includes three other Intel veterans: Atiq Bajwa, Chief Architect, and foremerly VP and GM of product architecture at Intel; Rohit Avinash Vidwans, Executive Vice President of Engineering, with 25 years experience at Intel including work on Xeon microprocessors for data center and enterprise servers; and Greg Favor, Senior Fellow, and 25 years experience at Intel including over 60 patents. Ampere is backed by The Carlyle Group.

Ampere said its processors will offer a high performance, custom core Armv8-A 64-bit server operating at up to 3.3 GHz, 1TB of memory at a power envelope of 125 watts. It will also offer mixed signal I/O features including PCIE Gen 3, SATA Gen 3, USB and workload accelerators, as well as the high-performance on-chip fabric. The processors are sampling now and will be in production in the second half of the year.

In October, The Carlyle Group acquired the compute business of AppliedMicro from MACOM, which earlier in 2017 acquired Applied Micro Circuits Corporation (AppliedMicro") in a deal was valued at approximately $770 million on the day it was announced.

In March 2017, AppliedMicro announced the sampling of its third generation 16-nanometer FinFET Server-on-a-Chip (SoC) solution, X-Gene 3.  The device is an ARMv8-A compatible processor that matches comparable x86 processors in CPU throughput, per-thread performance, and power efficiency while offering advantages in memory bandwidth and total cost of ownership. It features 32 ARMv8-A 64-bit cores operating at speeds up to 3.0 GHz, eight DDR4-2667 memory channels with ECC and RAS supporting up to 16 DIMMs and addressing up to 1TB of memory and 42 PCIe Gen 3 lanes with eight controllers.

“We have an opportunity with cloud computing to take a fresh approach with products that are built to address the new software ecosystem,” said James. “The workloads moving to the cloud require more memory, and at the same time, customers have stringent requirements for power, size and costs. The software that runs the cloud enables Ampere to design with a different point of view. The Ampere team’s approach and architecture meets the expectation on performance and power and gives customers the freedom to accelerate the delivery of the most memory-intensive applications and workloads such as AI, big data, storage and database in their next-generation data centers.”

Saturday, October 28, 2017

MACOM divests AppliedMicro’s Compute Business

MACOM reached an agreement to sell the Compute business it acquired in its AppliedMicro acquisition earlier this year to Project Denver Holdings LLC , a new company backed by The Carlyle Group. Financial terms were not disclosed, but MACOM said it will hold a minority equity ownership interest in the new company.

“After a thorough review process, we are very excited about the sale of the Compute business and the opportunity it provides for both the employees of that business and our shareholders,” said John Croteau, MACOM’s President and Chief Executive Officer. “The Carlyle Group is one of the world’s largest and most successful global investment firms, with over $170 billion in assets under management. I believe NewCo’s exceptional leadership team and Carlyle’s backing combines the necessary elements to make the business a long-term success, and through a minority equity ownership MACOM is positioned to participate in the company’s long-term value creation.”









  • In January 2017, MACOM Technology Solutions Holdings acquired Applied Micro Circuits Corporation (AppliedMicro") for approximately $8.36 per share, consisting of $3.25 in cash and 0.1089 MACOM shares per share of AppliedMicro. The deal was valued at approximately $770 million on the day it was announced..

    MACOM said it made the acquisition to accelerate its growth in optical technologies for Cloud Service Providers and Enterprise Network customers.
  • MACOM and AppliedMicro's pro forma combined TTM revenue was approximately $709 million including AppliedMicro's Compute business, or approximately $644 million excluding the Compute business
AppliedMicro's Connectivity business is highly complementary to MACOM's product portfolio, through the addition of market-leading OTN framers, MACsec Ethernet networking components, and the industry's leading single-lambda PAM4 platform.


  • In March 2017, the company announced the sampling of its third generation 16-nanometer FinFET Server-on-a-Chip (SoC) solution, X-Gene 3.  The device is an ARMv8-A compatible processor that matches comparable x86 processors in CPU throughput, per-thread performance, and power efficiency while offering advantages in memory bandwidth and total cost of ownership. It features 32 ARMv8-A 64-bit cores operating at speeds up to 3.0 GHz, eight DDR4-2667 memory channels with ECC and RAS supporting up to 16 DIMMs and addressing up to 1TB of memory and 42 PCIe Gen 3 lanes with eight controllers. The processor is expected to have a performance that is up to six times that of the currently shipping X-Gene family of products.

Wednesday, October 25, 2017

Intel ships Stratix 10 FPGA with ARM Cortex-A53

Intel has begun shipping a high-end FPGA family with an integrated quad-core ARM Cortex-A53. 

The ARM-based Intel Stratix 10 FPGA, which packs more than 1 million logic elements (MLE) with an integrated ARM processor, could serve multiple application categories, such as 5G wireless communication, software defined radios, secure computing for military applications, network function virtualization (NFV), and data center acceleration.

For NFV, Intel said its new FPGA can handle the high-speed data path while the integrated processors enable low latency transactions needed to manage flow tables for control plane processing. With hardware acceleration, Intel Stratix 10 SX FPGAs provide a heterogeneous computing environment to create optimized, low latency accelerators.

“With Stratix 10 SX, Intel reaffirms its ‘all in’ commitment to SoC FPGA devices combining integrated, flexible ARM cores with high-performance Intel Stratix® 10 FPGAs,” said Reynette Au, vice president of marketing, Intel Programmable Solutions Group. “We now provide a wide set of options for customer needing processors and FPGAs, with device offerings across the low-end, mid-range and now, high-end FPGA families, to enable their system designs.”

Sunday, October 22, 2017

Rambus validates DDR4 for Arm-based data centers

Rambus validated the interoperability of its DDR4 PHY and the Arm CoreLink DMC-620 Dynamic Memory Controller, which is a fast, single-port Coherent Hub Interface (CHI) for transferring data from its CoreLink CMN-600 (Coherent Mesh Network) to the Rambus DDR4 memory PHY. Together, these IP blocks offer speeds of up to 3200 Mbps, the highest performance memory speed available on the market. Both are DFI 4.0 compliant, allowing the PHY and memory controller to interoperate.

“Design teams face complex challenges in scaling the number of computing cores for advanced datacenter SoCs, while minimizing integration and testing time to ensure faster time-to-market,” said Jeff Defilippi, senior product manager, Infrastructure Business Unit, Arm. “Our collaboration with Rambus removes another degree of difficulty in designing purpose-built SoCs, resulting in higher-performing systems built for the most demanding cloud and enterprise workloads.”

https://www.rambus.com/


Wednesday, October 4, 2017

Mellanox announces software-defined SmartNIC adapters based on ARM

Mellanox Technologies announced its BlueField family of software-defined SmartNIC adapters, designed for scale-out server and storage applications.

The new adapters leverage embedded ARM processor cores based on the company's BlueField system-on-chip processors and accelerators in the network interface card (NIC).

Key features of the BlueField intelligent adapters:

  • 2 network ports of Ethernet or InfiniBand: 10G/25G, 40G, 50G or 100Gb/s options
  • RDMA support for both InfiniBand and RoCE from the leader in RDMA technology
  • Accelerators for NVMe-over-Fabrics (NVMe-oF), RAID, crypto and packet processing
  • PCI Express Gen3 and Gen4, with either x8- or x16-lane configurations
  • Integrated low-latency PCIe switch with up to 8 external ports for flexible topologies
  • Up to 16 ARMv8 Cortex A72 processors with 20MB of coherent cache
  • 8 – 32GB of on-board DDR4 DRAM
  • Comprehensive virtualization support with SR-IOV
  • Accelerated Switching and Packet Processing (ASAP2) OVS offloads
  • Multi-host and SocketDirect™ enabling a single adapter to support up to four CPU hosts
  • Multiple server form-factor options including half-height, half-length PCIe and other configurations


Mellanox said its new BlueField SmartNIC could be used for a range of applications, including Network Functions Virtualization (NFV), security and network traffic acceleration. The fully programmable environment and DPDK framework support a wide range of standard software packages running in the BlueField ARM subsystem. Examples include: Open vSwitch (OVS), Security packages such as L3/4 firewall, DDoS protection and Intrusion Prevention, encryption stacks (IPsec, SSL/TLS), traffic monitoring, telemetry and packet capture.

“Our BlueField adapters effectively place a Computer in Front of the Computer,” said Gilad Shainer, vice president marketing, Mellanox Technologies. “They provide the flexibility needed to adapt to new and emerging network protocols, and to implement complex networking and security functions in a distributed manner, right at the boundary of the server. This brings more scalability to the data center and enhances security by creating an isolated trust zone.”

Friday, May 5, 2017

China Mobile, ARM, Cavium and Enea Validate NFV test cases

China Mobile, the largest mobile carrier in China with over 850 million subscribers, ARM, Cavium and Enea announced an agreement covering collaboration in the China Mobile Open NFV Testlab leveraging Enea's OPNFV-based commercial NFV Core platform and Cavium's ARM-based ThunderX workload-optimised data centre server processors.

Through the agreement, China Mobile's Open NFV Testlab will host the platform as part of the operator's Telecom Integrated Cloud (TIC) initiative. The work will specifically encompass validation of a range of NFV tests cases, including virtualised CPE (vCPE), vBRAS, vEPC and vIMS, while also supporting development and integration within the Open Network Automation Platform (ONAP) project.

Enea NFV Core is a carrier-grade virtualisation software platform based on OPNFV and OpenStack. It is designed to enable the deployment and management of vCPE network functions in central offices and data centres utilising generic hardware platforms. The NFV Core is optimised for the vCPE use case and central office deployments and is designed to offer the performance, reliability and flexibility required in next generation telecom networks.
Commenting on the agreement, Raj Singh VP and GM, network and communication group at Cavium, said, "(The) collaboration with China Mobile will enable key NFV functionality and drive NFV towards large scale deployment… telco applications with compute, I/O and real time processing requirements demand scalable and optimised processing solutions… advanced COTS hardware such as Cavium's ThunderX server processors… provide scalable NFV solutions using standard software and ecosystems".


While Noel Hurley, VP and GM, networking and servers, business segments group at ARM, noted, "ARM and its ecosystem of partners (are) committed to enabling OPNFV to bring efficient and cost-effective compute power for data networks… the network pipeline must be expanded at scale to support new computing across all markets in the most efficient way... the ecosystem delivers efficiency through integrated solutions that demonstrate the performance-per-watt, density and TCO provided by ARM technology".

Tuesday, May 2, 2017

Iliad's Online launches cloud service based on Cavium ThunderX

Web hosting provider Online, a wholly-owned subsidiary of French telecom company Iliad Group, announced the commercial deployment of server platforms based on Cavium's ThunderX workload-optimised processors as part of its Scaleway cloud service offering.

Online offers a range of services to Internet customers worldwide including domain names, web hosting, dedicated servers and hosting in its data centre, and with several hundred thousand servers deployed is one of the largest web hosting providers in Europe.

For the deployment, Online is using dual socket, 96 core ThunderX based platforms as part of the Scaleway IaaS cloud offering. The Scaleway cloud platform is supported by Ubuntu 16.04 OS, including LAMP stack, Docker, Puppet, Juju, Hadoop and MAAS, and also provides support for standard features of the Scaleway cloud including flexible IPs, native IPv6, Snapshots and images.

Cavium's ThunderX products offer a 64-bit ARMv8-A based server processor designed for data centre and cloud applications. The devices feature custom cores, single and dual socket configurations, and high memory bandwidth and memory capacity. The products also include hardware accelerators, integrated high bandwidth network and storage IO, virtualised core and IO functionality and a scalable high bandwidth, low latency Ethernet fabric.

ThunderX products are compliant with ARMv8-A architecture specifications, as well as with ARM's SBSA and SBBR standards, and supported by major OS, hypervisor and software tool and application vendors.

Earlier in the year, Cavium announced it was collaborating with Microsoft to evaluate and enable a range of cloud workloads running on its flagship ThunderX2 ARMv8-A data centre processor for the Microsoft Azure cloud platform.


As part of the partnership, the companies demonstrated web services on a version of Windows Server developed for Microsoft's internal use running cloud services workloads on ThunderX2. The server platform was based on Microsoft Project Olympus open source, hyper-scale cloud hardware design.

Friday, March 24, 2017

Microsoft's Project Olympus provides an opening for ARM

A key observation from this year's Open Compute Summit is that the hyper-scale cloud vendors are indeed calling the shots in terms of hardware design for their data centres. This extends all the way from the chassis configurations to storage, networking, protocol stacks and now customised silicon.

To recap, Facebook's newly refreshed server line-up now has 7 models, each optimised for different workloads: Type 1 (Web); Type 2 - Flash (database); Type 3 – HDD (database); Type 4 (Hadoop); Type 5 (photos); Type 6 (multi-service); and Type 7 (cold storage). Racks of these servers are populated with a ToR switch followed by sleds with either the compute or storage resources.

In comparison, Microsoft, which was also a keynote presenter at this year's OCP Summit, is taking a slightly different approach with its Project Olympus universal server. Here the idea is also to reduce the cost and complexity of its Azure rollout in hyper-scale date centres around the world, but to do so using a universal server platform design. Project Olympus uses either a 1 RU or 2 RU chassis and various modules for adapting the server for various workloads or electrical inputs. Significantly, it is the first OCP server to support both Intel and ARM-based CPUs. 

Not surprisingly, Intel is looking to continue its role as the mainstay CPU supplier for data centre servers. Project Olympus will use the next generation Intel Xeon processors, code-named Skylake, and with its new FPGA capability in-house, Intel is sure to supply more silicon accelerators for Azure data centres. Jason Waxman, GM of Intel's Data Center Group, showed off a prototype Project Olympus server integrating Arria 10 FPGAs. Meanwhile, in a keynote presentation, Microsoft Distinguished Engineer Leendert van Doorn confirmed that ARM processors are now part of Project Olympus.

Microsoft showed Olympus versions running Windows server on Cavium's ThunderX2 and Qualcomm's 10 nm Centriq 2400, which offers 48 cores. AMD is another CPU partner for Olympus with its ARM-based processor, code-named Naples.  In addition, there are other ARM licensees waiting in the wings with designs aimed at data centres, including MACOM (AppliedMicro's X-Gene 3 processor) and Nephos, a spin-out from MediaTek. For Cavium and Qualcomm, the case for ARM-powered servers comes down to optimised performance for certain workloads, and in OCP Summit presentations, both companies cited web indexing and search as one of the first applications that Microsoft is using to test their processors.

Project Olympus is also putting forward an OCP design aimed at accelerating AI in its next-gen cloud infrastructure. Microsoft, together with NVIDIA and Ingrasys, is proposing a hyper-scale GPU accelerator chassis for AI. The design, code named HGX-1, will package eight of NVIDIA's latest Pascal GPUs connected via NVIDIA’s NVLink technology. The NVLink technology can scale to provide extremely high connectivity between as many as 32 GPUs - conceivably 4 HGX-1 boxes linked as one. A standardised AI chassis would enable Microsoft to rapidly rollout the same technology to all of its Azure data centres worldwide.

In tests published a few months ago, NVIDIA said its earlier DGX-1 server, which uses Pascal-powered Tesla P100 GPUs and an NVLink implementation, were delivering 170x of the performance of standard Xeon E5 CPUs when running Microsoft’s Cognitive Toolkit.

Meanwhile, Intel has introduced the second generation of its Rack Scale Design for OCP. This brings improvements in the management software for integrating OCP systems in a hyper-scale data centre and also adds open APIs to the Snap open source telemetry framework so that other partners can contribute to the management of each rack as an integrated system. This concept of easier data centre management was illustrated in an OCP keynote by Yahoo Japan, which amazingly delivers 62 billion page views per day to its users and remains the most popular website in that nation. The Yahoo Japan presentation focused on an OCP-compliant data centre it operates in the state of Washington, its only overseas data centre. The remote data centre facility is manned by only a skeleton crew that through streamlined OCP designs is able to perform most hardware maintenance tasks, such as replacing a disk drive, memory module or CPU, in less than two minutes.

One further note on Intel’s OCP efforts relates to its 100 Gbit/s CWDM4 silicon photonics modules, which it states are ramping up in shipment volume. These are lower cost 100 Gbit/s optical interfaces that run over up to 2 km for cross data centre connectivity.

On the OCP-compliant storage front not everything is flash, with spinning HDDs still in play. Seagate has recently announced a 12 Tbytes 3.5 HDD engineered to accommodate 550 Tbyte workloads annually. The company claims MTBF (mean time between failure) of 2.5 million hours and the drive is designed to operate 24/7 for five years. These 12 Tbyte enable a single 42 U rack to deploy over 10 Pbytes of storage, quite an amazing density considering how much bandwidth would be required to move this volume of data.


Google did not make a keynote appearance at this year’s OCP Summit, but had its own event underway in nearby San Francisco. The Google Cloud Next event gave the company an even bigger stage to present its vision for cloud services and the infrastructure needed to support it.

Wednesday, March 8, 2017

Microsoft's Project Olympus OCP Server Runs Qualcomm's ARM Processor

Qualcomm Datacenter Technologies (QCT) is working with Microsoft to enable a variety of Azure cloud workloads using its 10 nanometer Qualcomm Centriq 2400 ARM-based processor.

QCT has now joined the Open Compute Project and submitted a server specification using Centriq 2400, which offers up to 48 cores optimized for highly parallelized data center workloads.

Specifically, the Qualcomm Centriq 2400 Open Compute Motherboard server specification is based on the latest version of Microsoft’s Project Olympus. The companies have demonstrated Windows Server, developed for Microsoft’s internal use, powered by the Centriq 2400 processor.

“QDT is accelerating innovation in datacenters by delivering the world’s first 10nm server platform,” said Ram Peddibhotla, vice president, product management, Qualcomm Datacenter Technologies, Inc. “Our collaboration with Microsoft and contribution to the OCP community enables innovations such as Qualcomm Centriq 2400 to be designed in and deployed into the data centers rapidly. In collaborating with Microsoft and other industry leading partners, we are democratizing system design and enabling a broad-based ARM server ecosystem.”

“Microsoft and QDT are collaborating with an eye to the future addressing server acceleration and memory technologies that have the potential to shape the data center of tomorrow,” said Dr. Leendert van Doorn, distinguished engineer, Microsoft Azure, Microsoft Corp. “Our joint work on Windows Server for Microsoft’s internal use, and the Qualcomm Centriq 2400 Open Compute Motherboard server specification, compatible with Microsoft’s Project Olympus, is an important step toward enabling our cloud services to run on QDT-based server platforms.”

http://www.qualcomm.com

Cavium's ARM-based ThunderX2 Powers Microsoft's Project Olympus Server

At the Open Compute Summit in Santa Clara, California, Cavium announced that its ThunderX2 ARMv8-A Data Center processor is being tested by Microsoft for running a variety of workloads on the Microsoft Azure cloud platform.

The ThunderX2 product family is Cavium's second generation 64-bit ARMv8-A server processor SoCs for Data Center, Cloud and High Performance Computing applications. The family integrates fully out-of-order high performance custom cores supporting single and dual socket configurations. ThunderX2 is optimized to drive high computational performance delivering outstanding memory bandwidth and memory capacity.

Cavium said its hardware platform is fully compliant with Microsoft's Project Olympus which is one of the most modular and flexible cloud hardware designs in the data center industry. The platform integrates two ThunderX2 processors in a dual socket configuration. ThunderX2 SoC integrates a large number of fully out-of-order custom ARMv8-A cores with rich IO connectivity for accommodating a variety of peripherals for Azure, delivering excellent throughput and latency for cloud applications. The platform has been designed in collaboration with a leading server ODM supplier for Microsoft.

"Cavium is excited to work with Microsoft on ThunderX2," said Gopal Hegde, VP/GM, Data Center Processor Group at Cavium. "ARM-based servers have come a long way with first generation ThunderX-based server platforms being deployed at multiple data centers, which enabled a critical mass of ecosystem partners for ARM. We see the second generation products helping to drive a tipping point for ARM server deployment across a mainstream set of volume applications. Microsoft's support will help accelerate commercial deployment of ARMv8 server platforms for Data Centers and Cloud."

http://www.cavium.com

Wednesday, January 18, 2017

OpenSynergy Develops Automotive Safety Hypervisor for ARM Cortex-R52

OpenSynergy, a developer of automotive software based in Berlin, is working on a software hypervisor for the ARM Cortex-R52 processor. The hypervisor turns any chip based on the Cortex-R52 into several virtual machines capable of simultaneously executing separate software tasks.

ARM said this allows for the isolation of safety-critical functions from those that require less stringent control. In addition, it enables the consolidation of applications onto fewer electronic control units (ECUs) to both manage complexity and reduce cost.  Target applications include autonomous vehicles and industrial control systems.

"Mass-market autonomous vehicles will be engineered with greatly enhanced ECU compute capabilities and the ability to safely manage far more complex software stacks," said Richard York, vice president of embedded marketing, ARM. "The Cortex-R52 was purpose-built for this task, with hypervisor-enabled software separation protecting critical safety features while ensuring fast task execution. This will enable highly performant vehicles that can be fully trusted to take over from the driver."

"The ARM Cortex-R52 processor will bring virtualization technology to a much wider set of devices in the automotive market," said Stefaan Sonck Thiebaut, CEO, OpenSynergy. "In doing so, we look forward to enabling the next generation of vehicle architecture."

https://www.arm.com/about/newsroom/automotive-safety-hypervisor-announced-for-arm-cortex-r52.php

Thursday, December 8, 2016

China Unicom Works with Cavium on ARM-based Network Gear

Cavium announced an agreement with China Unicom to accelerate the design and development of Virtualized BBUs based on Cavium's ThunderX workload optimized data server processors, which are built on ARM architecture. In addition, Cavium has joined the China Unicom CORD Industry Alliance and will drive adoption of open source architecture and technologies in China together with China Unicom.

The companies said they plan to work together on new innovative fronthaul solutions, system architecture and vBBU performance and deployment. This collaboration allows Cavium to align with China Unicom's commercial networks technology development and innovation, research feasibility of Next Generation Virtualized Wireless Access Network, perform lab and field testing, evaluate results, drive deployment of developed technologies into commercial network, carry out lab and field performance test and assessment, accelerate pilot and application of new technical innovations in real-world networks.

"We are very pleased to collaborate with China Unicom in this critical area. As network capacity continues to be stretched and the user demands continue to grow the industry is faced with significant challenges which cannot be solved by traditional means," said Raj Singh General Manager of the Wireless Broadband Group at Cavium. "The use of advanced general purpose hardware such as Cavium's ThunderX workload optimized data severs allows us to provide a highly scalable virtualized solution for these requirements."

http://www.cavium.com

See also