Advertising

Showing posts with label 64-bit ARMv8. Show all posts
Showing posts with label 64-bit ARMv8. Show all posts

Tuesday, 20 September 2016

Unknown

ARM launches safety-centric Cortex-R52 processor

Cambridge-based and Japanese-owned chip designer ARM has announced the launch of its first ARMv8-R architecture IP, the ARM Cortex-R52, designed specifically to address functional safety in autonomous vehicle and other safety-critical applications.

 ARM has announced the launch of the Cortex-R52 processor 
ARM has announced the launch of the Cortex-R52 processor, its first to use the ARMv8-R architecture, with a focus on safety-critical tasks such as autonomous vehicles.
 

Based around an extension of the 64-bit ARMv8 microarchitecture, ARMv8-R introduces extensions for one simple purpose: to improve safety. The design introduces a hardware-enforced separation of software tasks, whereby a software hypervisor polices the execution and resource allocation to ensure that safety-critical code is run in a completely isolated environment away from any other code. The result, ARM claims, is a significant boost in security, a reduction in the amount of code that needs to go through safety certification, and the ability to run complex software without losing support for deterministic real-time operating system (RTOS) operation.

'The Cortex-R52 is the first processor built on the ARMv8-R architecture and it was designed from the ground up to address functional safety,' explained James McNiven, ARM's general manager for CPU and media processing groups, of the launch. 'We are helping partners to meet particular market opportunities, especially in fully autonomous vehicles and robotics systems where specific functionality is required for safety-critical tasks. By documenting the strict development process, fault modelling and supporting software isolation, ARM is enabling a faster route to market for partners addressing these applications.'

The first licensee for the Cortex-R52 design is STMicroelectronics, which is to build the part for its customers as a replacement for the older safety-centric Cortex-R5. Those looking to upgrade will, the company has claimed, see a 35 percent increase in performance as a result. 'The Cortex-R52 supports our Smart Driving vision by enabling a new range of high-performance, power-efficient SoCs for any in-vehicle application demanding real-time operation and the highest levels of functional safety, including powertrain, chassis and ADAS,' claimed Fabio Marchiò at STMicroelectronics. 'The Cortex-R52's ability to compartmentalise software provides our users with the best solution for safety without loss of determinism. Its virtualisation support simplifies the consolidation of applications and functions into a single processor, delivering a shorter integration time.'

Thus far, no company has stepped forward to suggest when the first devices based around the Cortex-R52 design will hit the open market.

ARM has announced the launch of the Cortex-R52 processor, its first to use the ARMv8-R architecture, with a focus on safety-critical tasks such as autonomous vehicles.
Read More

Friday, 22 July 2016

Unknown

New Cavium ThunderX2 adopts 64-bit ARM-based servers to address application and workload requirements

Semiconductor vendor Cavium announced Monday ThunderX2, its second generation of workload optimized ARM server SoCs that targets high performance volume servers deployed by public/private cloud and telecom communications data centers and high performance computing applications. It is optimized for data center workloads such as compute, security, storage, data analytics, network function virtualization and distributed databases.

The ThunderX2 line of processors currently includes four workload optimized processors targeting different workloads.

The ThunderX2_CP has been optimized for cloud compute workloads such as private and public clouds, web serving, web caching, web search, commercial HPC workloads such as computational fluid dynamics (CFD) and reservoir modeling. This line supports multiple 10/25/40/50/100 GbE network Interfaces and PCIe Gen3 interfaces. It also includes accelerators for virtualization and vSwitch offload.

The ThunderX2_ST has been optimized for big data, cloud storage, massively parallel processing (MPP) databases and Data warehousing workloads. This family supports multiple 10/25/40/50/100 GbE network interfaces, PCIe Gen3 interfaces and SATAv3 interfaces. It also includes hardware accelerators for data protection/ integrity/security, user to user efficient data movement.

The ThunderX2_SC has been optimized for secure web front-end, security appliances and cloud RAN type workloads. This family supports multiple 10/25/40/50/100 GbE interfaces and PCIe Gen3 interfaces. Integrated hardware accelerators include Cavium’s industry leading, 5th generation NITROX security technology with acceleration for IPSec, RSA and SSL.

The ThunderX2_NT has been optimized for media servers, scale-out embedded applications and NFV type workloads. This family supports multiple 10/25/40/50/100 GbE interfaces. It also includes OCTEON style hardware accelerators for packet parsing, shaping, lookup, QoS and forwarding.

“The Cavium ThunderX2 will expand the market opportunity for ARM-based server technologies by addressing demanding application and workload requirements for compute, storage networking and security,” said Simon Segars, CEO, ARM. “ThunderX2 demonstrates Cavium’s ability to deliver a combination of innovation and engineering execution and the new product family increases the momentum for server deployments powered by ARM processors in large scale data centers and end user environments.”

Cavium’s ThunderX2 SoC line is supported by a comprehensive software ecosystem ranging from platform level systems management and firmware to commercial operating systems, development environments and applications.

Cavium has actively engaged in server industry standards groups such as UEFI and delivered numerous reference platforms to an array of community and corporate partners. Cavium has also demonstrated its position in the open source software community driving upstream kernel enablement for ThunderX, actively contributing to Linaro’s enterprise and networking groups, investing in Linux Foundation projects such as Xen and OPNFV and sponsoring the FreeBSD Foundation’s ARMv8 server implementation.

ThunderX2 will deliver two to three times the performance across a range of standard benchmarks and applications compared to ThunderX, while boosting the market reach of the ThunderX line of processors by targeting applications that require high single thread performance such as web search, graph analytics, a variety of enterprise applications such as massively parallel processing (MPP) databases, data warehousing and enterprise HPC applications such as computational fluid dynamics (CFD) and reservoir modelling. ThunderX2 will deliver comparable performance at a better total cost of ownership compared to the next generation of traditional server processors.



Read More

Thursday, 21 July 2016

Unknown

Cavium ThunderX block diagrams for its four families

During our work with the Cavium ThunderX platform, we have had access to a single 48-core ARMv8 2.0GHz SKU. Cavium’s strategy is to target different processor models into different market. Aside from clock speed and core count, Cavium has feature-based differentiation for its different ThunderX product families. We received permission to publish information on the different SKUs and their unique features.

The four Cavium ThunderX families (ThunderX_CP, ThunderX_ST, ThunderX_SC, ThunderX_NT) have a similar base architecture. From this common architecture, each chip is tailored to its targeted application. There are a few common underpinnings but this is the basic block diagram that shows all potential components of the platform.
The ThunderX_CP family is targeted at compute workloads. These chips target public and private clouds, web caching and web serving, search, social media and similar applications. The main accelerator in this family is the vSwitch Offload Engine. If you think about the application workload this is targeted at, one will take advantage of the high speed networking, multitude of cores and RAM bandwidth.
With Cavium’s strong networking portfolio as well as 16x SATA III 6.0gbps ports (60% more than a comparable Intel single socket system), storage is an application where the ThunderX architecture is well suited. The ThunderX_ST family targets block, object and distributed file storage workloads, distributed database applications and Hadoop style workloads. It comes in 8-48 core variants and has additional hardware accelerators such as the compression engine available.

Building on the high core count and the high-end networking throughput, the Cavium ThunderX_SC is focused on applications like eCommerce web servers. Cavium includes hardware accelerators for SSL, IPSec, deep packet inspection, anti-virus and anti-malware to free CPU cycles. Cavium has IP that it uses in dedicated Nitrox III chips that it leverages here.
With Cavium’s background as a networking company, and given the specs of the ThunderX chips, one area we think the company will do well in is in the networking and SP cloud space. The ThunderX_NT targets network function virtualization (NFV) servers and the telecom clouds being deployed with a mix of high bandwidth along with hardware accelerators. Like the storage family, the networking focused family can scale from 8-48 cores.

Final Words

We have been working with a single socket Cavium ThunderX machine, the Gigabyte R120-T30 server thus far and the results have been impressive. Cavium is embracing a different approach to product differentiation than Intel. While Intel is generally focused on raw x86 performance, Cavium’s SoC design allows it to differentiate on features. This can include networking, storage accelerators and other vectors. The Cavium ThunderX family is the first ARMv8 chip we have seen, in production, that can legitimately match and outpace parts of the Intel Xeon E5 line in terms of performance.




Read More

Thursday, 6 November 2014

Unknown

Cavium Rolls Out ThunderX2 ARM Processor


Today Cavium announced ThunderX2, its second generation of Workload-Optimized ARM server SoCs. ThunderX2 targets high performance volume servers deployed by Public/Private Cloud and Telco data centers and high performance computing applications. It is optimized for key Data Center workloads such as compute, security, storage, data analytics, network function virtualization and distributed databases.


ARM-based server technologies
The Cavium ThunderX2 will expand the market opportunity for ARM-based server technologies by addressing demanding application and workload requirements for compute, storage networking and security,” said Simon Segars, CEO, ARM. “ThunderX2 demonstrates Cavium’s ability to deliver a combination of innovation and engineering execution and the new product family increases the momentum for server deployments powered by ARM processors in large scale data centers and end user environments.”
ThunderX2 will deliver two to three times the performance across a wide range of standard benchmarks and applications compared to ThunderX. It also significantly expands the market reach of the ThunderX family of processors by targeting applications that require high single thread performance such as web search, graph analytics, a variety of enterprise applications such as massively parallel processing (MPP) databases, data warehousing and enterprise HPC applications such as computational fluid dynamics (CFD) and reservoir modeling.
ThunderX2 will deliver comparable performance at a better total cost of ownership compared to the next generation of traditional server processors. ThunderX2 is the second generation Workload Optimized processor from Cavium. It is based on the latest ARMv8-A architecture and built on a 14nm FinFET process node. It is fully compliant with ARM’s Server Base System Architecture (SBSA) standard.

Key ThunderX2 features will include:
§  2nd generation of full custom Cavium ARM core o Multi-Issue, Fully OOO. o 2.4 to 2.8 GHz in normal mode, Up to 3 GHz in Turbo mode. o > 2X single thread performance compared to ThunderX.
§  Up to 54 cores per socket delivering 2-3X socket level performance compared to ThunderX
§  Cache o 64K I-Cache and 40K D-Cache, highly associative. o 32MB shared Last Level Cache (LLC).
§  Single and dual socket configuration support using 2nd generation of Cavium Coherent Interconnect with > 2.5X coherent bandwidth compared to ThunderX
§  System Memory o 6 DDR4 memory controllers per socket. o Dual DIMM per memory controller, for a total of 12 DIMMs per socket. o Up to 3200MHz in 1 DPC and 2966MHz in 2 DPC configuration. o Up to 3 TB of memory in dual socket configuration.
§  Full system virtualization for low latency from virtual machine to IO enabled through Cavium virtSOC technology
§  Next Generation IO o Integrated 10/25/40/50/100GbE network connectivity: Multiple integrated SATAv3 interfaces, Integrated PCIe Gen3 interfaces, x1, x4, x8 and x16 support.
§  Integrated Hardware Accelerators: OCTEON style packet parsing, shaping, lookup, QoS and forwarding, Virtual Switch (vSwitch) offload, Virtualization, storage and NITROX V security.

The ThunderX2 family of processors includes FOUR workload optimized processors targeting different workloads:
§  ThunderX2_CP: Optimized for cloud compute workloads such as private and public clouds, web serving, web caching, web search, commercial HPC workloads such as computational fluid dynamics (CFD) and reservoir modeling. This family supports multiple 10/25/40/50/100 GbE network Interfaces and PCIe Gen3 interfaces. It also includes accelerators for virtualization and vSwitch offload.
§  ThunderX2_ST: Optimized for big data, cloud storage, massively parallel processing (MPP) databases and Data warehousing workloads. This family supports multiple 10/25/40/50/100 GbE network interfaces, PCIe Gen3 interfaces and SATAv3 interfaces. It also includes hardware accelerators for data protection/ integrity/security, user to user efficient data movement.
§  ThunderX2_SC: Optimized for secure web front-end, security appliances and cloud RAN type workloads. This family supports multiple 10/25/40/50/100 GbE interfaces and PCIe Gen3 interfaces. Integrated hardware accelerators include Cavium’s industry leading, 5th generation NITROX security technology with acceleration for IPSec, RSA and SSL.
§  ThunderX2_NT: Optimized for media servers, scale-out embedded applications and NFV type workloads. This family supports multiple 10/25/40/50/100 GbE interfaces. It also includes OCTEON style hardware accelerators for packet parsing, shaping, lookup, QoS and forwarding. 

ThunderX2 combines our next generation core that will deliver significantly higher single thread performance with next generation IO and hardware accelerators to provide a compelling value proposition for the server market and greatly expand the serviceable server TAM,” said Syed Ali, President and CEO of Cavium. “ThunderX2 will enable flexible, scalable and fully optimizable servers for next generation software defined data centers.”

Cavium’s ThunderX2 SoC family is supported by a comprehensive software ecosystem ranging from platform level systems management and firmware to commercial Operating Systems, Development Environments and Applications. Cavium has actively engaged in server industry standards groups such as UEFI and delivered numerous reference platforms to a broad array of community and corporate partners. 

It is a honor to be able to rely on such a close business relationship with a partner like Cavium, we designed and delivered our own solutions based on Cavium ThunderX SoC for a while now, with extreme satisfaction in terms of performances, support and overall results. We can’t wait to get our hands the next gen SoC as we are confident it will provide an additional boost to an already strong and tested product line.” Cosimo Gianfreda, CTO E4 Computer Engineering. (Know more)
Read More