Arm Community
Arm Community
  • Site
  • User
  • Site
  • Search
  • User
  • Groups
    • Research Collaboration and Enablement
    • DesignStart
    • Education Hub
    • Innovation
    • Open Source Software and Platforms
  • Forums
    • AI and ML forum
    • Architectures and Processors forum
    • Arm Development Platforms forum
    • Arm Development Studio forum
    • Arm Virtual Hardware forum
    • Automotive forum
    • Compilers and Libraries forum
    • Graphics, Gaming, and VR forum
    • High Performance Computing (HPC) forum
    • Infrastructure Solutions forum
    • Internet of Things (IoT) forum
    • Keil forum
    • Morello Forum
    • Operating Systems forum
    • SoC Design and Simulation forum
    • 中文社区论区
  • Blogs
    • AI and ML blog
    • Announcements
    • Architectures and Processors blog
    • Automotive blog
    • Graphics, Gaming, and VR blog
    • High Performance Computing (HPC) blog
    • Infrastructure Solutions blog
    • Innovation blog
    • Internet of Things (IoT) blog
    • Operating Systems blog
    • Research Articles
    • SoC Design and Simulation blog
    • Tools, Software and IDEs blog
    • 中文社区博客
  • Support
    • Arm Support Services
    • Documentation
    • Downloads
    • Training
    • Arm Approved program
    • Arm Design Reviews
  • Community Help
  • More
  • Cancel
Arm Community blogs
Arm Community blogs
Architectures and Processors blog Arm joins CXL - establishing key industry standard to streamline heterogeneous compute
  • Blogs
  • Mentions
  • Sub-Groups
  • Tags
  • Jump...
  • Cancel
More blogs in Arm Community blogs
  • AI and ML blog

  • Announcements

  • Architectures and Processors blog

  • Automotive blog

  • Embedded blog

  • Graphics, Gaming, and VR blog

  • High Performance Computing (HPC) blog

  • Infrastructure Solutions blog

  • Internet of Things (IoT) blog

  • Operating Systems blog

  • SoC Design and Simulation blog

  • Tools, Software and IDEs blog

Tell us what you think
Tags
  • Hardware Accelerators
  • Heterogeneous Computing
  • Processor Architecture
  • interconnect
Actions
  • RSS
  • More
  • Cancel
Related blog posts
Related forum threads

Arm joins CXL - establishing key industry standard to streamline heterogeneous compute

Richard Grisenthwaite
Richard Grisenthwaite
September 11, 2019
2 minute read time.

Compute Express Link (CXL) is a new high-speed CPU interconnect interface to accelerators and memory expansion. Arm joins forces in CXL with industry leaders Alibaba, Facebook, Google and Microsoft and hardware providers Cisco, Dell EMC, Hewlett Packard Enterprise, Huawei and Intel Corporation, aiming to develop the necessary specifications.

As the industry moves toward heterogenous compute for key workloads, such as networking, storage, graphics, and machine learning, CXL reinforces the need for standardization and is aiming for a high-speed, low latency interconnect between the CPU and a growing variety of workload accelerators. Standards are important to enable innovative system designs by ensuring interoperability with different optimized system components.  Arm joined the CXL Consortium both to ensure low friction CXL development to the Arm ecosystem and to contribute to the future definition with CXL member companies.

Arm has been actively participating in the CXL work groups and has dedicated key resources to promote CXL, including several Arm Fellows and Principal Engineers.

Fitting with Arm’s standards strategy

Arm brings to the table its track record in establishing a full software framework and aims to promote the migration to CXL as the interface to the accelerators and memory expansion. We expect to maintain CCIX to support inter-package chip-to-chip interface that is currently not in the scope of CXL. We will continue to support customer solutions based on existing CCIX hardware.

Arm has a proven track record developing standards such as AMBA, an open standard for on-chip communication. AMBA architecture provides the on-chip interconnect definitions for highly scalable SoCs in the datacenter, from compute servers to networking, storage, security and other accelerators. This includes an associated ecosystem with development tools from all leading EDA vendors. The most prevalent on-chip interface for 3rdparty PCIe IP is AMBA. Arm collaborates in industry multi-chip interconnect standards to optimize performance between on-chip and off-chip domains. Arm is also a board member in the PCI SIG and the Gen-Z Consortium. Gen-Z is the fabric that allows for rack and row connectivity for CPUs, accelerators, and memory.

Arm also invests in software architecture focused on portability and simplified deployment by working within the UEFI Forum. This forum provides the necessary interfaces between the firmware and the operating systems to support the accelerator and memory expansion use cases across different implementations. Arm is to lead the UEFI sub-team in the System and Software Work Group, after contributing the platform interfaces developed for CCIX that equally map to CXL. Arm also sees the opportunity for harmonizing open source development in addition to firmware standards, for example, a common accelerator framework and an accelerator library that can foster adoption of both CXL and CCIX in the short term, just like PMDK does for persistent memory.

Advancing the ecosystem

We have a well-established ecosystem and have its best interests at heart. We hope to be joined by many of our partners within the consortium, to represent the views of the Arm ecosystem and develop this key standard for the benefit of heterogeneous compute. Arm is a strategic choice of IP provider for the ecosystem. Our interconnect IP (CMN) is committed to enable CXL for both processor and end-point devices based on our customer input.

Anonymous
Architectures and Processors blog
  • Using Perf to enable PMU functionality on Armv8-A CPUs: Stage 3 and Stage 4

    Jiaming Guo
    Jiaming Guo
    Part 3. This blog introduces how to enable PMU feature in Armv8-A CPUs and provides an example of performance analysis workflow on real Armv8-A platform.
    • August 22, 2023
  • Using Perf to enable PMU functionality on Armv8-A CPUs: Stage 1 and Stage 2

    Jiaming Guo
    Jiaming Guo
    Part 2. This blog introduces how to enable PMU feature in Armv8-A CPUs and provides an example of performance analysis workflow on real Armv8-A platform.
    • August 15, 2023
  • Using Perf to enable PMU functionality on Armv8-A CPUs: Enable Arm PMU support for the kernel and install Linux Perf tool

    Jiaming Guo
    Jiaming Guo
    This blog post introduces how to enable PMU feature in Armv8-A CPUs and provides an example of performance analysis workflow on real Armv8-A platform.
    • August 8, 2023