Beaverton, OR, USA – September 16, 2019 –Today, Alibaba, Cisco, Dell EMC, Facebook, Google, Hewlett Packard Enterprise, Huawei, Intel Corporation and Microsoft announce the incorporation of the Compute Express Link (CXL) Consortium, and unveiled the names of its newly-elected members to its Board of Directors.
The core group of key industry partners announced their intent to incorporate in March 2019, and remain dedicated to advancing the CXL standard, a new high-speed CPU-to-Device and CPU-to-Memory interconnect which accelerates next-generation data center performance.
The five new CXL board members are as follows: Steve Fields, Fellow and Chief Engineer of Power Systems, IBM; Gaurav Singh, Corporate Vice President, Xilinx; Dong Wei, Standards Architect and Fellow at ARM Holdings; Nathan Kalyanasundharam, Senior Fellow at AMD Semiconductor; and Larrie Carr, Fellow, Technical Strategy and Architecture, Data Center Solutions, Microchip Technology Inc.
Chairman of the CXL Consortium, Jim Pappas, discussed the recent developments:
“CXL is enjoying unprecedented momentum with participation of leading companies from across all segments of the computer industry,” said Pappas. “It is extremely gratifying to see the level of investment and focus on the success of CXL. We have a world-class technical team covering all of the needed disciplines to ensure success. On behalf of the CXL Board, I am delighted to welcome these new directors onto the CXL Board of Directors and look forward to their contributions to this important new technology.”
CXL Specification 1.1
As accelerators are increasingly used to complement CPUs in support of emerging applications such as Artificial Intelligence and Machine Learning, CXL was designed to be an industry open standard interface for high-speed communications. The specification delivers breakthrough performance for emerging usage models while supporting an open ecosystem for data center accelerators and other high-speed enhancements—it was ratified in March 2019.
CXL technology maintains memory coherency between the CPU memory space and memory on attached devices, which allows resource sharing for higher performance, reduced software stack complexity, and lower overall system cost. This permits users to simply focus on target workloads as opposed to the redundant memory management hardware in their accelerators.
The specification is currently available to companies that join the CXL Consortium.
CXL Newly-Elected Board Members’ Statements upon Election
The newly-elected Board Members to CXL made the following comments upon their confirmation to the BoD:
"CXL has achieved enough participation that it is now time for IBM to bring the experience gained from 6+ years of open coherent acceleration with CAPI and OpenCAPI and to help ensure that CXL is an open architecture-agnostic standard." --Steve Fields, Fellow and Chief Engineer of Power Systems, IBM
“High performance, adaptable acceleration is going to be critical to meet the increasing demand for performance across a wide variety of markets,” said Gaurav Singh, corporate vice president of Architecture, Verification and Systems Engineering, Xilinx, Inc. “At Xilinx, we recognized the importance of coherent links to connect processors, memory and accelerators and I am excited to join the CXL board to drive the standard forward. With the backing of hyperscale cloud providers, system OEMs, processors vendors and technology leaders like Xilinx, CXL will become a ubiquitous processor interface for memory and accelerators.” --Gaurav Singh, Corporate Vice President, Xilinx
“I am delighted to be appointed as a new board member of the CXL Consortium. Arm’s membership at the board level not only reflects the industry’s confidence in our strong contribution to this important, emerging technology, but a shared vision for aligning to a truly global multi-CPU standard applicable to all processor architectures.” --Dong Wei, Standards Architect and Fellow at Arm Holdings, Inc.
“AMD is a strong believer in the need for heterogeneous processing to meet the exponential demand for compute, and we are pleased to join with other industry leaders to drive alignment on a common coherent interconnect standard in CXL,” said Nathan Kalyanasundharam, Senior Fellow at AMD. “We look forward to collaborating with all the members of the CXL community to enable a tighter coupling and coherency between processors and accelerators, simplify programming models, and enable a new wave of innovation in accelerator and storage technologies.” --Nathan Kalyanasundharam, Senior Fellow at AMD
“I am honored to be elected to the CXL Board of Directors. CXL will be a key technology to enable breakthrough memory-centric infrastructure solutions with the goal of improving datacenter performance and efficiency. Microchip looks forward to contributing to the evolution of the CXL standard, and the existing collaboration with the Gen-Z organization. --Larrie Carr, Fellow, Technical Strategy and Architecture, Data Center Solutions, Microchip Technology Inc.
Join CXL Industry Leaders
The founding promoter companies represent a wide range of industry expertise and include leading Cloud Service Providers, Communications OEMs, and System OEMs. CXL welcomes new members. Work is now underway to expand innovative use cases that leverage CXL technology as well as the development of CXL Specification 2.0 to further the standard’s technical features and capabilities.
About Compute Express Link
Compute Express Link (CXL) Consortium was founded in 2019 and is dedicated to advancing CXL technology, a high-speed CPU interconnect technology that accelerates next generation data center performance. CXL Consortium is led by key industry leaders Alibaba, Cisco, Dell EMC, Facebook, Google, Hewlett Packard Enterprise, Huawei, Intel Corporation and Microsoft. These companies originally formed an open industry standard group to develop technical specifications that facilitate breakthrough performance for emerging usage models while supporting an open ecosystem for data center accelerators and other high-speed enhancements. For more information on the CXL Consortium, please visit www.computeexpresslink.org.