Compute Express Link™: The Breakthrough CPU-to-Device Interconnect
A look into the CXL device ecosystem and the evolution of CXL use cases
Airing Date: January 26, 2023
Compute Express Link™ (CXL™) is an industry supported cache-coherent interconnect for processors, memory expansion, and accelerators. CXL enables a high-speed, efficient interconnect between the CPU and platform enhancements and workload accelerators, such as GPUs, FPGAs, and other purpose-built accelerator solutions. At Supercomputing 2022 (SC’22), CXL Consortium members showcased CXL technology demonstrations on memory solutions, IP, compliance testing, fabric implementations, switches, and software solutions.
The webinar will provide an update from the CXL Consortium, highlight the industry ecosystem and technology demos shared at SC’22, and discuss evolving CXL use cases. A Q&A session featuring member company representatives who showcased a CXL demo at SC’22 will close out the webinar.
Presenters: Kurtis Bowman and Kurt Lender, CXL Consortium MWG Co-Chairs
CXL 1.1 vs. CXL 2.0 – What’s the difference?
Airing Date: June 15, 2022
This webinar will share a high-level overview of CXL 1.1, and the enhancements made in CXL 2.0 focusing on switching, memory pooling, Single Logical Devices (SLD) vs. Multiple Logical Devices (MLD), and fabric management. The presentation will also explore managed hot-plug, memory QoS telemetry, speculative reads, and security enhancements.
Presenters: Danny Volkind and Elad Shliselberg, UniFabriX
An Overview of the Compute Express Link™ (CXL™) 2.0 ECNs
Airing Date: December 9, 2021
In November 2020, the CXL Consortium released the CXL 2.0 specification which introduces support for switching, memory pooling, and support for persistent memory – all while preserving industry investments by supporting full backward compatibility. Based on member feedback, CXL 2.0 ECNs made significant improvements to the specifications in the areas of device management, RAS, Security, memory interleaving and others.
This webinar will review the key CXL 2.0 specification ECNs and the new usages they enable.
Ishwar Agarwal, Microsoft and Mahesh Natu, Intel
Compute Express Link™ (CXL™): Supporting Persistent Memory
Airing Date: June 15, 2021
Compute Express Link™ (CXL™) is an open industry-standard interconnect offering coherency and memory semantics using high-bandwidth, low-latency connectivity between the host processor and devices such as accelerators, memory buffers, and smart I/O devices. The CXL 2.0 specification introduces support for switching, memory pooling, and persistent memory – all while preserving industry investments by supporting full backward compatibility.
An increasing number of applications — ranging from databases to AI workloads — are being enhanced to take advantage of persistent memory. This webinar will explore how the CXL specification has evolved to support persistent memory devices in a way that preserves the established software model. This webinar will cover enhancements to the CXL protocol, error handling, and standardized configuration interface, enabling innovative designs that are based on a variety of non-volatile media and form factors.
Introducing the Compute Express Link™ 2.0 Specification
Airing Date: December 10, 2020
The CXL™ Consortium is proud to announce the second-generation Compute Express Link™ (CXL™). CXL is an open industry-standard interconnect offering coherency and memory semantics using high-bandwidth, low-latency connectivity between host processor and devices such as accelerators, memory buffers, and smart I/O devices.
This webinar will explore the new features in the CXL 2.0 specification including support for switching for fan-out to connect to more devices; memory pooling for increased memory utilization efficiency and providing memory capacity on demand; and support for persistent memory – all while preserving industry investments by supporting full backward compatibility with CXL 1.1 and 1.0.
Exploring Coherent Memory and Innovative Use Cases
Airing Date: March 12, 2020
Siamak Tavallaei, CXL Consortium Technical Task Force Co-Chair and Principal Architect, Microsoft Azure, Rob Blankenship, Processor Architect and Principal Engineer, Intel, and Kurt Lender, CXL Consortium Marketing Working Group Co-Chair and Senior Ecosystem Enabling Manager, Data Center Group, Intel, take listeners into a deep dive about CXL technology maintains memory coherency between the CPU memory space and memory on attached devices.
The webinar also details several representative CXL use cases – Caching Devices/Accelerators, Accelerators with Memory, and Memory Buffers.
Introducing CXL 3.0: Enabling composable systems with expanded fabric capabilities
Airing Date: October 6, 2022
CXL 3.0 expands on previous technology generations to increase scalability and optimize system level flows with advanced switching and fabric capabilities, efficient peer-to-peer communications, and fine-grained resource sharing across multiple compute domains. It also doubles the data rate to 64 GT/s with no added latency over CXL 2.0, while maintaining backward compatibility with previous specifications.
This webinar will introduce the CXL 3.0 specification and explore the new features including:
• Fabric capabilities and management
• Dynamically composable systems
• Improved memory sharing and pooling
• Enhanced coherency
• Peer-to-peer communication
• Doubled data rate with no added latency
Presenters: Dr. Debendra Das Sharma, Intel and Danny Moore, Rambus
Introduction to the Compute Express Link™ (CXL™) Fabric Manager
Airing Date: March 22, 2022
This webinar will introduce the CXL™ Fabric Manager (FM), explore Multi-Logical Device (MLD) management, and detail the Component Command Interface (CCI) transport protocols, background operations and categories of requests. The presentation will also share an architectural overview of the CXL 2.0 management framework and how it addresses the requirements of enterprise and datacenter deployments.
Presenter: Vincent Haché, Rambus
Compute Express Link™ (CXL™): Link-level Integrity and Data Encryption (CXL IDE)
Airing Date: September 30, 2021
Security is a key cornerstone for any technology to be successful and CXL is making great strides in security by working collaboratively with other industry-standard bodies such as PCI-SIG and DMTF to ensure a seamless user experience with the best security mechanisms. CXL 2.0 enhances the security mechanism from CXL 1.1 and 1.0 by adding link-level Integrity and Data Encryption (CXL IDE) to provide confidentiality, integrity and replay protection for data transiting the CXL link.
This webinar will explore CXL IDE usage models and how security is managed across CXL.io, CXL.mem, CXL.cache and CXL Switches. The webinar will also explore a Device’s responsibility to maintain security.
Compute Express Link™ 2.0 Specification: Memory Pooling
Airing Date: March 23, 2021
Compute Express Link™ (CXL™) is an open industry-standard interconnect offering coherency and memory semantics using high-bandwidth, low-latency connectivity between the host processor and devices such as accelerators, memory buffers, and smart I/O devices. In November 2020, the CXL Consortium announced the CXL 2.0 specification which introduces support for switching, memory pooling, and support for persistent memory – all while preserving industry investments by supporting full backward compatibility.
In this webinar, Mahesh Wagh (Intel) and Rick Sodke (Microchip), will explore how CXL 2.0 supports memory pooling for multiple logical devices (MLD) as well as a single logical device with the help of a CXL switch. This presentation will also introduce the standardized fabric manager for inventory and resource allocation to enable easier adoption and management of CXL-based switch and fabric
Memory Challenges and CXL Solutions
Airing Date: August 6, 2020
Compute Express Link™ (CXL™) is an industry supported cache-coherent interconnect for processors, memory expansion and accelerators. CXL supports dynamic multiplexing between a rich set of protocols that includes I/O (CXL.io, based on PCIe®), caching (CXL.cache) and memory (CXL.memory) semantics. CXL.mem allows a host processor to access memory attached to a CXL device. CXL.mem transactions are simple memory load and store transactions that run downstream from the host processor which takes care of all the associated coherency flows.
This webinar will explore how the CXL.mem protocol can deliver power-efficient performance with emerging applications such as AI, HPC, DL and comms with a coherent interface and low latency. This webinar will also explore type-3 devices and use cases for memory bandwidth expansion, memory capacity expansion and storage class memory.
Introduction to Compute Express Link™ (CXL)
Airing Date: December 12, 2019
A highly informative webinar about the CXL Consortium™ and its groundbreaking technology. Join Glenn Ward, CXL Consortium’s MWG Co-Chair and Chief of Staff, Cloud Server Infrastructure for Microsoft; Debendra Das Sharma, Intel Fellow at Intel Corporation; and Kurtis Bowman, CXL Consortium Board Member and Director, Server Architecture and Technologies at Dell EMC during this thought-provoking webinar about the organization and its innovative technology.
Topics covered include the industry landscape, features and benefits, use cases and more!