AsteraLabs-Blue-Grey-Hz
  • Applications
  • Products
    • Product Overview

      Built in the cloud, for the cloud.

      • Hardware Solutions
      • Design Services
      • Buy and Sample
    • Aries PCIe®/CXL™ Smart Retimers

      Industry-proven Smart Retimers for PCI Express® (PCIe) 4.0, PCIe 5.0, and Compute Express Link™ (CXL) systems

    • Taurus Ethernet Smart Cable Modules™

      Overcome reach, signal integrity, and bandwidth utilization issues for 100G/Lane Ethernet connectivity

    • Leo CXL Memory Connectivity Platform Pre-production

      CXL-attached memory expansion, pooling, and sharing for cloud servers

  • Cloud-Scale Interop Lab
    • Aries PCIe/CXL Smart Retimers

      Learn how Astera Labs enables you to deploy PCIe and CXL systems with confidence.

    • Leo CXL Memory Connectivity Platform

      Learn how Astera Labs enables you to deploy CXL-attached memory with confidence.

  • Technology Insights
    • Articles

      Browse our knowledge base articles for information about our products and technologies.

    • Video Center

      Explore our educational and technical video center to support your design needs.

    • Document Library

      Find app notes, white papers and more in our technical resource library.

    • FAQs

      Find answers to the most frequently asked questions about our products and technologies.

    • Webinars

      Created by engineers for engineers, our webinar series explores the most important topics related to hyperscale datacenters.

  • Careers
  • About
    • About Us
    • Team
    • Support Portal    
    • Quality
    • News & Articles
    • Events
  • Contact

Home » Products » CXL Memory Platform » Leo CXL Memory Connectivity Platform

Leo CXL Memory Connectivity Platform

Home » Products » CXL Memory Platform » Leo CXL Memory Connectivity Platform

Leo CXL™ Memory Connectivity Platform

 

Astera Labs delivers Leo CXL™ Memory Connectivity Platform to eliminate memory bandwidth bottlenecks and capacity constraints for compute-intensive workloads such as AI and ML.

Leo CXL Memory Connectivity Platform is the industry’s first purpose-built solution to support memory expansion, memory pooling and memory sharing using CXL 1.1 and 2.0. Leo includes server-grade customizable Reliability, Availability and Serviceability (RAS), end-to-end security, extensive fleet management capabilities and seamless interoperability with all major CPU, GPU and memory vendors for cloud-scale deployment.

  • Purpose-built for cloud scale deployment targeting workloads such as AI and Machine Learning
  • Memory expansion, pooling and sharing for heterogeneous CPU/GPU topologies
  • CXL Type-3 device supporting CXL 1.1 and 2.0 memory expansion
  • CXL interface up to 32 GT/s per lane, up to 16 lanes
  • Multiple DDRx channels to increase memory capacity up to 2TB
  • Up to 5600 MT/s per memory channel to fully utilize available bandwidth of CXL 1.1 and 2.0 interface
  • Server-grade customizable RAS and software APIs to integrate with fleet management services
  • Seamless interoperation with all major CPU, GPU, and memory vendors
  • Flexible and scalable memory interface with a low-latency data path
  • Best-in-class security features to ensure end-to-end data integrity and protection

Leo CXL Memory Connectivity Platform includes a comprehensive portfolio of controllers and hardware solutions:

  • Leo E-Series CXL Smart Memory Controller supporting memory expansion
  • Leo P-Series CXL Smart Memory Controller supporting memory expansion, pooling, and sharing
  • Aurora A-Series Smart Memory Hardware Solutions in add-in-card form factor supporting memory expansion, pooling and sharing
Download Product Brief
Order Now
More About the Leo Platform
    System Block Diagram
    LEO-Diagram-NoPCIe

    Use Cases

    Memory Expansion
    Memory Expansion

    Reduce TCO and increase CPU efficiency with memory bandwidth and capacity expansion

    Memory Pooling
    Memory Pooling

    Reduce TCO by optimizing memory resource utilization

    Memory Sharing
    Memory Sharing

    Optimize data flow efficiency with low latency

    E-Series vs. P-Series

    Astera Labs offers two versions of the Leo CXL Memory Connectivity Platform to unlock the benefits of the CXL.mem protocol: the E-Series, which offers memory expansion, and the P-Series, which adds memory sharing and pooling capabilities. Both deliver cloud service providers tools and diagnostics essential for advanced fleet management.

    Leo E-Series

    E-Series

    • Memory Expansion

    Leo P Series Chip

    P-Series

    • Memory Expansion

    • Memory Pooling

    • Memory Sharing

    leo-aurora-board-front-675x550

    Aurora A-Series Hardware Solutions

    Aurora A-Series CXL Smart Memory Hardware Solutions offer all the benefits of Leo Controllers and enable quick plug-and-play deployment with faster time-to-market for system OEMs and data centers.

    • PCIe x16 CEM add-in card form factor
    • Lower TCO via more granular serviceability - can replace single failing DIMM
    • Configurable memory capacity due to plug-and-play capability of DIMMs
    • Up to 4x DDR5 RDIMMs supporting up to 2TB
    • On-board debug connectors for fleet management on Cloud Servers
    • Temperature and health monitoring of Leo controllers and memory
    • RDIMM fault isolation and error correction
    • High volume production-qualified solutions with robust supply chain

    Why Use Leo CXL Memory Connectivity Platform?

    Astera Labs Leo CXL Memory Connectivity Platform is the industry's first purpose-built solution that supports both memory expansion and memory pooling to solve performance bottlenecks and capacity constraints in cloud servers.

    leo-comprehensive
    Purpose-Built for Cloud

    Comprehensive portfolio of purpose-built SoCs and hardware solutions for cloud-scale deployment targeting workloads such as AI and ML

    leo-secure
    Customizable RAS & Security

    Server-grade customizable RAS, end-to-end security features, and software tools to integrate with fleet-management services

    leo-scalable
    Low-Latency DDRx & Custom Memory

    Flexible and scalable memory interface with low-latency data path to support JEDEC DDRx and custom memory interfaces

    leo-interoperable
    Seamless Interoperability

    Seamless interoperability with all major CPU, GPU and memory vendors, making it easy to manage, debug, and deploy at scale

    Orderable Information

    Leo CXL Smart Memory Controllers

    Orderable Part NumberImageDocumentsDescriptionCXL SpecCXL LinkMemoryCapacityExpansionPooling / SharingOrdering
    CM5082E-*CM5082E-*https://www.asteralabs.com/wp-content/uploads/2022/08/Astera_Labs_Leo_Aurora_Product_FINAL.pdfLeo E-Series CXL 2.0 x8 Smart Memory ControllerCXL 1.1/2.08x32G2ch DDRx Up To 5600MT/s2TBYesNohttps://www.asteralabs.com/product-details/cm5082e/
    CM5162E-*CM5162E-*https://www.asteralabs.com/wp-content/uploads/2022/08/Astera_Labs_Leo_Aurora_Product_FINAL.pdfLeo E-Series CXL 2.0 x16 Smart Memory ControllerCXL 1.1/2.016x32G2ch DDRx Up To 5600MT/s2TBYesNohttps://www.asteralabs.com/product-details/cm5162e/
    CM5162P-*CM5162P-*https://www.asteralabs.com/wp-content/uploads/2022/08/Astera_Labs_Leo_Aurora_Product_FINAL.pdfLeo P-Series CXL 2.0 x16 Smart Memory ControllerCXL 1.1/2.016x32G2ch DDRx Up To 5600MT/s2TBYesYeshttps://www.asteralabs.com/product-details/cm5162p/

    Aurora Smart Memory Hardware Solutions

    Orderable Part NumberImageDocumentsDescriptionCXL SpecCXL LinkMemoryCapacityExpansionPooling / SharingOrdering
    A1000-1P4AAA1000-1P4AAhttps://www.asteralabs.com/wp-content/uploads/2022/08/Astera_Labs_Leo_Aurora_Product_FINAL.pdfAurora A-Series CXL Smart Memory Add-in CardCXL 1.1/2.016x32G4x DDR5-5600 RDIMM slots2TBYesYeshttps://www.asteralabs.com/product-details/aurora-a-series/

    Evaluation Kits

    Orderable Part NumberImageDocumentsDescriptionCXL SpecCXL LinkMemoryCapacityExpansionPooling / SharingOrdering
    Leo-SVB-RevALeo-SVB-RevAhttps://www.asteralabs.com/wp-content/uploads/2022/08/Astera_Labs_Leo_Aurora_Product_FINAL.pdfLeo CXL Smart Memory Controller - System Validation BoardCXL 1.1/2.016x32G (PCIe CEM)4x DDR5-5600 RDIMM slots2TBYesYeshttps://www.asteralabs.com/product-details/leo-system-validation-board/

    Application Notes

    Videos

    Leo thumbnail
    09 Feb
    Demo: CXL Memory Pooling

    The industry’s first CXL memory pooling solution to reduce memory stranding, optimize memory utilization and reduce TCO for cloud servers from Astera Labs.

    Why we Test_ Leo Memory Connectivity Platform - asteralabs
    30 Jan
    How We Test: Leo Memory Connectivity Platform

    Learn how our comprehensive interoperability testing reduces design challenges, so you can accelerate time-to-market, streamline development efforts and reduce costs for designing and deploying heterogeneous infrastructure with CXL technology.

    Interop Testing with CXL 1.1 Host CPU’s and Popular DDR5 Memory Modules - asteralabs
    30 Jan
    Interop Testing with CXL 1.1 Host CPU’s and Popular DDR5 Memory Module

    After establishing the foundation of our rigorous testing, we worked with our customers to determine the most popular memory configurations for their systems and applications, to which we’ve included in our initial interop reports. We include 64GB DDR5-4800 RDIMMs from Micron, Samsung, and SK Hynix, each of which are tested with CXL 1.1-capable CPUs from AMD and Intel.

    • 1
    • 2
    • 3
    • Next »
    Load More loader

    FAQ

    Why is CXL™ important?

    CXL™ is needed to overcome CPU-memory and memory-storage bottlenecks faced by computer architects. Future data centers need heterogeneous compute, new memory and storage hierarchy, and an agnostic interconnect to tie it all together. CXL maintains memory coherency between the processor memory space and memory on attached devices to enable pooling and sharing of resources to provide higher performance, reduce software stack complexity, and lower overall system cost.​

    What kind of memory is supported over CXL™?

    Traditional DRAM and persistent storage class memory (SCM) are supported, allowing for flexibility between performance and cost.

    What is CXL™?

    Compute Express Link™ (CXL™) is an open industry standard interconnect offering high-bandwidth, low-latency connectivity between the host processor and devices including accelerators, memory expansion, and smart I/O devices. CXL utilizes the PCIe® 5.0 physical layer infrastructure and the PCIe alternate protocol to address the demanding needs of high-performance computational workloads in Artificial Intelligence, Machine Learning, communication systems, and HPC through the enablement of coherency and memory semantics across heterogeneous processing and memory systems.​

    What are the different devices that the CXL™ protocol supports?

    The CXL™ protocol supports three different type of devices:

    • Type 1 Caching Devices / Accelerators
    • Type 2 Accelerators with Memory
    • Type 3 Memory Buffer
    What are some CXL™ based applications?
    • Memory tiering in which additional capacity is applied with a variable mix of lower-latency direct-attached memory and higher-latency large capacity memory
    • Higher VM density per system by having more memory capacity attached
    • Large databases can use a caching layer provided by SCM to improve the performance
    What are the 3 CXL™ protocols?
    • CXL.io is used for initialization, link-up, device discovery and enumeration, and register access. It provides a non-coherent load/store interface for I/O devices similar to PCIe® 5.0.
    • CXL.cache defines interactions between a Host and Device, which allows CXL devices to cache host memory with low latency.
    • CXL.mem provides a Host processor with direct access to Device-attached memory using load/store commands.
    What kind of electrical signals support CXL™?

    CXL™ runs on PCIe® 5.0 electrical signals. CXL runs on PCIe PHY and supports x16, x8, and x4 link widths natively.

    What are the new features in the CXL™ 2.0 specification?

    CXL™ 2.0 adds support for switching, persistent memory, and security as well as memory pooling support to maximize memory utilization, reducing or eliminating the need to over-provision memory.

    What is the difference between memory pooling and sharing?

    In traditional servers, memory is directly connected to a specific CPU or GPU (i.e., locked behind the host) and can result in over-provisioning of memory resources when applications are not using the available memory. When the memory is over-provisioned to a specific host, the memory is now stranded and cannot be accessed by other hosts, thereby increasing data center costs. In addition, when memory is locked behind a host, the data being processed by the application needs to be copied through high latency interconnects if a different CPU or GPU needs access to the data.

    Memory pooling allows multiple hosts in a heterogenous topology to access a common memory address range with each host being assigned a non-overlapping address range from the “pool” of memory resources. Memory pooling allows system integrators to dynamically allocate memory from this pool, which reduces costs by reducing stranded memory and increasing memory utilization. Memory pooling is part of a growing trend for resource disaggregation or composability for heterogeneous solutions.

    Memory sharing allows multiple hosts in a heterogeneous topology to access a common memory address range with each host being assigned the same address range as the other host. This improves memory utilization similar to memory pooling, but also provides an added benefit of data flow efficiency since multiple hosts can access the same data. With memory sharing, coherency needs to be managed between the hosts to ensure data is not overwritten by another host incorrectly.

    How is data security managed with CXL™?

    CXL™ 2.0 supports Integrity and Data Encryption (IDE) and key exchange protocols for to provide end-to-end protection of data on the CXL link.

    What is RAS, and why is it important?

    “RAS” is the ability of the system to provide resilience starting from the underlying hardware all the way to the application software through three components collectively referred to as “RAS” features: ​

    1. Reliability: the ability of the system to detect and correct faults​
    2. Availability: how the system guarantees uninterrupted operation with minimal degradation
    3. Serviceability: the ability of the system to proactively diagnose, repair, upgrade or replace components at scale​

    Have more questions about general topics, such as PCIe 5.0, ordering, quality, etc.? See the full list of FAQs.

    The Importance of Security Features in a CXL Memory Controller to Protect Mission-Critical Cloud Data

    The explosion of modern applications such as Artificial Intelligence, Machine Learning and Deep Learning is changing the very nature of computing and transforming businesses. These applications have opened myriad ways for companies to improve their business development processes, operations, and security and to provide better customer experiences. To support these applications, platforms are being designed… Read More »

    Read More

    Unlocking Cloud Server Performance with CXL

    The increasing volume of data and complexity of models requires advancements in cloud server architecture to remove memory bottlenecks and unlock the performance for compute-intensive workloads, such as Artificial Intelligence and Machine Learning. Leo Memory Connectivity Platform for CXL 1.1 and 2.0 eliminates the memory bottlenecks inherent in today’s architectures and enables new heterogeneous infrastructure to increase performance and reduce costs for cloud-scale deployment.

    Read More

    Connectivity Is Key to Harnessing the Data Reshaping Our World

    As our appetite for creating and consuming massive amounts of data continues to grow, so too will our need for increased cloud capacity to store and analyze this data. Additionally, the server connectivity backbone for data center infrastructure needs to evolve as complex AI and ML workloads become mainstream in the cloud.

    Read More

    Hardware Solutions

    Rapidly implement diverse system topologies using our plug-and-play connectivity system boards. Our solutions include:

    • CXL Memory Cards: Increase cloud server performance and reduce total cost of ownership through memory expansion, pooling and sharing.
    • Smart Cable Modules: Active Copper-Based Solution to Address Reach, Signal Integrity and Bandwidth Utilization Issues for 100G/Lane Ethernet Switch-to-Switch and Switch-to-Server Interconnects.
    • Riser Cards: Extend PCIe/CXL technology slots and enable incredibly complex multi-connector topologies.
    • PCIe-Over-Cable Extender Cards: Connect a server head-node to a JBoF or JBoG without sacrificing speed.
    • GPU Booster Cards: Support external graphics (eGPUs) and enhance the gaming experience.

    See our Hardware Solutions >

    Services

    Rapidly design and achieve signal integrity peace of mind and avoid wondering if a PCIe 4.0/5.0, CXL 1.1/2.0 or 100G/Lane Ethernet technology-based design will work. Our team can help develop first-pass design success and accelerate time to market for your systems and boards.

    Contact us for more information about our design and testing services >

    • Aries PCIe/CXL Smart Retimers
    • Taurus Ethernet Smart Cable Modules
    • Leo CXL Memory Connectivity Platform
    • Cloud-Scale Interop Lab
    • Applications
    • Quality
    • Technology Insights
    • Contact Us
    • Careers
    • News & Articles
    • Support Portal    
    Subscribe for Updates
    Please enter your name.
    Please enter a valid email address.
    Subscribe

    Thanks for subscribing! 

    Something went wrong. Please check your entries and try again.

    By submitting this form, you are consenting to receive emails from Astera Labs. You can revoke your consent at any time by using the Unsubscribe link found at the bottom of every email.

    AsteraLabs-WhitewBug-Hz

    Copyright © 2023 Astera Labs, Inc. All rights reserved.

    Site Map I Privacy Policy I Terms of Use | Terms of Sale