Addressing Endpoint-Induced Congestion in Scale Out Accelerator Domain
Discover innovative RDMA mechanisms for handling endpoint congestion in accelerator domains, focusing on UDP/IP/Ethernet implementation and efficient congestion control for improved network performance.
Booting Operating Systems Using NVMe Over Fabrics - Network Boot Specification
Dive into the standardized approach for booting operating systems from networked NVMe devices, exploring the NVMe Boot Specification and its integration with UEFI and ACPI protocols.
Open MPI and Libfabric Integration for the Frontier Supercomputer
Delve into high-performance computing optimization through Open MPI and libfabric implementation on Frontier supercomputer, focusing on AMD MI250X architecture and HPE Slingshot NICs integration.
OpenFabrics Interfaces Support and GPU Integration in MPICH
Discover the latest updates on OFI integration in MPICH, focusing on GPU support implementation, network API enhancements, and Libfabric development for high-performance computing systems.
DPU-Powered File System Virtualization - Improving Cloud Infrastructure Efficiency
Discover how DPU-powered file system virtualization improves cloud infrastructure efficiency by offloading client-side drivers, enhancing tenant isolation, and delivering 4.4× CPU efficiency gains without host modifications.
Status of OpenFabrics Interfaces (OFI) Support in MPICH
Discover the latest developments in OpenFabrics Interfaces (OFI) support within MPICH, exploring implementation status and advancements in high-performance computing infrastructure.
Engineering CXL Memory Expander Device for Production - Challenges and Solutions
Dive into the engineering challenges and solutions for CXL memory controllers, focusing on interoperability, scalability, and efficient deployment strategies for Type-3 implementations.
Rack Manifold Requirements and Qualification for Liquid Cooled Systems
Explore design guidelines and testing methodologies for rack-mounted manifold systems in liquid-cooled IT equipment, covering material selection, performance standards, and implementation.
PCIe Extended Connectivity Requirements for Rack-Level Disaggregation
Explore PCIe Extended Connectivity requirements for rack-level disaggregation, covering NVMe and CXL interconnects for compute, acceleration, memory, and storage modules using Gen5 & Gen6 technologies.
Signal Integrity in Immersion - Overcoming Challenges for High-Performance IT Platforms
Explore signal integrity challenges in immersion cooling systems, examining testing methods and design strategies for optimizing high-performance IT platform performance and data transmission.
Hardware Fault Management Standards and Interfaces - Project Update
Discover the latest developments in hardware fault management standards, including RAS API workstream, FMFM initiatives, and industry approaches to error injection standardization.
Frame-Based Deployment of Rear Door Heat Exchanger for Data Center Cooling Efficiency
Discover how to efficiently deploy Rear Door Heat Exchangers in data centers using frame-based systems to improve cooling and reduce energy consumption while maintaining hardware flexibility.
LTPI Reference Implementation for DC-SCM 2.0 - Overview and Design Guide
Discover Intel's LTPI reference implementation for DC-SCM 2.0, exploring its scalable architecture, verification testing package, and practical hardware implementation guidelines.
Flash-Centric Storage Architecture Enabled by E1.S Form Factor
Explore flash-centric storage architecture and EDSFF form factors, focusing on NAND technology optimization, data center efficiency, and advanced features for enhanced storage performance.
A Reliable and Low Latency Ethernet Hardware Transport
Discover Google's innovative approach to achieving reliable, low-latency Ethernet hardware transport through advanced networking protocols and optimization techniques.
Adapting the Data Center to New AI Trends
Explore how AI trends reshape data center infrastructure, focusing on high-density server racks, liquid cooling solutions, and innovative approaches to meet evolving computational demands.
Generative AI and the Outsized Role of Serdes in Data Center Performance
Explore how AI's data center evolution puts interconnect and serdes technology at the forefront of scalability, examining power efficiency challenges and performance impacts across multiple scales.
Memory Tiering and Persistence Enablement with CXL Memory Module
Explore Samsung's CXL Memory Module solution for optimizing compute architectures, featuring memory tiering and persistence capabilities designed for AI inference and in-memory database applications.
Next Steps in AI Technology Adoption - Data Center Challenges and Solutions
Explore key challenges and solutions for AI technology adoption in data centers, from power management to cooling systems and global deployment strategies.
Revolutionizing SONiC Routing - Tackling Challenges in FRR Integration and Scaling up Routing Features
Explore the evolution of SONiC routing through SRv6 implementation, FRR integration challenges, and scaling solutions for handling millions of routes in data center networks.
Next Generation Fabric for AI/ML Use Cases - Disaggregated Scheduled Fabric Using SONiC
Explore how Disaggregated Scheduled Fabric with SONiC optimizes network infrastructure for AI/ML applications, offering superior bandwidth management compared to traditional ECMP-based IP fabrics.