Overview
Explore
Services
Partners
24 results found
Revving Up ECMP Routing for AI/ML Workloads
Dive into the world of high-performance networking. In this article, you will discover solutions to optimize Ethernet for AI/ML and HPC environments, ensuring seamless GPU synchronization and enhanced network performance.
How Rafay Simplifies Multi-Tenant GPU Workload Orchestration and Governance at Scale
Rafay Systems revolutionizes AI infrastructure by simplifying multi-tenant GPU orchestration and governance. It unifies complex environments into a secure, self-service platform, maximizing GPU utilization and enabling rapid AI innovation. Rafay's approach empowers platform teams to deliver efficient, compliant and scalable AI solutions, transforming operational challenges into strategic advantages.
Cisco UCS M8 Series: 8th-Generation Compute Unleashed
From AI and analytics to virtualization and edge computing, Cisco's UCS M8 portfolio offers a comprehensive range of platforms designed to meet the demands of nearly any modern workload.
Priority Flow Control: The Stop Light of the AI Networking Highway
Priority Flow Control (PFC) acts like a traffic light, pausing specific data flows to prevent congestion and data loss in AI networks. By enabling lossless Ethernet fabrics, PFC enhances performance in AI workloads, crucial for applications like RDMA. Explore its implementation across various networking hardware for optimal network efficiency.
The Quantum Computing Threat - Part 1
As we become more "any connection from anywhere" the means of secure transactions becomes even more critical.
Workload Management & Orchestration Series: NVIDIA Run:ai
This series explores various workload management and/or orchestration tools at a high level to help you understand each tool's characteristics beyond the marketing hype. This blog covers NVIDIA's Run:ai.
Managing Network Congestion in AI: The Role of Explicit Congestion Notification (ECN)
This post delves into Explicit Congestion Notification (ECN) as a tool for managing network congestion in AI and high-performance computing environments. By slowing down packet transmission, ECN enhances performance, though its implementation poses challenges, necessitating careful configuration and tuning for optimal results.
Scalable RoCEv2 Deployments: Overcoming Network Challenges in AI Data Centers
The blog discusses the deployment and testing of RoCEv2 technology in AI data centers, highlighting the challenges and solutions provided by Spirent.
Hands-On Learning of Quantum Computing
Quantum computing holds long-term strategic promise, but near-term progress is constrained by scarce, noisy and immature hardware. This blog presents a quantum computation simulator that allows organizations to engage with quantum computing today, without reliance on physical quantum devices. The simulator represents quantum states as explicit, inspectable data structures, enabling engineers and technical leaders to directly observe and reason about superposition, interference, entanglement and measurement as executable program behavior.
Workload Management & Orchestration Series: Slurm Workload Manager
This series explores various workload management and/or orchestration tools at a high level to help you understand each tool's characteristics beyond the marketing hype. This blog covers Slurm.
Workload Management & Orchestration Series: ClearML
This series explores various workload management and/or orchestration tools at a high level to help you understand each tool's characteristics beyond the marketing hype. This blog covers ClearML.
HPE Aspire 2022 Takeaways
Hewett Packard Enterprise (HPE) recently completed its in-person and virtual Aspire 2022 event (May 9-13th, 2022 at the Gaylord Opryland Resort & Convention Center in Nashville, TN). The event allowed HPE internal and channel professionals to gather for three-plus days to learn, connect and grow one's technical knowledge of HPE products and solutions. What mattered to me most was access to updated content, industry experts and HPE engineers skilled in high-performance computing (HPC).