Intel Omni-Path: The Next-Generation Fabric
Nor-Tech’s Demo Cluster
Intel Omni-Path, along with Intel Orchestrator will be integrated into our innovative demo cluster. As a widely acknowledged pioneer and leading expert in the HPC space, Nor-Tech has the hardware, software, and expertise to demonstrate the immediate ROI of running your applications on a Nor-Tech cluster. This is a no-cost, no-strings opportunity.
To Take Advantage of the Demo Cluster,
Or call 952-808-1000/877-808-1010
Intel Omni-Path Architecture (Intel OPA), an element of Intel Scalable System Framework, delivers the performance for tomorrow’s high performance computing (HPC) workloads and the ability to scale to 10s of 1,000s of nodes—and eventually more—at a price competitive with today’s fabrics. The Intel OPA 100 Series product line is an end-to-end solution of PCIe adapters, silicon, switches, cables, and management software. As the successor to Intel True Scale Fabric, it is built on a combination of enhanced IP and Intel technology.
The Future of High Performance Fabrics
Intel OPA is designed specifically to address issues of poor HPC performance and expensive scaling in existing standards-based high performance fabrics. The enhancements of Intel OPA will facilitate the progression toward Exascale while cost-effectively supporting clusters of all sizes. Intel OPA is designed to provide:
- Features and functionality at both the host and fabric levels to greatly raise levels of scaling.
- CPU and fabric integration necessary for the increased computing density, improved reliability, reduced power, and lower costs required by significantly larger HPC deployments.
- Fabric tools to readily install, verify, and manage fabrics at this level of complexity.
Key Features and Innovations
- Cost Reduction: Reducing fabric costs was a major design goal for Intel OPA. Using denser switching with a 48-port switch chip delivers a 33% cost savings over the standard 36-port alternative–fewer switches and cables are required. Power and maintenance costs are reduced as well.
- Packet Integrity Protection: This is an area where Intel OPA really shines. It allows for rapid and transparent recovery of transmission errors between sender and receiver on an Intel OPA link. It enables recovery of transient errors whether between a host and switch or between switches. This eliminates the need for transport level timeouts and end-to-end retries and the burden of a heavy latency penalty associated with alternate error recovery approaches.
- Adaptive Routing: This monitors the performance of the possible paths between fabric end-points and selects the least congested path to rebalance the packet load. Adaptive Routing scales as fabrics grow larger and more complex
- Dispersive Routing: Intel OPA fabric supports a variety of routing methods including defining alternate routes that disperse traffic flows for redundancy, performance, and load balancing. Instead of sending all packets via a single path, Dispersive Routing distributes traffic across multiple paths.
- Traffic Flow Optimization: This optimizes the service quality of messages sent on an egress port based on priority, virtual lane, or service level.
- Dynamic Lane Scaling: This allows an operation to continue even if one or more lanes of a 4X link fail, eliminating the need to restart or go to a previous checkpoint to keep the application running; the job can run to completion before taking action to resolve the issue.