4.2. Use Case 2: Using Red Hat Storage for High Performance Computing

Enabling technologies and techniques has become more accessible, and in data-intensive industries such as financial services, energy, life sciences, High Performance Computing (HPC) has gained popularity in recent years. These industries continue to face several challenges as they use HPC to gain an edge in their respective industry.
The key challenges enterprises face are scalability and performance. Red Hat Storage Server family provides an open source, scale-out network attached storage (NAS) and object storage software solution that is designed to work seamlessly with industry standard x86 servers.
Red Hat Storage Server is built on Red Hat Enterprise Linux operating system. It provides freedom of choice to customers by allowing them to deploy cost-effective, scalable and highly-available storage without compromising on scale or performance. It can easily be deployed on-premise, in private clouds, in public cloud infrastructures or hybrid cloud environments and is optimized for high-performance computing workloads that demand high bandwidth and throughput performance.
Use Case 2: Using Red Hat Storage for High Performance Computing

Figure 4.2. Use Case 2: Using Red Hat Storage for High Performance Computing

4.2.1. Key Features of Red Hat Storage Server for High Performance Computing Use Case

This section describes the key features of Red Hat Storage for High Performance Computing use case.
  • Petabyte Scalability
    Red Hat Storage Server’s fully distributed architecture and advanced file management algorithms allow it to support multi-petabyte repositories with ease.
  • High Performance with no bottleneck
    Red Hat Storage Server enables quick fast file access by algorithmically spreading files evenly throughout the system, without a centralized metadata server. As the nodes can access storage nodes directly, hot spots, choke points, and other I/O bottlenecks are eliminated. Hence, contention for data is reduced and there is no single point of failure.
  • Elastic Scalability
    Storage volumes are abstracted from hardware, allowing each to be managed independently. Storage can be added or removed from the storage pools while data continues to be available, with no application interruption. Volumes can be expanded or shrink across machines. It can be migrated within the system to rebalance capacity or add/remove systems on-the-fly, allowing HPC environments to scale seamlessly.
  • Infiniband Support
    Red Hat Storage Server supports IP over Infiniband (IPoIB). Infiniband as a back-end interconnect for the storage pool is recommended as it provides additional options for maximizing performance. Using RDMA as a mount protocol for its native client is a technology preview feature.
  • Compatibility
    Due to native POSIX compatibility and support for the SMB, NFS and HTTP protocols, Red Hat Storage Server supports existing applications with no code changes required.