HPC  >  Clusters

Unified Cluster Portfolio

» 

High Performance Computing

Products & Solutions

» Systems
» Clusters
» Storage
» Visualization
» Software
» Grid computing
» Services and Support
» Solutions
» Power and Cooling

Resources

» News & Promotions and Case studies
» HPC Cluster Edge Blog
» Customer successes
» HPC Library
» HPC Podcasts
» HPC events
» Benchmarks
» How to Buy
» Subscribe to newsletter

Learn more about

» HP-CAST User Group meetings
» Catalysts for Innovation Program
» HP Labs
» HP Partners
» Explore the HP Alternative program
» Contact HP
Content starts here

» HP Cluster Platforms

» Operating Environments

» Cluster Management

Storage

» Applications & Tools

» Cluster Interconnects

Storage

HP offers customers choice and flexibility in their storage solutions including HP StorageWorks X9000 Network Storage Systems and Linux-based Lustre technologies for scalable storage.

The HP X9000 Network Storage Systems offer customers flexibility for a variety of environments and configurations. The X9000 systems provide scalable aggregate bandwidth across many files, with scalable capacity, native Linux and Windows clients, NFS and CIFS export allowing access by a large number of loosely coupled systems including desktops & servers outside the datacenter. The HP X9000 storage systems are best for the broad SCI market and HPC including Media, FSI, Bioinformatics, GeoScience, Hosting/Web, and Cloud.

For more information on HP X9000 storage options visit www.hp.com/go/x9000

HP offers Lustre-based storage options through a relationship with Data Direct Networks (DDN). Lustre-based storage is optimized for large single files with scalable bandwidth, sequential throughput. These are generally closely coupled with an HPC cluster, sharing the cluster’s high performance interconnect, often InfiniBand.

Some cases for which Lustre is optimized include:

  • Large single files with scalable bandwidth, sequential throughput
  • Maximum throughput per server (native IB protocol , direct I/O)
  • Requires storage optimized for large sequential I/O
  • Storage optimized for large sequential I/O
  • Where parallel applications require many distributed processes writing to a single, large file. Examples include MPI I/O

Lustre is best used for distributed, parallel applications in traditional HPC applications where closely coupled solutions sharing the cluster’s high performance interconnect are required.

»  Unified Cluster Portfolio overview
Printable version