Jump to content United States-English
HP.com Home Products and Services Support and Drivers Solutions How to Buy
» Contact HP
HP.com home
HPC  >  Tools

HPC solutions: LSF on AlphaServer systems

» 

High Performance Computing

Products & Solutions

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

Resources

» News & Promotions
» Subscribe to newsletter
» HPC Library
» HPC Podcasts
» HPC events
» Benchmarks
» How to Buy

Learn more about

» Customer successes
» HP Collaboration & Competency Network
» HP-CAST User Group meetings
» Catalysts for Innovation Program
» HP Labs
» HP Partners
» Explore the HP Alternative program
» Contact HP
Content starts here

Optimal solutions for high performance network computing

Combine Platform Computing with HP, a recognized leader in high-performance computing, and you can expect the most from your UNIX and Windows NT distributed computing environment. Platform Computing's LSF Suite is the industry standard for workload management. LSF Suite gives you a reliable, scalable means of harnessing and effectively managing your computing resources in support of business-critical applications.

LSF highlights

LSF delivers substantial improvement in overall system performance by dynamically distributing batch, interactive, and parallel jobs across the network. As a result, LSF helps you get maximum value from your workstations, servers, and software licenses.

Primary features

  • Interactive load sharing
    - monitors the load on every machine in a cluster and provides dynamic access to remote hardware and software resources.

  • Batch scheduling
    - smoothly balances the load over time, providing site-configurable control of resource sharing.

  • Parallel computing
    - supports PVM, MPI, and HPF

  • Fault tolerance
    - system is available as long as one host is up; no job lost even if all hosts are down
    - job checkpointing and migration

Primary benefits

Encompassing distributed load sharing and dynamic job scheduling for heterogeneous UNIX and Windows NT computing environments, Platform's LSF Suite delivers:

  • faster product development
  • higher quality of service
  • shorter time-to-market
  • improved products
  • greater customer satisfaction
  • increased profits

LSF Batch
- seamless, highly configurable batch job scheduling for distributed and heterogeneous environments

LSF JobScheduler
- integrates heterogeneous servers into a highly robust "virtual mainframe". It supports calendar-based and event-based job scheduling

LSF MultiCluster
- extends the Workload Management benefits and functionality of the LSF Suite across WANs and remote clusters

LSF Analyzer
- provides comprehensive workload analysis across a cluster of computing resources, enabling system administrators and managers to make informed capacity-planning decisions, remove bottlenecks, and tune overall system performance

LSF Make
- a distributed, parallel Make program that simultaneously dispatches tasks to multiple hosts, to expedite software builds, regression tests, and inter-dependent groups of simulation job runs

LSF diagram
        LSF and HP are optimal solutions for network computing

Your needs How LSF helps

Control the number of jobs on a server to avoid host overload

LSF's batch scheduling smoothes the workload over time and controls the sharing of valuable resources

Share the load across several hosts and present a single system image

LSF's support for load sharing and distributed batch scheduling turns a cluster into a scalable mainframe or supercomputer

Integrate your servers with desktop workstations

Configure all desktops and servers into an LSF cluster so that jobs can be transparently scheduled to run on the best machines

The HP advantage

HP's high-bandwidth, low latency Memory Channel enables multiple shared-memory systems to be interconnected to form supercomputer-class platforms, at a fraction of the cost of traditional supercomputers.

More applications are being developed using message-passing libraries to run on computers with physically distributed memory. These applications can use HP's highly-optimized message-passing libraries to run Memory Channel clusters of a AlphaServers. Scalable applications and Memory Channel allow up to 96 powerful Alpha microprocessors to be applied to single problems.

> Further strengths include:

  • Choice of Alpha and Intel processors, UNIX and Windows NT operating systems for flexible deployment

  • The fastest microprocessor in the industry, for unmatched application performance

  • Advanced 64-bit computing to handle large, complex problems

  • PowerStorm graphics, the best in price/performance

  • Leadership GIGAswitch and MultiSwitch 900 network backbones for the ultimate throughput in high performance networked computing

  • High performance, high capacity StorageWorks products to archive large volumes of data

  • The scalability and continuing evolution of Alpha architecture ensures growth to protect and enhance your investment

  • Unmatched systems support and integration services ensure the highest availability and the most efficient use of your equipment

Printable version
Privacy statement Using this site means you accept its terms
© 2008 Hewlett-Packard Development Company, L.P.