hero-portable


Launch presentation

Portable Clusters, high-performance mobile supercomputing

With a portable cluster by Nor-Tech, power, possibility and potential can travel anywhere.

Ushering in a new era of onsite, high-performance computing, Portable Clusters by Nor-Tech are highly ruggedized and extremely mobile. The portable HPC clusters are lightweight and available in military shock mounted 19″ rackmount cases that adhere to MIL-SPEC standards. The mobile clusters run cool, save energy and are able to operate up to 128 processor cores on just two 120VAC 20-amp circuits.

It’s a high-performance solution that can tackle any global challenge in today’s computing environment.

Benefits of a Portable Cluster over large data clusters

  • Cost of Ownership is reduced: less cooling = lower power consumption
  • Because of the relatively low initial purchase price the funding complexities are much reduced versus a large cluster.
  • Shorter queue wait times, smaller user groups
  • Departmental or individual use means fewer security vulnerabilities.

Clusters are available with any of the following software installed and configured:

  • Windows Server 2008/HPC
  • Various Linux Distro’s of CentOS, RHEL, SUSE, Fedora Core, etc.
  • Warewulf Cluster Management Suite
  • Rocks Cluster Management Suite
  • Sun Grid Engine
  • Torque
  • MPI, MPICH
  • Ganglia
  • InfiniBand Intra-cluster Networking
  • Mryinet Intra-cluster Networking
  • Viz Cluster options are available
  • IPMI management options are available
  • What sets Nor-Tech Apart from every other HPC cluster builder?

    Click here for the full rundown.

    In a nutshell:

    • We start with the right tools for the job such as HPC-class RAM memory and Enterprise/Datacenter hard drives.
    • All clusters leave Nor-Tech already set to the client’s IP and DNS scheme; setup is plug & play at the client site.
    • We ensure each cluster is fully configured and ready to deploy when it leaves our facility.
    • We provide a complete set of references including user manuals, a network topology guide, and a Visio drawing.
    • We label all cables with ID tags on both ends.
    • We provide a customized bare metal cluster recovery disk.
    • Every client is asked to log into their new cluster just after burn-in completion to verify operation and setup users.
    • We assist clients with job script creation.
    • Our clusters all ship with Storage Guard to alert clients before their clusters run out of space; and ambient temperature monitoring and protection to protect clusters in the event air conditioning fails.
    • Our clients have the option of an onsite spare parts closet for minimum down time in the event of a hardware failure.
    • We guarantee no wait-time support.
    • Everything we sell is built by us in Minnesota and supported by us around the world.
    • Request Info

      * indicates required field