Go Zero-to-HPC with “Cluster Kickstart”

Picture 1
As cluster computing rapidly takes over the high performance computing (HPC) market, the need for enterprise class services and supported solutions is becoming paramount. Clustercorp leverages its extensive experience in deploying efficient HPC Clusters with the Rocks™ cluster distribution to take its customers from Zero-to-HPC with a packaged service and support program we call “Cluster Kickstart”.

“Cluster Kickstart” includes onsite cluster deployment that starts with a freshly racked system and ends with a complete Rocks+ installation validated with a series of HPC benchmarks (including a tuned HPL run). “Cluster Kickstart” includes the configuration of all applications, compilers and hardware supported by Rocks+. The package also includes up to 6 months of Rocks+Support, which ensures end-users start off on strong footing with a single 800 number and email address to contact for all support inquiries.

Customers can extend “Cluster Kickstart” services by adding a multi-year support contract or adding extra onsite time for custom configurations. Call for pricing on extras or for deployments larger than 256 nodes. Check our Rocks+Hardware page for recommended solutions.

rocksplussupport5

“Cluster Kickstart” is built around the Rocks+ cluster distribution, which adds two critical components necessary for deploying Rocks as a commercial solution:

Rocks+Support - Rocks+Support addresses the biggest problem in using an open source solution such as Rocks – “the support gap”. While community support for active open source projects like Rocks is a great start, a complex, high-value deployment such as an HPC Cluster practically requires the formal backing of an experienced support team. Clustercorp was founded by the core members of the Rocks community for this exact purpose. “Cluster Kickstart” includes an introductory Rocks+Support contract, which can be extended to 1, 2, or 3 years.

Rocks+Rolls - The second obstacle to a successful cluster deployment commonly pops up shortly after cluster delivery, when system administrators begin the burdensome process of application installation and configuration across their new cluster. Clustercorp leverages Rocks Rolls to automate the installation of software across a cluster and prevent long-term “software skew”.

This product includes software developed by the Rocks Cluster Group at the San Diego Supercomputer Center at the University of California, San Diego and its contributors.