Last week, Wolfram Research announced that it was developing a cloud computing service for users of its Mathematica software used to graph and understand complex mathematics, physics, and engineering problems.Together with Nimbis Services and R Systems, Wolfram is creating a service that allows users of the Mathematica product line to access high-performance computing (HPC) services in the cloud. The models that Mathematica users create are typically computationally intensive, often taking hours or even days to execute, which make them well-suited to consume cloud-based HPC services.

The collaboration of Wolfram Research, Nimbis Services, a clearing-house for accessing third-party computing resources and commercial software, and R Systems, a provider of computing resources to the commercial and academic research community, will make it easier for Mathematica users to make the transition from desktop technical computing to HPC systems without necessarily having to connect to powerful and expensive supercomputers. According to Deborah Wince-Smith, president of the Council on Competitiveness, a Washington, D.C.-based nonprofit whose goal is to increase the United States' economic competitiveness in the global market, "High-performance computing systems remain a largely underutilized competitiveness asset in the United States for the majority of companies. Opening access to HPC represents a huge productivity opportunity for the nation and a competitiveness transformation challenge."

Nimbis Services, a startup company in the digital analysis computing industry, will enable the Mathematica cloud service to access many diverse HPC systems, including TOP500 supercomputers and the Amazon Elastic Compute Cloud ("EC2"), according to Robert Graybill, Nimbis president and CEO. "Our partnership with Wolfram Research immensely benefits software users attempting to increase efficiency and capacity," adds R Systems founder Brian Kucic. "As Mathematica users seek to extend resource capacity, the exceptionally large memory of our multicore HPC resources and the double-data rate and quad-data rate InfiniBand network will increase performance."

The role of the R Systems InfiniBand network within the Mathematica service is important, since a recent paper, Cloud Computing for Parallel Scientific HPC Applications: Feasibility of Running Coupled Atmosphere-Ocean Climate Models on Amazon's EC2, concluded that "the performance of running an HPC climate model on an EC2 computing cloud on-demand cluster system was below the level seen at dedicated, supercomputer centers." However, the report said that cloud systems more closely targeted to HPC applications that feature a specialized interconnect such as Myrinet or InfiniBand, "could be a very powerful addition to the HPC landscape in the very near future."