Platform Computing will this week [23-24 October] showcase new software as it anticipates virtualisation driving demand for grid computing across enterprises.

Grid is often associated with high-performance computing (HPC) and niches such as academic research but in an interview with IT Week today, Platform chief executive Songnian Zhou said the picture is changing.

“Adoption of grid is going from HPC and verticals to broad commercial deployments and virtualisation is going to be a big boost for grid computing because it allows you to allocate resources and move applications around without disrupting the application in the VM container,” Zhou said.

Other IT trends such as low-cost blade servers, open-source software stacks and standardised switches are also helping make grids a reality at financial services companies such as Citigroup and investment banks like Lehman Brothers, Zhou claimed.

Platform’s EGO software helps firms orchestrate and manage grid services and Zhou insisted that operating system, systems management vendors or others are unlikely to infringe on the space.

“CIOs are not paying attention to individual boxes anymore so the datacentre needs to have an operating environment. This is a very different beast to anything that I’ve seen in the systems management market. IBM can extend WebSphere but that’s just creating another silo. We’re complementary to all these things.”

Zhou also said he is relaxed about grid-computing, usually used to mean huge projects outside the organisation’s wide-area network, and server clustering terms, being used interchangeably.

“Even the grid buzzwords are short for other buzzwords so if we focus on the business need, the rest does not matter,” Zhou said.

“SETI@Home [a well-known grid-based research project] is an extreme example. What’s needed is a model for running applications in a distributed environment with a service-oriented infrastructure governed by policy.”

At the Platform European Grid Conference in Paris today and tomorrow, the company will show off the recently announced Open Cluster Stack for Linux and a forthcoming upgrade to the LSF product for grid-enabling products.

LSF 7.0, due to be formally announced in early November, improves scalability, robustness and adds more sophisticated scheduling, Platform said.