The Stack Archive News Article

Verne Global unveils HPC as-a-service because of huge demand

Wed 6 Dec 2017

Engineers and scientists can take advantage of a new high performance computing platform thanks to a release from Verne Global.

Verne bosses say the new high performance computing (HPC) as-a-service platform has been released as a result of massive demand from its existing customers.

People specialising in computer-aided engineering, genomic sequencing, molecular modelling, grid computing, artificial intelligence and machine learning are likely to find the release useful, given the need for massive compute power in those fields.

High performance computing refers to systems that can carry out many more operations at once than regular computers due to an aggregation of processors. This level of compute power comes in particularly handy when tackling massive problems in science, such as mapping the origins of the universe.

Scaling HPC

The platform, called hpcDirect, is described by Verne as ‘powerful, agile and efficient’, and has been built to meet the increasingly challenging requirements of industries driven by data.

“Building hpcDirect was a direct response to overwhelming demand from our customers and tightly correlated with the market’s desire to move from a CapEx to an OpEx model for high performance computing,” said Dominic Ward, Verne Global managing director.

“With hpcDirect, we take the complexity and capital costs out of scaling HPC and bring greater accessibility and more agility in terms of how IT architects plan and schedule their workloads,” he added.

HPCAzure and Cray partner to bring supercomputing to the cloud

Innovation: HPE releases machine learning platform for the data centre

Scalability is an important part of the platform. According to the firm, it will give customers a bare metal service that comes with the ability to quickly give access to the full performance of HPC servers, ‘uncontended and in a secure manner.’

People wanting to use the service will be able to do so in a few ways. It can be added incrementally to existing HPC systems, or can fully support massive processing operations, working with petaflops of compute.

Customers with differing requirements can take advantage of the flexible nature of the service, which can be quickly provisioned to the size needed. The service’s clusters use architectures such as Intel Xeon processors and Mellanox Infiniband for quick connectivity between cores. There are different storage and memory options depending on what the customer needs.

Verne was founded in 2012 and builds new and innovative technology for the data centre industry, including facilities fuelled only be green energy. The firm’s bosses state that this release is another step in its work on smarter data centre technology.


analytics business Data Centre news UK
Send us a correction about this article Send us a news tip