Harnessing hybrid cloud for HPC | SUSE Communities

Harnessing hybrid cloud for HPC


Harnessing Hybrid Cloud for High Performance Compute

From Boffins to Big Business

As a grizzled veteran of the IT industry, I have been involved in many high performance computing (HPC) projects over the years, both from a hardware and software perspective. I have always found them to be intensely interesting mainly because the projects were deeply scientific in nature, whether it be decoding the human genome, designing better, more efficient vehicles or even deep space research.

What’s different now is the emergence of HPC into the mainstream. Instead of it just being the preserve of academics, scientists and other boffins, normal commercial organisations are trying to harness the power of HPC to solve their business issues, notably through its application to AI and Machine Learning.

As today’s technology creates vast hordes of unstructured data, unlocking the business value therein has become a key competitive advantage and almost the Holy Grail of Digital Transformation for many organisations. HPC has a key part to play in this as deriving insight from large data sets has been a major component of scientific research for many years.

An Elastic, Hybrid, High Performance Compute Cloud Solution

SUSE has a long and distinguished history in HPC, both as a provider of a dedicated HPC Operating System and more latterly with our CEPH-based software defined storage, as a target for all that lovely data.

In the UK, we are lucky to partner with some great organisations in this space. Dublin-based Securelinx, have an outstanding track record of delivering Linux and HPC projects and have a highly skilled team focussing on HPC and AI solutions. Their latest innovation, showcased at SUSECON in Nashville earlier this year, is a Hybrid Cloud solution for HPC workloads, using Slurm and cloud native APIs to create an elastic, hybrid, high performance compute cloud.

All the Results + Cost Savings

This solution has the crucial advantage of allowing HPC projects to make use of cloud resources for burst capacity. Rather than building huge compute farms at potentially exorbitant costs just for the duration of the project, using cloud resources only when required can have a real beneficial impact on the project budgets.

It’s well worth taking a look at this video for all the details.

And if you have any questions or would like more information, get in touch with one of us:




Leave a Reply

Your email address will not be published. Required fields are marked *

No comments yet

Avatar photo