Red Hat Debuts Multi-Architecture Solutions for HPC at ISC 2017

Print Friendly, PDF & Email

In this video, Yan Fisher from Red Hat describes the company’s multi-architecture solutions that were on display at ISC 2017. As Yan wrote in his recent blog, this is the first time Red Hat has had an exhibit at ISC:

Twice a year the most prominent supercomputing sites in the world get to showcase their capabilities and compete for a Top500 spot. With Linux dominating the list, Red Hat is paying close attention to the latest changes announced at the ISC 2017 in Frankfurt, Germany. While supercomputers of the past were often proprietary, the trend of building them out of commodity components has dominated the landscape in the past two decades. But recently the definition of “commodity“ in HPC has been morphing. Traditional solutions are routinely augmented by various acceleration technologies, cache-coherent interconnects are becoming mainstream and boutique hardware and software technologies previously reserved for highly specialized solutions are being adopted by major HPC sites at scale.

Developing new and adapting existing highly scalable applications to take advantage of the new technological advances across multiple deployment domains is the greatest challenge facing HPC sites. This is where the operating system can provide a unified interface to the underlying hardware and interconnects and serve as a foundation for modular and standardized application stacks that take advantage of enhanced system capabilities.

Life Across Architectures

In November 2016, Red Hat joined the OpenHPC Project, bringing its expertise and leadership in the open source world to one of the leading open supercomputing communities. Red Hat continues to help drive open innovation and standardization in high-performance computing;

  • We’ll be showcasing the power and flexibility of Red Hat Enterprise Linux across multiple architectures, including ARM v8-A, x86 and POWER little endian, by way of the Game of Life. Adrian Reber created an intriguing demo that you could experience first hand in our booth or read about it in Adrian’s post.
  • Another great example of partner-driven collaboration is the advanced technology demonstration by Hewlett Packard Enterprise, Cavium, and Red Hat. Come to the booth and see how our efforts deliver inherent value to HPC users.

Open Source, Innovation and the Future of HPC

Historically, HPC workloads have had to rely on custom-built software stacks, most of which relied on proprietary software and overly-specialized hardware. Red Hat is bringing change to the supercomputing arena, from the world’s leading enterprise Linux platform tailored for HPC workloads to massively scalable, fully open cloud infrastructure, along with the management and automation technologies needed to keep these deployments running smoothly.

At ISC, Red Hat showcased the following technologies:

See our complete coverage of ISC 2017

Sign up for our insideHPC Newsletter