ARM-based HPC systems will be available by the end of 2017. At ISC High Performance 2017 Penguin Computing and Bull both announce that they will deliver products tailored for the HPC industry. Both companies systems will feature Cavium’s ARMv8-based ThunderX2 platform.
The announcement from the Atos owned company Bull comes after years of development of ARM systems through the EU funded Mont Blanc project.
Scientists and researchers working on the Mont-Blanc project built and designed the first ARM-based HPC cluster based on Bull blades, demonstrating the viability of using ARM technology for HPC.
The project has been in operation since 2011, initially aimed to define a computing architecture capable of delivering HPC scale and performance using ARM processors. A three-year extension was added to address the development of parallel programming and support for the ARMv8 64-bit processors.
In October 2015, a third phase of the Mont-Blanc project - coordinated by Bull - was initiated to ensure that the system would be ready to support HPC applications. This included the design and of a ‘high-end’ HPC platform.
‘The Mont-Blanc partners are delighted by this announcement. We have long been convinced that, due to their energy-efficiency, ARM processors offer tremendous potential for High-Performance Computing. We would like to give special thanks to the European Commission for their support throughout our project’ said Etienne Walter, coordinator of the Mont-Blanc project.
The new Bull system, the Sequana X1310 is being introduced part of the Bull X1000 range of supercomputers.
The Bull Sequana X1310 blade includes three compute nodes, each equipped with two latest generation 64 bit ThunderX2 processors from Cavium, based on the ARMv8 instruction set. The new model will be available in Q2 2018.
Penguin Computing announced the availability of its Tundra Extreme Scale (ES) server platforms based on Cavium, second generation ThunderX2 processors. Tundra ES Valkre servers powered by ThunderX2 processors are now available for public order, with the standard 19” rack mount models shipping in the Q3 2017.
This new system is Penguin Computing’s second generation of ARM-based server platforms. The new ThunderX2-based system will focus on highly-scalable Hyperscale and HPC-type workloads including big data, large-scale graph analytics, molecular dynamics, and Ceph/Cloud storage. The performance will be driven by ThunderX2 ARMv8 SOCs optimised for these workloads, with high-performance custom cores, dual socket coherent connectivity and high memory bandwidth and capacity.
‘By extending our product roadmap to Cavium’s second generation 64-bit ARMv8 CPUs in our Tundra family of Open Compute servers we again step up our leadership position. Our customers get outstanding value from the efficiency and flexibility enabled by OCP infrastructure combined with best-in-class compute performance coming from Cavium’s ThunderX2 offering’ said Jussi Kukkonen, vice president, advanced solutions, Penguin Computing.