Exascale HPC Needs An Application Innovation Spark
Every evolution in computing hardware brings with it big challenges for software developers. …
Every evolution in computing hardware brings with it big challenges for software developers. …
The cadence of server processor launches by the remaining companies that still etch their chips has slowed in recent years, starting first with the low-volume players like IBM, Oracle, and Fujitsu and now possibly spreading to Intel with its Xeon line and already baked in with its Xeon Phi line with a roughly three year span between generations. …
Lining up the architectures of future supercomputers is interesting because it gives us a glimpse of what may be in the corporate datacenter many more years out. …
When it comes to high performance computing, IBM is in a phase change that will take it several years to complete with its key OpenPower infrastructure partners, Nvidia and Mellanox Technologies. …
Hot on the heels of the closing of the deal that divests its semiconductor business and places it in the hands of Globalfoundries, the former chip making business of AMD that is controlled by the government of Abu Dhabi, IBM and its academic and chip industry partners have announced that they have successfully etched chips with transistors that are 7 nanometers in size – significantly smaller than current processes and extending the Moore’s Law curve one more step. …
Just as the fervor died down around the massive deals for forthcoming pre-exascale supercomputers in the United States following the CORAL procurements (most recently, with the announcement of Aurora—the only one of three such HPC deals that is not betting the future on IBM OpenPower systems) the supercomputing spark was stoked again, this time, from across the pond. …
Burst buffers are a hot topic these days. At the most simple level, a burst buffer consists of the combination of rapidly accessed persistent memory with its own processing power. …
The rollout of Power8 systems that IBM started last year is nearly completed as the company has put its largest Power E880 configurations into the field, giving an upgrade path to customers who had been using Power7 and Power7+ systems to run large databases and online transaction processing workloads and to others who are looking for more scalable machines to run in-memory databases and their applications like SAP’s HANA stack. …
When one looks at major milestones in supercomputer history, one of the relatively recently markers rests with IBM, which began work its work to create a massively parallel system for protein folding research. …
In the wake of today’s announcement of the codes selected to be tuned for the 2018 Summit supercomputer at Oak Ridge National Lab, questions around how application teams are thinking about the massive amount of memory across new tiers, the swapping out of PCIe for a new interconnect, and the pervasive GPUs were first to mind. …
All Content Copyright The Next Platform