After 12 years of working in the high performance computing sector as both an industry analyst and journalist, I’m delighted to join The Next Platform as a senior editor.
My new role also reunites me with some old friends, who just happen to be some of the best writers in the business.
Not surprisingly, my focus will be on HPC and all things related. These days, that covers a lot of territory since in many ways HPC has provided a useful model for other subject matter that is under the purview of The Next Platform, namely hyperscale, cloud, and high-end enterprise computing.
The applications in these domains, though different from one another, are driven by one primary challenge: how to deliver performance and throughput at scale. HPC is the only community that has been doing this for more than 50 years, and in my (admittedly biased) opinion, has been rather successful at it.
One topic I intend to write about quite a bit is the ramifications of the death of Moore’s Law. For the entire history of HPC, it’s been the most important driver of performance growth, and its slow demise is already impacting the sector profoundly.
The uncertain path to exascale computing, the explosion of hardware diversity, and the challenge of rising energy bills in the supercomputing center, can to a large extent be traced back to the deterioration of Moore’s Law. While this sounds dire, I believe we’re on the cusp of a new era of innovation in which transistor shrinkage will be supplanted with a lot more interesting set of technologies that will make the last half century of computing seem like the Dark Ages.
Naturally, I’ll be covering other critical trends in HPC as well – the intermingling of machine learning and data analytics with traditional modeling, the advance of heterogeneous computing, the adoption of the cloud delivery model, and the expansion of the high-performance technologies into new application domains, just to name a few.
I hope to bring some clarity to these topics in my writings, while making them enjoyable to read. And in the process, I intend to have some fun myself.
Michael Feldman has covered the high performance computing industry for more than 12 years, first as the editor-in-chief of HPCwire, as a market research analyst at Intersect360 Research, and as the managing editor for TOP500 News. During that time, he established himself as a trusted voice in the HPC community. Prior to his work as a journalist and industry analyst, Michael was a software engineer, a career that spanned 25 years, and included projects in compiler and runtime platforms, software development tools, and real-time embedded systems.