Morton Ordering on the Intel Xeon Phi

The Morton order is a mapping of multidimensional data to one dimension that preserves locality of the data. This is also known as Z-order. “By using Morton ordering as an alternative to row-major or column-major data storage, significant speedups can be achieved on the Intel Xeon Phi coprocessor or Intel Xeon CPU when performing matrix multiplies or matrix transposes.”

Interview: Next-Generation Cori Supercomputer Coming to NERSC

“We need to emphasize here that the Knights Landing processor is self-hosted, and so that means it’s not an accelerator. It’s not a coprocessor and the particular kernel processor that will be having for NERSC-8, will have more than 60 cores and it will have multiple hardware threads for the core. That’s a lot, right? Having 60 cores per node with multiple hardware thread. That a significant increase from both our Hopper and Edison system, which has 24 cores each. So we’re going to be working with our users to figure out what’s the right amount of parallelism that they need to expose in their application. That’s one really big difference.”

Interview: Raj Hazra and Charlie Wuischpard on Steering the Next Generation of Intel HPC

“I brought Charlie on because my philosophy is to put people that are better than me on things that are critical to the mission at hand. That is, Charlie’s focus is to run the ship hard, fast, lean and get it ready as it goes through that turn. And my job then is to work with the larger Intel organization at the strategy office up to the exec office in laying track for that new direction. And so together I think we’ve got a very good partnership to lead what has really become one of the strongest teams in HPC going forward.”