The National Aeronautics and Space Administration (NASA) and Hewlett Packard Enterprise (HPE) have joined together in the creation of a groundbreaking supercomputer. The state of the art computational monster will be put to use in NASA’s Ames Research Center. It will be used to create models and simulations of landing procedures for the Artemis Moon missions which are projected for the year 2024.
The supercomputer known as Aitken – named after the American astronomer, Robert Grant Aitken, who specialized in the study of double stars – has the ability to run simulations at a computing speed of 3.69 petaFLOPs (1015 floating-point operations per second). The NASA-HPE supercomputer is custom-made to work with the Ames modular data center, a project it began in 2017 as an effort to significantly decrease the quantity of water and energy used to keep its supercomputing hardware at a functional temperature.
Aitken boasts second-gen Intel Xeon processors, Mellanox InfiniBand high-speed networking, and over 220 terabytes of on-board memory. It’s the product of a four year vision between HPE and NASA that is now coming to life. Aitken will model varying entry possibilities, descent and landing for the Artemis Moon missions, and simulations to assess potential results and assist in deciding an optimized approach.
This is not the sole project that NASA and HPE have been working on together. Another supercomputer was made to endure the harshness of space and was sent to the International Space Station in 2017 for testing before its possible utilization on longer missions.
HPE also revealed near the start of this year that it was acquiring Cray for $1.3 billion. Cray Inc. is an American supercomputer manufacturing company headquartered in Seattle, Washington, and has been partnered with NASA for some time. Their relationship goes back to NASA’s founding of a dedicated computational modelling department, as well as the creation of their Central Computing Facility at Ames Research Center.