NASA’s HPC infrastructure is currently in need of serious modernization as it cannot meet the needs of organizations within the US National Aeronautics and Space Administration. This conclusion was reached during an audit by the Office of Inspector General, The Register reported.

It is emphasized that NASA’s LDC infrastructure is morally outdated and cannot effectively support modern workloads. For example, NASA’s Advanced Supercomputing Center uses 18,000 CPUs and only 48 GPU-based accelerators.

On the other hand, available computing power is lacking for all consumers. Therefore, some departments of NASA and science centers are forced to purchase their own equipment and create a local LDC infrastructure. Notably, the Space Launch System team alone spends $250,000 a year for these purposes rather than connecting to a central system. With the exception of Goddard Space Flight Center and Stennis Space Center, nearly every NASA structural unit has its own independent computing infrastructure.

Another reason for deploying in-house talent is confusion around cloud resources and NASA policy, which makes planning and estimating financial costs difficult. The audit also revealed security problems at NASA’s supercomputer park. For example, some systems accessed by foreign users cannot be properly monitored.

The generally observed picture leads to delays and additional costs in the implementation of space missions. To eliminate the shortcomings, it is recommended that NASA management undertake a comprehensive reform of the LDC sector, which includes inventory of assets, identification of technological gaps and cyber risks. It is also necessary to develop a strategy to improve the allocation of available computing power.