NASA has taken a major leap in high-performance computing with the launch of Athena, its most powerful and efficient supercomputer to date. Officially announced in late January 2026 and made fully available to users starting January 14, 2026, Athena represents a significant upgrade in the agency’s High-End Computing Capability (HECC) program. Housed at NASA’s Ames Research Center in California’s Silicon Valley, this cutting-edge system is already supporting critical missions, including preparations for Artemis II and beyond.
A New Era at Ames Research Center
Athena resides in the Modular Supercomputing Facility (MSF), a purpose-built site located less than a mile from the main NASA Advanced Supercomputing (NAS) building. This strategic placement enables seamless integration with existing infrastructure while allowing for modular scalability.
The system was developed by HPE using Cray EX4000 architecture and features 1,024 compute nodes powered by AMD EPYC “Turin” processors (specifically the 9745 model with 128 cores per socket). Each node delivers dual processors for a total of 256 physical cores, resulting in an impressive 262,144 cores across the entire cluster. With 768 GB of DDR5 memory per node (3 GB per core), the system boasts a total memory capacity of 786 TB.
Athena achieves a theoretical peak performance of 20.132 petaflops in double-precision floating-point operations—translating to over 20 quadrillion calculations per second. This raw power surpasses previous NASA flagships like Aitken and the now-decommissioned Pleiades, while delivering superior energy efficiency and lower operational costs.
The interconnect fabric relies on Cray Slingshot 11 technology, providing 200 Gbit/s per interface (400 Gbit/s per node total) in a Dragonfly topology that ensures low-latency, all-to-all communication within and between node groups—essential for large-scale parallel simulations.
From Beta to Full Production
After a successful beta testing phase with select users, Athena transitioned to full production access on January 14, 2026. Jobs are submitted via PBS Pro from Athena front-end systems (ATHFEs) to dedicated queues: devel, normal, long, and wide. The current Standard Billing Unit (SBU) rate stands at 12.19, reflecting optimized resource allocation.
NASA encourages users to review detailed configuration guides and attend training webinars, such as the “Introduction to the New Athena Cluster,” for best practices in porting code, optimizing performance, and managing allocations.
Driving Mission-Critical Science
Athena’s immense capabilities directly support NASA’s core objectives:
- Rocket and spacecraft simulations — enabling precise modeling of launch trajectories, aerodynamics, and re-entry dynamics for Artemis missions.
- Aeronautics research — designing fuel-efficient, next-generation aircraft and advancing sustainable aviation technologies.
- AI and data-intensive science — training large-scale foundation models to analyze vast datasets, uncover patterns in Earth climate data, exoplanet atmospheres, astrophysics observations, and more.
- Hybrid computing integration — complementing on-premises resources with cloud-based platforms for flexible, mission-tailored workflows.
By handling tasks that would take a standard desktop computer centuries in mere days, Athena accelerates discovery and reduces development timelines for ambitious projects like lunar landings, Mars exploration, and deep-space science.
Efficiency and Sustainability in Focus
Beyond sheer speed, Athena emphasizes smarter computing. Its modular design minimizes energy waste, lowers utility costs, and aligns with NASA’s broader sustainability goals in an era of exploding data volumes and AI demands.
The Road Ahead
As part of the ongoing HECC portfolio, Athena sets the stage for future expansions in exascale-class computing and hybrid architectures. With missions growing more complex, systems like this will remain indispensable for turning ambitious visions into achievable realities.
In the words of NASA’s announcement: Athena isn’t just hardware—it’s the computational foundation for a new generation of exploration, helping humanity push farther into the cosmos while solving pressing challenges here on Earth.
