This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| Theta (supercomputer) | |
|---|---|
| Name | Theta |
| Manufacturer | Cray |
| Operator | Argonne National Laboratory |
| Location | Argonne, Illinois |
| Architecture | Intel Xeon Phi (Knights Landing) |
| Memory | 2816 GB |
| Storage | 11 PB |
| Peak | 11.69 PFLOPS (double) |
| Purpose | Scientific computing |
| Dates | 2017–2020s |
Theta (supercomputer) was a high-performance computing system deployed at Argonne National Laboratory and managed by the Argonne Leadership Computing Facility for scientific simulation, data analysis, and machine learning. It served as one of the flagship platforms in the CORAL procurement alongside systems at Oak Ridge National Laboratory and Lawrence Livermore National Laboratory, providing resources to projects funded by the U.S. Department of Energy, the National Nuclear Security Administration, and collaborating universities such as the University of Chicago and University of Illinois Urbana–Champaign. Theta emphasized manycore processor parallelism and accelerator-style workflows used in computational chemistry, climate modeling, and astrophysics.
Theta was a production supercomputer based on Intel's Xeon Phi Knights Landing manycore processors and delivered to Argonne National Laboratory through a contract with Cray Inc. and Intel. It targeted capacity-class workloads and supported campaigns from agencies including the Office of Science (United States Department of Energy) and partnerships with centers like the Leadership Computing Facility and the National Energy Research Scientific Computing Center. The system complemented capability machines such as Summit and Sierra and fit into national strategies articulated by the Exascale Computing Project and the National Strategic Computing Initiative.
Theta's nodes combined manycore Intel Xeon Phi processors with high-speed interconnects from Cray Aries network topology and a Lustre-based parallel file system from vendors used by facilities like Oak Ridge Leadership Computing Facility. Each compute node housed a 64-core Knights Landing processor, multiple gigabytes of high-bandwidth MCDRAM and DDR4 memory, and PCIe fabric similar to designs in systems from Hewlett Packard Enterprise and Dell EMC. The system topology used a dragonfly-inspired layout akin to interconnects in machines such as Piz Daint and employed service nodes for provisioning resembling architectures at NERSC.
Theta ran a Linux-based operating system consistent with HPC environments at Oak Ridge National Laboratory and supported programming models including MPI, OpenMP, and offload paradigms promoted by Intel tooling such as the Intel compiler suite and libraries. The software stack integrated workflow managers used by projects at Lawrence Berkeley National Laboratory and container technologies similar to efforts at Argonne for reproducible research with tools comparable to Singularity and Spack. Debugging and performance analysis used toolchains related to offerings from Allinea/Arm Forge and Intel VTune, similar to practices at Los Alamos National Laboratory.
Theta achieved peak theoretical performance measured in petaflops and was benchmarked on suites familiar to DOE centers, including runs of the High Performance LINPACK (HPL) and scientific workloads akin to those used on Titan and Mira. Benchmark results influenced procurement comparisons across systems at Oak Ridge and Lawrence Livermore, and were included in studies by the Top500 and Green500 projects that profile sustained performance and energy efficiency for supercomputers like Sequoia and Sierra.
Theta was installed at the Argonne Leadership Computing Facility located on the Argonne National Laboratory campus within the Greater Chicago science corridor, fitting into data center infrastructure similar to deployments at Brookhaven National Laboratory and Los Alamos National Laboratory. The facility provided chilled water systems, raised-floor power distribution, and redundancy modeled after designs by engineering teams that worked on sites such as Lawrence Livermore National Laboratory and Sandia National Laboratories. Site commissioning included collaboration with Cray engineers and Argonne computing staff who previously supported systems like Mira.
Theta supported multidisciplinary science including projects in computational fluid dynamics used by collaborators at the University of Michigan, climate simulations related to work at the National Center for Atmospheric Research, materials design connected to Northwestern University and MIT, and cosmological simulations tied to teams at Princeton University and Caltech. It enabled machine learning workloads in partnership with researchers at Argonne and corporate partners similar to collaborations with NVIDIA and supported data analytics pipelines like those developed for experiments at Fermilab and the Large Hadron Collider research community.
Theta originated from the CORAL procurement process that sourced next-generation systems for DOE laboratories, following earlier Argonne systems such as Mira and aligning with national compute roadmaps from the Exascale Computing Project. Contracting and delivery involved Cray Inc. (later acquired by Hewlett Packard Enterprise) and Intel, with installation milestones in 2017 and early science runs through partnerships with institutions such as the University of Chicago and federal research programs like the Advanced Scientific Computing Research office.
Over time Theta was succeeded by capability systems and exascale preparations at DOE labs, handing off workflows to successors at Argonne and other facilities including future machines developed under the Exascale Computing Project and procurement efforts by DOE Office of Science. Its legacy includes contributions to software portability, manycore optimization strategies shared with projects at Sandia National Laboratories and broader impacts on HPC curriculum at universities such as University of Illinois Urbana–Champaign and University of Chicago.