COMPUTE ATLAS Supercomputer supply-chain graph
1153 systems 482 sites Sourced data

Parts/Accelerators/NVIDIA L40S

Accelerators · Ada Lovelace · 2023

NVIDIA L40S

General-purpose data center GPU combining rendering, AI inference/training and graphics workloads on Ada Lovelace silicon.

Why this part

Used across several university and national-lab clusters in this dataset for mixed AI/visualization workloads.

Systems using this part

Systems
17
Rmax underneath
0.0 GFlop/s
Units recorded
2,014
14 of 17 edges disclose a quantity
System Role Country Units System Rmax Supplied by Rank
Aire (Leeds)
2025-02-06
Accelerator United Kingdom 84 - NVIDIA -
Insomnia (Columbia)
2024-02
Accelerator USA 18 - NVIDIA -
SCINet Atlas
2020-09-30
Accelerator USA 48 - NVIDIA -
Magerit-3
2020
Accelerator Spain 20 - NVIDIA -
farm22
Accelerator United Kingdom 40 - NVIDIA -
Killarney
Accelerator Canada 672 - NVIDIA -
Vulcan (Amii)
Accelerator Canada 820 - NVIDIA -
Jetstream2
Accelerator USA 32 - NVIDIA -
Pelle
Accelerator Sweden 40 - NVIDIA -
Imperial CX3
Accelerator United Kingdom 56 - NVIDIA -
Discovery (USC)
Accelerator USA - - NVIDIA -
Resnick HPC Cluster
Accelerator USA 16 - NVIDIA -
Katana
Accelerator Australia - - NVIDIA -
Engaging (MIT)
Accelerator USA - - NVIDIA -
Skipjack (JHU)
Accelerator USA 64 - NVIDIA -
CSF3
Accelerator United Kingdom 88 - NVIDIA -
GenomeDK
Accelerator Denmark 16 - NVIDIA -

“Units recorded” sums only the edges where a public source states a quantity. It is a floor, not a total, and should never be read as installed base.

Further reading