GPU Specifications

Host Details

• 19g (2019 GeForce RTX 2080) -- 32 nodes, octa RTX 2080 GPU, 24 Intel(R) Xeon(R) Gold 5118 cores, 196 GB memory, Omni-Path fabric (100 Gb/s), 1TB disk.

• 12k (2012 Kepler) -- 46 nodes, 16 Intel 2.0 GHz cores, 128 GB memory, quad GPU (Kepler K20m), FDR (56 Gb/s) Infiniband, 1 TB disk.

GPU cards

Tesla cards (w/ ECC memory)

• K20m -- Tesla Kepler card, GK110 GPU, 13 SMX processors, 2496 stream processors @ 706 MHz, 5 memory processors, 5 GB memory (4.4 with ECC enabled), 1.17 TFlops peak double precision, 3.5 single precision, 145 GB/s memory bandwidth (stream triad, ECC on).
• C2050 -- Tesla Fermi card, 448 stream processors @ 1.15 GHz, 3 GB memory (2.625 GB with ECC on), 1.03 TFlops single precision, 100 GB/s memory bandwidth (stream triad, ECC on).

RTX cards (no ECC)

• RTX-2080 -- Turing GPU architecture, each has 2944 cores with 1515 MHz base clock speed, 8 GB GDDR6 memory, 448 GB/s memory bandwidth.

Since the gaming cards do not have ECC memory, they should only be used for matrix inversions where a quick test of correctness at the end of the inversion is possible.