Sanjaya Lohani, PhD - Machines/SoC

Considering joining the group ..

Machines/SoCs

Overview

GPU Access: NVIDIA DGX A100, NVIDIA CLARA AGX (RTX 6000), RTX 5000 ada, RTX 3080

CPU Access: Intel Xeon w9 (56 cores), AMD EPYC 7763 (128 cores), Intel Xeon x5 (Dual 20 cores), Intel Xeon x3 (Dual 16 cores), Intel Xeon x8 (Dual 14 cores), Intel Xeon x7 (Dual 12 cores), Carmel ARM v8.2 (8 cores)

FPGA: AMD Ultrascale+ MPSoC ZCU102

Lab's Computing Machines/SoCs

AI and Virtual Quantum Computing Workstation

Processor: Intel® Xeon® w9-3495X (105 MB cache, 56 cores, 112 threads, up to 4.8 GHz Turbo, 350 W).
GPU: Dual NVIDIA® RTXTM 5000 Ada Generation (x2), 32 GB GDDR6, 4 DP.
NVIDIA RTX 5000 ada.
Memory: 128 GB: 8 x 16 GB, DDR5, 4800 MT/s, RDIMM, ECC.
Storage: 2 TB M.2, 6 TB SATA.

Image from Nvidia

NVIDIA Clara AGX

CPU: 8-core Carmel ARM v8.2 64-bit CPU, 8MB L2 + 4MB L3.

Memory: 32GB 256-Bit LPDDR4x, 136.5GB/s.

GPU: RTX 6000, 24 GB GDDR6, 672 GB/s, 4,608 CUDA cores, 576 Tensor Cores.

Expansion: PCIe x8 (single slot width) to Xavier, PCIe x8 (single slot width) to RTX6000.

Image from Xilinx

AMD Zynq UltraScale+™ MPSoC ZCU102

DDR4 SODIMM – 4GB 64-bit w/ ECC attached to processing system (PS).

DDR4 Component – 512MB 16-bit attached to programmable logic (PL).

Port: PCIe® Root Port Gen2 x4, USB3, Display Port, and SATA. 4x SFP+ interfaces for Ethernet

2x FPGA Mezzanine Card (FMC).

I/O interfaces: includes 16 16.3Gb/s GTH transceivers and 64 user-defined differential I/O signals.

SMU Hardware (Total access)

NVIDIA DGX SuperPOD

Computational Ability: 1,644 TFLOPS.
Number of Nodes: 20 NVIDIA DGX A100.

CPU Cores: 2,560

GPU Accelerator Cores: 1,392,640

Total Memory: 52.5 TB

Useful Links:
https://www.smu.edu/oit/services/superpod
https://southernmethodistuniversity.github.io/hpc_docs/

M3 HPC Cluster

Nodes: 170

Processors: AMD EPYC 7763

CPUs/Node: 2

Cores/Node: 128

Memory/Node: 512 GB

Useful Links:
https://www.smu.edu/oit/services/m3

S2 Server (GENUSE Linux Server)

Processors:

Dual 20 core Intel Xeon (x5)
Dual 16 core Intel Xeon (x3)

Dual 14 core Intel Xeon (x8)

Dual 12 core Intel Xeon (x7)

Useful Links:
https://www.smu.edu/oit/services/s2
https://www.smu.edu/oit/services/genuse/computers

Google Sites

Report abuse