Home

ACCESS Operations API

v3.58.1

Compute Comparison

IACS
Ookami
Indiana U
Jetstream2 LM
Indiana U
Jetstream2 CPU
Indiana U
Jetstream2 GPU
NCSA
Delta CPU
NCSA
DeltaAI
NCSA
Delta GPU
NSF NCAR
Derecho-GPU
NSF NCAR
Derecho
OSG
OSPool
PSC
Neocortex SDFlex
PSC
Bridges-2 GPU
PSC
PSC Neocortex CS-2
PSC
Bridges-2 RM
PSC
Bridges-2 EM
Purdue U
Anvil AI
Purdue U
Anvil GPU
Purdue U
Anvil CPU
SDSC
Expanse CPU
SDSC
SDSC Voyager AI System
SDSC
Expanse GPU
TACC
TACC Stampede3
Texas A&M U
TAMU Launch
Texas A&M U
FASTER
Texas A&M U
ACES
Texas Tech U
REPACCS CPU
Texas Tech U
REPACSS GPU
U Kentucky
KYRIC
UD
DARWIN Compute Nodes
UD
DARWIN GPU Nodes
advance_max_reservable_su None None None None None None None None None 0.0 None None None None None None None None 512.0 None None None None None None None None None None None
advance_reservation_support False False False False False None False False False False None False None False False None False False True False False False None False False None None False False False
batch_system SLURM None None None None PBS Pro PBS Pro HTCondor SLURM (with interactive access) Slurm (with interactive access) None Slurm (with interactive access) Slurm (with interactive access) None Slurm Slurm SLURM None SLURM Slurm SLURM SLURM SLURM Slurm Slurm SLURM Slurm (with interactive access) Slurm (with interactive access)
community_software_area False False False False False None False False False False None False None False False None False False True False False False None False False None None False False False
cpu_count_per_node 48 128 128 128 128 288 64 64 128 1 None 40 None 128 96 None 128 128 128 None 40 112 32 64 96 256 64 40 64 64
cpu_speed_ghz 1.8 2.0 2.0 2.0 2.45 None 2.45 2.25 2.45 2.4 None 2.5 None 2.25 2.4 None 2.45 2.45 2.25 None 2.5 None 3.0 2.2 3.8 2.25 2.1 2.0 2.5 2.5
cpu_type Fujitsu A64FX-FX700 AMD Milan 7713 AMD Milan 7713 AMD Milan 7713 AMD Milan NVIDIA Grace AMD Milan AMD EPYC 7742 (Milan) AMD EPYC 7763 (Milan) x86_64 None Intel Xeon Gold 6248 None AMD EPYC 7742 Intel Xeon Platinum 9260M None AMD Milan AMD Milan AMD EPYC 7742 (Rome) Intel Cascade Lake Intel Xeon 6248 Sapphire Rapids, Ice Lake, Skylake AMD EPYC 9124 16-Core Intel Ice Lake Intel Sapphire Rapids AMD EPYC 9754 Intel Xeon Gold 6448Y Broadwell class AMD EPYC™ 7502
AMD EPYC™ 7502 or Intel® Xeon® Pl…
atinum 8260
disk_size_tb 788.0 None None None None None None None None 0.0 None None None None None None None None 12000.0 None None None None None None None None 300.0 None None
display_in_xsede_su_calculator True True True True True None True False False True None True None True True None True True True False True False None True True None None True True True
gpu_description None (4) NVIDIA A100 40gb GPUs per node. None NVIDIA Hopper NVIDIA A40 and A100 NVIDIA A100 (40 GB) None Various None NVIDIA Tesla V100-32GB SXM2 None None None None NVIDIA A100 None None None
Each node has four NVIDIA V100s (…
32 GB SMX2), connected via NVLINK.
Ponte Vecchio (PVC) 2 NVIDIA A30's per gpu node NVIDIA A100, A40, A30, A10, and T4
NVIDIA H100 & A30, Intel Data Cen…
ter GPU (PVC)
None NVIDIA H100 NVL None None
DARWIN contains three kinds of GP…
U nodes: NVIDIA Tesla V100s, NVIDIA Tesla T4s, and one AMD Radeon Instinct MI50. The V100 nodes hold 4 GPUs each, which can be allocated separately. The T4 and MI50 nodes each hold a single GPU.
graphics_card None None None None None None None None None None None None None None None None None None 4 GPUs per node, 128 GB HBM 2e None None None None None None None
NVIDIA Tesla V100 / NVIDIA Tesla …
T4 / AMD Radeon Instinct MI50
info_resourceid ookami.sbu.access-ci.org jetstream2-lm.indiana.access-ci.org jetstream2.indiana.access-ci.org
jetstream2-gpu.indiana.access-ci.…
org
delta-cpu.ncsa.access-ci.org deltaai.ncsa.access-ci.org delta-gpu.ncsa.access-ci.org derecho-gpu.ncar.ucar.edu derecho.ncar.ucar.edu grid1.osg.access-ci.org neocortex-sdflex.psc.access-ci.org bridges2-gpu.psc.access-ci.org neocortex.psc.access-ci.org bridges2-rm.psc.access-ci.org bridges2-em.psc.access-ci.org anvil-ai.purdue.access-ci.org anvil-gpu.purdue.access-ci.org anvil.purdue.access-ci.org expanse.sdsc.access-ci.org voyager.sdsc.access-ci.org expanse-gpu.sdsc.access-ci.org stampede3.tacc.access-ci.org launch.tamu.access-ci.org faster.tamu.access-ci.org aces.tamu.access-ci.org repacss-cpu.ttu.access-ci.org repacss-gpu.ttu.access-ci.org kyric.uky.access-ci.org darwin.udel.access-ci.org darwin-gpu.udel.access-ci.org
interconnect Mellanox HDR100 Ethernet (100GigE) Ethernet (100GigE) Ethernet (100GigE) Slingshot Slingshot Slingshot Slingshot-11 Slingshot-11 None None Dual-rail HDR-200 InfiniBand None HDR-200 InfiniBand Dual-rail HDR-200 InfiniBand None 100 Gbps Infiniband 100 Gbps Infiniband
Hybrid Fat-Tree, HDR100 InfiniBan…
d nodes, HDR200 switches
400G RoCE None Omni-Path (OPA) NVIDIA HDR100 HDR 100 InfiniBand NVIDIA NDR200 GB Nvidia ConnectX-7 Nvidia ConnectX-7 Ethernet InfiniBand InfiniBand
job_manager SLURM None slurm slurm Slurm PBS Pro PBS Pro condor SLURM (with interactive access) Slurm (with interactive access) None Slurm (with interactive access) Slurm (with interactive access) None Slurm Slurm None Kubernetes None Slurm None None None Slurm Slurm None Slurm Slurm
local_storage_per_node_gb 512.0 None None None 800.0 3500.0 1700.0 None None 10.0 None 7680.0 None 3840.0 7680.0 None None None 1024.0 6400.0 1600.0 150.0 200.0 960.0 1600.0 1920.0 480.0 6000.0 1774.0 1774.0
machine_type Cluster None None Cluster Cluster Cluster Cluster MPP MPP Cluster Cluster Cluster None Cluster Cluster None Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster Cluster
manufacturer Cray Dell Dell Dell HPE HPE HPE HPE HPE Various HPE HPE None HPE HPE None Dell Dell Dell Intel Habana Dell Dell Dell Dell Dell Dell Dell Dell Atipa Atipa
max_reservable_su None None None None None None None None None 0.0 None None None None None None None None 512.0 None None None None None None None None None None None
memory_per_cpu_gb 32.0 8.0 4.0 4.0 2.0 None None 8.0 2.0 2.0 None 12.8 None 2.0 42.6 None None 2.0 2.0 None 9.6 128.0 None 4.0 5.33 6.0 8.0 75.0 256.0 256.0
model Apollo 80 None Apollo 2000 Cray EX2500 Apollo 6500 Cray EX Cray EX None None HPE Apollo 6500 Gen 10 None HPE Apollo 2000 Gen 11 HPE ProLiant DL560 Gen10 None None None None None None None None None PowerEdge R7625 PowerEdge R760XA None Procyon QE228-16 and SE128-32
Altezza SX126-12G4SXM2 GPU and Pr…
ocyon SE123-32 and SE228-32
nfs_network NA None None None None None None None None None None None None None None HDR InfiniBand None None None None None None None 100G Ethernet 10 Gigabit Ethernet 10 Gigabit Ethernet
nickname Ookami Jetstream2 LM Jetstream2 CPU Jetstream2 GPU Delta CPU DeltaAI Delta GPU Derecho-GPU Derecho OSPool Neocortex SDFlex Bridges-2 GPU PSC Neocortex CS-2 Bridges-2 RM Bridges-2 EM Anvil AI Anvil GPU Anvil CPU Expanse CPU SDSC Voyager AI System Expanse GPU TACC Stampede3 TAMU Launch FASTER ACES REPACCS CPU REPACSS GPU KYRIC DARWIN Compute Nodes DARWIN GPU Nodes
node_count 176.0 32.0 384.0 90.0 124.0 114.0 206.0 82.0 2488.0 60000.0 None 24.0 None 504.0 4.0 None 16.0 1000.0 732.0 42.0 52.0 1848.0 45.0 180.0 110.0 110.0 8.0 5.0 92.0 13.0
operating_system Rocky Linux 8.4 User-defined per VM User-defined per VM User-defined per VM Red Hat Enterprise v8 SLES Red Hat Enterprise v8 None CentOS RHEL CentOS 8 None CentOS 8 CentOS 8 None Rocky Linux 8.4 Rocky Linux 8.4 Linux (CentOS) Ubuntu Linux (CentOS) Rocky Linux RHEL 8.10 Rocky Linux release 8.6 RHEL 8.8 Rocky Linux 9.4 Rocky Linux 9.4 Centos 8.2 CentOS 7 CentOS 7
parallel_file_system None None None None None None None None None No parallel file system None None None None None None None None None None None None None None None None None None None None
peak_teraflops 486.0 131.0 9093.0 7020.0 None None None None None 50.0 None None None None None None 1500.0 5300.0 3373.0 None 1664.0 None None None None None None None None None
platform_name Ookami, A64FX Testbed Indiana Jetstream2 Large Memory Indiana Jetstream2 CPU Indiana Jetstream2 GPU None None None Derecho-GPU Derecho Linux None HPE Apollo 6500 Gen10 None HPE Apollo 2000 Gen 11 HPE ProLiant DL560 Gen10 None
Dual AMD Milan CPUs with four NVI…
DIA A100 GPUs
Dual AMD Milan CPUs None None None
TACC Dell/Intel Sapphire Rapids, …
Ice Lake, Skylake (Stampede3)
TAMU Launch None TAMU ACES REPACSS REPACSS None None None
primary_storage_shared_gb 788 14000 14000 14000 None None None None No shared storage None 15000000 None 15000000 15000000 None None None None 324TB None 2,000 None 2,300,000 None None 400000 1001478 1001478
resource_type Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute Compute
rmax None None None None None None None 4.58 10.32 None None None None None None None None None None None None None None None None 869.0 816.0 None None None
rpeak None 131.0 9093.0 7020.0 None None None 5.04 12.4 None None None None None None None None None None None None None None None None 1014.0 1088.0 None None None
storage_network Infiniband None None None None None None None Dual-rail HDR-200 InfiniBand None HDR-200 InfiniBand Dual-rail HDR-200 InfiniBand None None None HDR InfiniBand None None VAST None HDR 100 InfiniBand NDR 200GB Infiniband None 100G Ethernet InfiniBand InfiniBand
supports_sensitive_data False False False False False None False False False False None True None True True None False False False False False False None False False None None False False False
xsedenet_participant True False False False True None True False False False None True None False True None True True True False True False None True False None None False False False