HACC Cluster Node Status:
Active Nodes on HACC Cluster and Associated Compute Resources
Node Cores/Threads Memory Accelerators Status External Use Node Type
head-000-4 2/4 4 GB N/A UP Y General Compute
head-000-5 36/72 368 GB N/A UP Y Development + General Compute
login-0-01 1/2 2 GB N/A UP Y Login Node
login-1-01 1/2 2 GB N/A UP Y Login Node
login-2-01 1/2 2 GB N/A UP Y Login Node
web-0-1 1/1 4 GB N/A UP Y Web Login Node
c-000-2 4/8 40 GB Alveo U250 UP Y FPGA + Compute
c-000-3 4/8 40 GB Alveo U250 UP Y FPGA + Compute
c-000-4 4/8 40 GB Alveo U250 DOWN/MAINTENANCE Y FPGA + Compute
c-001-2 4/8 40 GB Alveo U280 UP Y FPGA + Compute
c-002 32/64 192 GB N/A DOWN N FPGA + Compute
c-003 32/64 192 GB 1 x U280 + Dev Boards + 2 x U250 UP/Reserved N FPGA + Compute
c-004 40/80 128 GB 1 Alveo U250 + 2x Tesla V100 UP/Reserved N FPGA + GPU + Compute
HACC Cluster FPGA Status:
List of current publicly available FPGAs denoting the the FPGA name, the installed shell, and the VM it is currently passed through to. Note that the VM name shows which physical machine the FPGA is connected to, which can be used to determine what type of access is available on the FPGA i.e. if it is connected to the 100G switch.
FPGA Name Shell Installed Virtual Machine Name
U250-002 xilinx_u250_gen3x16_xdma_4_1_202210_1 xacc-comp-000-3
U250-003 xilinx_u250_gen3x16_xdma_4_1_202210_1 xacc-comp-000-2
U250-004 xilinx_u250_gen3x16_xdma_4_1_202210_1 xacc-comp-000-4
U280-002 xilinx_u280_gen3x16_xdma_1_202211_1 xacc-comp-001-3 or xacc-comp-001-4 (mutually exclusive)
U280-003 xilinx_u280_gen3x16_xdma_1_202211_1 xacc-comp-001-2 or xacc-comp-0001-4 (mutually exclusive)
U280-000 xilinx_u280_gen3x16_xdma_1_202211_1 xacc-comp-002-4
HACC mlCluster Node Status:
Active Nodes on HACC mlCluster and Associated Compute Resources
| Node | Cores | Memory | Accelerators | Status | External Use? | Node Type |
|---|---|---|---|---|---|---|
| Development VM | 90 | 220 GB | - | UP | Y | Login/Development |
| mlcluster 1 | 128 | 500 GB | 1x VCK5000, 1x U55C, 4x MI210 GPU | UP | Y(via Kubernetes) | Worker |
| mlcluster 2 | 128 | 500 GB | 1x VCK5000, 2x U55C, 4x MI210 GPU | UP | Y(via Kubernetes) | Worker |
| mlcluster 3 | 128 | 500 GB | 1x VCK5000, 2x U55C, 4x MI210 GPU | UP | Y(via Kubernetes) | Worker |