Firewalled Computing Resources Overview - Revision history

Weiler: /* The Phoenix Cluster */

2026-04-29T15:18:53Z

The Phoenix Cluster

← Older revision		Revision as of 08:18, 29 April 2026
Line 74:		Line 74:
	\| 1.7 PB of Clustered Ceph NVMe Solid State Storage		\| 1.7 PB of Clustered Ceph NVMe Solid State Storage
	\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)		\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)
	\| ~~10G~~ (nodes) / ~~200G~~+ (storage)		\| 10 Gb/s (nodes) / 200 Gb/s+ (storage)
	\|}		\|}

2026-04-29T13:59:15Z

The Phoenix Cluster

← Older revision		Revision as of 06:59, 29 April 2026
Line 74:		Line 74:
	\| 1.7 PB of Clustered Ceph NVMe Solid State Storage		\| 1.7 PB of Clustered Ceph NVMe Solid State Storage
	\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)		\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)
	\| 10G (nodes) / ~~100G~~+ (storage)		\| 10G (nodes) / 200G+ (storage)
	\|}		\|}

2026-04-29T13:47:08Z

The Phoenix Cluster

← Older revision		Revision as of 06:47, 29 April 2026
Line 56:		Line 56:
	This is a cluster of 25 Ubuntu 22.04 nodes, some of which have GPUs in them. Each node generally has about 2TB RAM and 256 cores, although the cluster is heterogeneous and has multiple node types. You interact with the Phoenix Cluster via the Slurm Job Scheduler. You must specifically request access to use Slurm on the Phoenix Cluster, just email '''cluster-admin@soe.ucsc.edu''' for access.		This is a cluster of 25 Ubuntu 22.04 nodes, some of which have GPUs in them. Each node generally has about 2TB RAM and 256 cores, although the cluster is heterogeneous and has multiple node types. You interact with the Phoenix Cluster via the Slurm Job Scheduler. You must specifically request access to use Slurm on the Phoenix Cluster, just email '''cluster-admin@soe.ucsc.edu''' for access.

	This cluster is a fairly modern cluster comprised of:		This HPC cluster is a fairly modern cluster comprised of:

	{\| class="wikitable" style="text-align:center;"		{\| class="wikitable" style="text-align:center;"

2026-04-29T13:44:53Z

The Phoenix Cluster

← Older revision		Revision as of 06:44, 29 April 2026
Line 71:		Line 71:
	\| 49.9 TB		\| 49.9 TB
	\| 16 TB (NVMe) per node, 400 TB total		\| 16 TB (NVMe) per node, 400 TB total
	\| 8 GPUs per node (A100, A5500), ~~for~~ 48 GPUs total		\| 8 GPUs per node (A100, A5500), 48 GPUs total
	\| 1.7 PB of Clustered Ceph NVMe Solid State Storage		\| 1.7 PB of Clustered Ceph NVMe Solid State Storage
	\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)		\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)

2026-04-29T13:44:18Z

The Phoenix Cluster

← Older revision		Revision as of 06:44, 29 April 2026
Line 76:		Line 76:
	\| 10G (nodes) / 100G+ (storage)		\| 10G (nodes) / 100G+ (storage)
	\|}		\|}

			Specific breakdown of node architecture:

	{\| class="wikitable" style="text-align:center;"		{\| class="wikitable" style="text-align:center;"

2026-04-29T13:43:24Z

The Phoenix Cluster

← Older revision		Revision as of 06:43, 29 April 2026
Line 70:		Line 70:
	\| 7040		\| 7040
	\| 49.9 TB		\| 49.9 TB
	\| 14 TB (NVMe) per node, ~~350~~ TB total		\| 16 TB (NVMe) per node, 400 TB total
	\| 8 GPUs per node (A100, A5500), for 48 GPUs total		\| 8 GPUs per node (A100, A5500), for 48 GPUs total
	\| 1.7 PB of Clustered Ceph NVMe Solid State Storage		\| 1.7 PB of Clustered Ceph NVMe Solid State Storage
	\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)		\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)
	\| 10G~~/40G~~/100G		\| 10G (nodes) / 100G+ (storage)
	\|}		\|}

2026-04-29T13:41:51Z

The Phoenix Cluster

← Older revision		Revision as of 06:41, 29 April 2026
Line 68:		Line 68:
	! Network Backbone		! Network Backbone
	\|-		\|-
	~~\| style="text-align:left;" \| phoenix-[22-24]~~
	~~\| style="text-align:left;" \| Ubuntu 22.04~~
	\| 7040		\| 7040
	\| 49.9 TB		\| 49.9 TB

2026-04-29T13:41:15Z

The Phoenix Cluster

← Older revision		Revision as of 06:41, 29 April 2026
Line 58:		Line 58:
	This cluster is a fairly modern cluster comprised of:		This cluster is a fairly modern cluster comprised of:

	~~'''~~Total CPU Cores~~:''' 7040~~		{\| class="wikitable" style="text-align:center;"
			\|- style="font-weight:bold;"
	~~'''~~Total Memory:~~'''~~ 49.9 TB		! Total CPU Cores
			! Total Memory
	~~'''Local Scratch Storage:'''~~ 14 TB (NVMe) per node, 350 TB total		! Local Scratch Storage
			! Six GPU Nodes
	~~'''Six GPU Nodes:'''~~ 8 GPUs per node (A100, A5500), for 48 GPUs total		! Primary Storage
			! Secondary Storage
	~~'''Primary Storage''':~~ 1.7 PB of Clustered Ceph NVMe Solid State Storage		! Network Backbone
			\|-
	~~'''Secondary Storage''':~~ 2.5 PB Spinning Disk Storage (Ceph + ZFS)		\| style="text-align:left;" \| phoenix-[22-24]
			\| style="text-align:left;" \| Ubuntu 22.04
			\| 7040
			\| 49.9 TB
			\| 14 TB (NVMe) per node, 350 TB total
			\| 8 GPUs per node (A100, A5500), for 48 GPUs total
			\| 1.7 PB of Clustered Ceph NVMe Solid State Storage
			\| 2.5 PB Spinning Disk Storage (Ceph + ZFS)
			\| 10G/40G/100G
			\|}

	{\| class="wikitable" style="text-align:center;"		{\| class="wikitable" style="text-align:center;"

2026-04-29T13:37:12Z

The Phoenix Cluster

← Older revision		Revision as of 06:37, 29 April 2026
Line 59:		Line 59:

	'''Total CPU Cores:''' 7040		'''Total CPU Cores:''' 7040

	'''Total Memory:''' 49.9 TB		'''Total Memory:''' 49.9 TB

	'''Local Scratch Storage:''' 14 TB (NVMe) per node, 350 TB total		'''Local Scratch Storage:''' 14 TB (NVMe) per node, 350 TB total

	'''Six GPU Nodes:''' 8 GPUs per node (A100, A5500), for 48 GPUs total		'''Six GPU Nodes:''' 8 GPUs per node (A100, A5500), for 48 GPUs total

	'''Primary Storage''': 1.7 PB of Clustered Ceph NVMe Solid State Storage		'''Primary Storage''': 1.7 PB of Clustered Ceph NVMe Solid State Storage

	'''Secondary Storage''': 2.5 PB Spinning Disk Storage (Ceph + ZFS)		'''Secondary Storage''': 2.5 PB Spinning Disk Storage (Ceph + ZFS)

2026-04-29T13:36:54Z

The Phoenix Cluster

← Older revision		Revision as of 06:36, 29 April 2026
Line 55:		Line 55:

	This is a cluster of 25 Ubuntu 22.04 nodes, some of which have GPUs in them. Each node generally has about 2TB RAM and 256 cores, although the cluster is heterogeneous and has multiple node types. You interact with the Phoenix Cluster via the Slurm Job Scheduler. You must specifically request access to use Slurm on the Phoenix Cluster, just email '''cluster-admin@soe.ucsc.edu''' for access.		This is a cluster of 25 Ubuntu 22.04 nodes, some of which have GPUs in them. Each node generally has about 2TB RAM and 256 cores, although the cluster is heterogeneous and has multiple node types. You interact with the Phoenix Cluster via the Slurm Job Scheduler. You must specifically request access to use Slurm on the Phoenix Cluster, just email '''cluster-admin@soe.ucsc.edu''' for access.

			This cluster is a fairly modern cluster comprised of:

			'''Total CPU Cores:''' 7040
			'''Total Memory:''' 49.9 TB
			'''Local Scratch Storage:''' 14 TB (NVMe) per node, 350 TB total
			'''Six GPU Nodes:''' 8 GPUs per node (A100, A5500), for 48 GPUs total
			'''Primary Storage''': 1.7 PB of Clustered Ceph NVMe Solid State Storage
			'''Secondary Storage''': 2.5 PB Spinning Disk Storage (Ceph + ZFS)

	{\| class="wikitable" style="text-align:center;"		{\| class="wikitable" style="text-align:center;"