Supported CPU machines for Cluster Director

This document describes the CPU-based Compute Engine machines that are available in Cluster Director. You can use these general-purpose machines to run high performance computing (HPC) workloads and cluster management roles.

C4A machine series

Cluster Director supports the C4A machine series. This machine series runs on Arm-based Google Axion processors and delivers consistent performance for Arm-based workloads. You must pair login nodes that use the C4A machine series with compute nodes that use the A4X Max or A4X machine series.

The machine types in the C4A series differ by the amount of DDR5 memory per vCPU. These memory configurations are as follows:

standard: 4 GB of memory per vCPU
highmem: 8 GB of memory per vCPU
highcpu: 2 GB of memory per vCPU

Each family offers a range of machine sizes, from 1 to 72 vCPUs. You can't use machine types that have Local SSD disks attached (-lssd) or run on bare metal (-metal) for login nodes. C4A instances support Hyperdisk Balanced or Hyperdisk Extreme boot disks.

The following table summarizes the C4A machine types that are available in Cluster Director:

C4A standard

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`c4a-standard-1`	1	4	Up to 10	N/A
`c4a-standard-2`	2	8	Up to 10	N/A
`c4a-standard-4`	4	16	Up to 23	N/A
`c4a-standard-8`	8	32	Up to 23	N/A
`c4a-standard-16`	16	64	Up to 23	N/A
`c4a-standard-32`	32	128	Up to 23	Up to 50
`c4a-standard-48`	48	192	Up to 34	Up to 50
`c4a-standard-64`	64	256	Up to 45	Up to 75
`c4a-standard-72`	72	288	Up to 50	Up to 100

^* A vCPU represents an entire physical core because simultaneous multithreading (SMT) isn't supported. For more information, see CPU platforms.
^‡ Maximum egress bandwidth can't exceed the number given. Actual egress bandwidth depends on the destination IP address and other factors. For more information, see Network bandwidth.
^# Supports high-bandwidth networking for larger machine types.

C4A high-mem

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`c4a-highmem-1`	1	8	Up to 10	N/A
`c4a-highmem-2`	2	16	Up to 10	N/A
`c4a-highmem-4`	4	32	Up to 23	N/A
`c4a-highmem-8`	8	64	Up to 23	N/A
`c4a-highmem-16`	16	128	Up to 23	N/A
`c4a-highmem-32`	32	256	Up to 23	Up to 50
`c4a-highmem-48`	48	384	Up to 34	Up to 50
`c4a-highmem-64`	64	512	Up to 45	Up to 75
`c4a-highmem-72`	72	576	Up to 50	Up to 100

C4A high-cpu

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`c4a-highcpu-1`	1	2	Up to 10	N/A
`c4a-highcpu-2`	2	4	Up to 10	N/A
`c4a-highcpu-4`	4	8	Up to 23	N/A
`c4a-highcpu-8`	8	16	Up to 23	N/A
`c4a-highcpu-16`	16	32	Up to 23	N/A
`c4a-highcpu-32`	32	64	Up to 23	Up to 50
`c4a-highcpu-48`	48	96	Up to 34	Up to 50
`c4a-highcpu-64`	64	128	Up to 45	Up to 75
`c4a-highcpu-72`	72	144	Up to 50	Up to 100

N2 machine series

Cluster Director supports the N2 machine series. This series offers a balance of price and performance suitable for a variety of workloads, such as CPU-bound computational tasks or general-purpose HPC workloads. You can use the N2 machine series for compute nodes or login nodes. N2 login nodes support clusters with any supported GPU, TPU, or CPU compute nodes, including A4X Max, A4X, A4, A3 Ultra, A3 Mega, TPU7x, TPU v6e, TPU v5p, and N2.

The N2 machine types differ by the amount of memory configured per vCPU. These memory configurations are as follows:

standard: 4 GB of memory per vCPU
highmem: 8 GB of memory per vCPU
highcpu: 1 GB of memory per vCPU

Each family offers a range of machine sizes, from 2 to 128 vCPUs. For login nodes, Google recommends that you use N2 standard machine types with 32 or fewer vCPUs to avoid unnecessary charges for compute capacity that you don't need. N2 instances support Persistent Disk, Hyperdisk Extreme, or Hyperdisk Throughput boot disks.

The following table summarizes the available N2 machine types in Cluster Director:

N2 standard

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`n2-standard-2`	2	8	Up to 10	N/A
`n2-standard-4`	4	16	Up to 10	N/A
`n2-standard-8`	8	32	Up to 16	N/A
`n2-standard-16`	16	64	Up to 32	N/A
`n2-standard-32`	32	128	Up to 32	Up to 50
`n2-standard-48`	48	192	Up to 32	Up to 50
`n2-standard-64`	64	256	Up to 32	Up to 75
`n2-standard-80`	80	320	Up to 32	Up to 100
`n2-standard-96`	96	384	Up to 32	Up to 100
`n2-standard-128`	128	512	Up to 32	Up to 100

^* A vCPU is implemented as a single hardware thread, or logical core, on one of the available CPU platforms.
^‡ Maximum egress bandwidth can't exceed the number given. Actual egress bandwidth depends on the destination IP address and other factors. For more information, see Network bandwidth.
^# Supports high-bandwidth networking for larger machine types. For Windows OS images, the maximum network bandwidth is limited to 50 Gbps.

N2 high-mem

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`n2-highmem-2`	2	16	Up to 10	N/A
`n2-highmem-4`	4	32	Up to 10	N/A
`n2-highmem-8`	8	64	Up to 16	N/A
`n2-highmem-16`	16	128	Up to 32	N/A
`n2-highmem-32`	32	256	Up to 32	Up to 50
`n2-highmem-48`	48	384	Up to 32	Up to 50
`n2-highmem-64`	64	512	Up to 32	Up to 75
`n2-highmem-80`	80	640	Up to 32	Up to 100
`n2-highmem-96`	96	768	Up to 32	Up to 100
`n2-highmem-128`	128	864	Up to 32	Up to 100

N2 high-cpu

Machine types	vCPUs^*	Memory (GB)	Default egress bandwidth (Gbps)^‡	Tier_1 egress bandwidth (Gbps)^#
`n2-highcpu-2`	2	2	Up to 10	N/A
`n2-highcpu-4`	4	4	Up to 10	N/A
`n2-highcpu-8`	8	8	Up to 16	N/A
`n2-highcpu-16`	16	16	Up to 32	N/A
`n2-highcpu-32`	32	32	Up to 32	Up to 50
`n2-highcpu-48`	48	48	Up to 32	Up to 50
`n2-highcpu-64`	64	64	Up to 32	Up to 75
`n2-highcpu-80`	80	80	Up to 32	Up to 100
`n2-highcpu-96`	96	96	Up to 32	Up to 100