a3-highgpu-1g as a GKE node
26 vCPU234 GiBx86_641 GPUGoogle Cloud us-central1A3 Highgpu 1g
Where the a3-highgpu-1g goes
Default GKE setup: GKE ≤ 1.36.
CPU25.66 of 26 vCPU left for your pods
Memory222.17 GiB of 234 GiB left for your pods
your podsDaemonSets (200m, 300 MiB assumed)kubelet reservedeviction
110max pods per node
$923.23on demand / month · $1.26/h
$525.17spot / month (varies)
a3-highgpu-1g GKE max pods and allocatable memory — quick answers
- How many pods can a a3-highgpu-1g run as a node with GKE ≤ 1.36? Up to 110; kubelet keeps 11815 MiB of memory and 135m CPU, leaving 222.46 GiB and 25.86 vCPU for pods.
- How many pods can a a3-highgpu-1g run as a node with GKE 1.37+ (Container-Optimized OS)? Up to 110; kubelet keeps 2250 MiB of memory and 135m CPU, leaving 231.80 GiB and 25.86 vCPU for pods.
- How many 500m / 1 GiB pods fit? 51 per node after DaemonSets; 1 nodes for 10 replicas, about $923.23/month on demand.
What kubelet reserves on a3-highgpu-1g
| Setup | Max pods | kube-reserved CPU | Memory reserved + eviction | Allocatable CPU | Allocatable memory | Memory usable |
|---|---|---|---|---|---|---|
| GKE ≤ 1.36 tiered memory reservation | 110 | 135m | 11715 Mi + 100 Mi | 25.86 | 222.46 GiB | 95.1% |
| GKE 1.37+ (Container-Optimized OS) 15 MiB × max pods + 500 MiB, CPU capped at 1 vCPU | 110 | 135m | 2150 Mi + 100 Mi | 25.86 | 231.80 GiB | 99.1% |
How these numbers are computed: reservation formulas. Check a live node with kubectl describe node → Allocatable.
How many pods fit on one a3-highgpu-1g
Assumes DaemonSets (CNI, kube-proxy, log and metrics agents) request 200m CPU, 300 Mi and 3 pods per node. Try your own numbers in the node calculator.
| Pod requests (CPU / memory) | GKE ≤ 1.36 | GKE 1.37+ (Container-Optimized OS) | $ per pod / month |
|---|---|---|---|
| 100m / 128 Mi | 107 | 107 | $8.63 |
| 250m / 256 Mi | 102 | 102 | $9.05 |
| 250m / 512 Mi | 102 | 102 | $9.05 |
| 500m / 512 Mi | 51 | 51 | $18.10 |
| 500m / 1024 Mi | 51 | 51 | $18.10 |
| 1000m / 1024 Mi | 25 | 25 | $36.93 |
| 1000m / 2048 Mi | 25 | 25 | $36.93 |
| 2000m / 4096 Mi | 12 | 12 | $76.94 |
| 4000m / 8192 Mi | 6 | 6 | $153.87 |
Similar GKE node types
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| a2-highgpu-2g | 24 | 170 | 23.87 | 159.74 GiB | 110 | $1,079.64 | $6.76 |
| a2-ultragpu-2g | 24 | 340 | 23.87 | 326.34 GiB | 110 | $1,665.45 | $5.10 |
| g2-standard-32 | 32 | 128 | 31.85 | 118.58 GiB | 110 | $857.27 | $7.23 |
| g2-standard-48 | 48 | 192 | 47.81 | 181.30 GiB | 110 | $1,285.90 | $7.09 |
| g2-standard-24 | 24 | 96 | 23.87 | 88.50 GiB | 110 | $642.95 | $7.26 |
| a2-highgpu-4g | 48 | 340 | 47.81 | 326.34 GiB | 110 | $2,159.27 | $6.62 |
| a2-ultragpu-1g | 12 | 170 | 11.90 | 159.74 GiB | 110 | $832.72 | $5.21 |
| a3-highgpu-2g | 52 | 468 | 51.80 | 451.78 GiB | 110 | $1,846.46 | $4.09 |
Closest match on other clouds
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| g3.8xlarge | 32 | 244 | 31.85 | 240.97 GiB | 250 | $1,664.40 | $6.91 |
| p3.8xlarge | 32 | 244 | 31.85 | 241.14 GiB | 234 | $8,935.20 | $37.05 |
| gr6.8xlarge | 32 | 256 | 31.85 | 253.14 GiB | 234 | $1,785.87 | $7.05 |
| g6e.8xlarge | 32 | 256 | 31.85 | 253.14 GiB | 234 | $3,305.85 | $13.06 |
| g7e.8xlarge | 32 | 256 | 31.85 | 250.22 GiB | 506 | $3,845.82 | $15.37 |
| p5.4xlarge | 16 | 256 | 15.89 | 254.39 GiB | 118 | $5,022.40 | $19.74 |