a3-ultragpu-8g-nolssd as a GKE node
224 vCPU2952 GiBx86_648 GPUGoogle Cloud us-central1A3 Ultragpu 8g nolssd
Where the a3-ultragpu-8g-nolssd goes
Default GKE setup: GKE ≤ 1.36.
CPU223.17 of 224 vCPU left for your pods
Memory2,885.81 GiB of 2952 GiB left for your pods
your podsDaemonSets (200m, 300 MiB assumed)kubelet reservedeviction
110max pods per node
$8,954.18on demand / month · $12.27/h
$5,092.23spot / month (varies)
a3-ultragpu-8g-nolssd GKE max pods and allocatable memory — quick answers
- How many pods can a a3-ultragpu-8g-nolssd run as a node with GKE ≤ 1.36? Up to 110; kubelet keeps 67479 MiB of memory and 630m CPU, leaving 2,886.10 GiB and 223.37 vCPU for pods.
- How many pods can a a3-ultragpu-8g-nolssd run as a node with GKE 1.37+ (Container-Optimized OS)? Up to 110; kubelet keeps 2250 MiB of memory and 630m CPU, leaving 2,949.80 GiB and 223.37 vCPU for pods.
- How many 500m / 1 GiB pods fit? 107 per node after DaemonSets; 1 nodes for 10 replicas, about $8,954.18/month on demand.
What kubelet reserves on a3-ultragpu-8g-nolssd
| Setup | Max pods | kube-reserved CPU | Memory reserved + eviction | Allocatable CPU | Allocatable memory | Memory usable |
|---|---|---|---|---|---|---|
| GKE ≤ 1.36 tiered memory reservation | 110 | 630m | 67379 Mi + 100 Mi | 223.37 | 2,886.10 GiB | 97.8% |
| GKE 1.37+ (Container-Optimized OS) 15 MiB × max pods + 500 MiB, CPU capped at 1 vCPU | 110 | 630m | 2150 Mi + 100 Mi | 223.37 | 2,949.80 GiB | 99.9% |
How these numbers are computed: reservation formulas. Check a live node with kubectl describe node → Allocatable.
How many pods fit on one a3-ultragpu-8g-nolssd
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 | $83.68 |
| 250m / 256 Mi | 107 | 107 | $83.68 |
| 250m / 512 Mi | 107 | 107 | $83.68 |
| 500m / 512 Mi | 107 | 107 | $83.68 |
| 500m / 1024 Mi | 107 | 107 | $83.68 |
| 1000m / 1024 Mi | 107 | 107 | $83.68 |
| 1000m / 2048 Mi | 107 | 107 | $83.68 |
| 2000m / 4096 Mi | 107 | 107 | $83.68 |
| 4000m / 8192 Mi | 55 | 55 | $162.80 |
Similar GKE node types
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| a3-ultragpu-8g | 224 | 2952 | 223.37 | 2,886.10 GiB | 110 | $9,914.18 | $3.44 |
| a3-edgegpu-8g-nolssd | 208 | 1872 | 207.41 | 1,827.70 GiB | 110 | $6,905.84 | $3.78 |
| a3-edgegpu-8g | 208 | 1872 | 207.41 | 1,827.70 GiB | 110 | $7,385.84 | $4.04 |
| a3-highgpu-8g | 208 | 1872 | 207.41 | 1,827.70 GiB | 110 | $7,385.84 | $4.04 |
| a3-megagpu-8g | 208 | 1872 | 207.41 | 1,827.70 GiB | 110 | $7,385.84 | $4.04 |
| a2-megagpu-16g | 96 | 1360 | 95.69 | 1,325.94 GiB | 110 | $6,421.79 | $4.84 |
| a2-ultragpu-8g | 96 | 1360 | 95.69 | 1,325.94 GiB | 110 | $6,661.79 | $5.02 |
| a3-highgpu-4g | 104 | 936 | 103.67 | 910.42 GiB | 110 | $3,692.92 | $4.06 |
Closest match on other clouds
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| p6-b300.48xlarge | 192 | 4096 | 191.45 | 4,093.53 GiB | 198 | $103,963.68 | $25.40 |
| g7e.48xlarge | 192 | 2048 | 191.45 | 2,040.86 GiB | 632 | $24,195.35 | $11.86 |
| p5.48xlarge | 192 | 2048 | 191.45 | 2,046.58 GiB | 100 | $40,179.20 | $19.63 |
| p5en.48xlarge | 192 | 2048 | 191.45 | 2,045.53 GiB | 198 | $46,206.08 | $22.59 |
| p6-b200.48xlarge | 192 | 2048 | 191.45 | 2,045.53 GiB | 198 | $83,170.94 | $40.66 |
| g6e.48xlarge | 192 | 1536 | 191.45 | 1,530.37 GiB | 492 | $21,995.76 | $14.37 |