a2-megagpu-16g as a GKE node
96 vCPU1360 GiBx86_6416 GPUGoogle Cloud us-central1A2 Megagpu 16g
Where the a2-megagpu-16g goes
Default GKE setup: GKE ≤ 1.36.
CPU95.49 of 96 vCPU left for your pods
Memory1,325.65 GiB of 1360 GiB left for your pods
your podsDaemonSets (200m, 300 MiB assumed)kubelet reservedeviction
110max pods per node
$6,421.79on demand / month · $8.80/h
$3,853.12spot / month (varies)
a2-megagpu-16g GKE max pods and allocatable memory — quick answers
- How many pods can a a2-megagpu-16g run as a node with GKE ≤ 1.36? Up to 110; kubelet keeps 34875 MiB of memory and 310m CPU, leaving 1,325.94 GiB and 95.69 vCPU for pods.
- How many pods can a a2-megagpu-16g run as a node with GKE 1.37+ (Container-Optimized OS)? Up to 110; kubelet keeps 2250 MiB of memory and 310m CPU, leaving 1,357.80 GiB and 95.69 vCPU for pods.
- How many 500m / 1 GiB pods fit? 107 per node after DaemonSets; 1 nodes for 10 replicas, about $6,421.79/month on demand.
What kubelet reserves on a2-megagpu-16g
| Setup | Max pods | kube-reserved CPU | Memory reserved + eviction | Allocatable CPU | Allocatable memory | Memory usable |
|---|---|---|---|---|---|---|
| GKE ≤ 1.36 tiered memory reservation | 110 | 310m | 34775 Mi + 100 Mi | 95.69 | 1,325.94 GiB | 97.5% |
| GKE 1.37+ (Container-Optimized OS) 15 MiB × max pods + 500 MiB, CPU capped at 1 vCPU | 110 | 310m | 2150 Mi + 100 Mi | 95.69 | 1,357.80 GiB | 99.8% |
How these numbers are computed: reservation formulas. Check a live node with kubectl describe node → Allocatable.
How many pods fit on one a2-megagpu-16g
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 | $60.02 |
| 250m / 256 Mi | 107 | 107 | $60.02 |
| 250m / 512 Mi | 107 | 107 | $60.02 |
| 500m / 512 Mi | 107 | 107 | $60.02 |
| 500m / 1024 Mi | 107 | 107 | $60.02 |
| 1000m / 1024 Mi | 95 | 95 | $67.60 |
| 1000m / 2048 Mi | 95 | 95 | $67.60 |
| 2000m / 4096 Mi | 47 | 47 | $136.63 |
| 4000m / 8192 Mi | 23 | 23 | $279.21 |
Similar GKE node types
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| 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 |
| a2-highgpu-8g | 96 | 680 | 95.69 | 659.54 GiB | 110 | $4,318.55 | $6.55 |
| 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 |
| g2-standard-96 | 96 | 384 | 95.69 | 369.46 GiB | 110 | $2,571.80 | $6.96 |
Closest match on other clouds
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| p4d.24xlarge | 96 | 1152 | 95.69 | 1,143.74 GiB | 737 | $16,029.08 | $14.01 |
| p4de.24xlarge | 96 | 1152 | 95.69 | 1,143.74 GiB | 737 | $20,036.35 | $17.52 |
| g7e.24xlarge | 96 | 1024 | 95.69 | 1,016.86 GiB | 632 | $12,097.68 | $11.90 |
| g6e.24xlarge | 96 | 768 | 95.69 | 762.37 GiB | 492 | $10,997.88 | $14.43 |
| p3dn.24xlarge | 96 | 768 | 95.69 | 759.74 GiB | 737 | $22,784.76 | $29.99 |
| g6e.48xlarge | 192 | 1536 | 191.45 | 1,530.37 GiB | 492 | $21,995.76 | $14.37 |