m3-ultramem-64 as a GKE node
64 vCPU1952 GiBx86_64Google Cloud us-central1M3 Ultra Memory 64
Where the m3-ultramem-64 goes
Default GKE setup: GKE ≤ 1.36.
CPU63.57 of 64 vCPU left for your pods
Memory1,905.81 GiB of 1952 GiB left for your pods
your podsDaemonSets (200m, 300 MiB assumed)kubelet reservedeviction
110max pods per node
$8,893.15on demand / month · $12.18/h
$4,562.11spot / month (varies)
m3-ultramem-64 GKE max pods and allocatable memory — quick answers
- How many pods can a m3-ultramem-64 run as a node with GKE ≤ 1.36? Up to 110; kubelet keeps 46999 MiB of memory and 230m CPU, leaving 1,906.10 GiB and 63.77 vCPU for pods.
- How many pods can a m3-ultramem-64 run as a node with GKE 1.37+ (Container-Optimized OS)? Up to 110; kubelet keeps 2250 MiB of memory and 230m CPU, leaving 1,949.80 GiB and 63.77 vCPU for pods.
- How many 500m / 1 GiB pods fit? 107 per node after DaemonSets; 1 nodes for 10 replicas, about $8,893.15/month on demand.
What kubelet reserves on m3-ultramem-64
| Setup | Max pods | kube-reserved CPU | Memory reserved + eviction | Allocatable CPU | Allocatable memory | Memory usable |
|---|---|---|---|---|---|---|
| GKE ≤ 1.36 tiered memory reservation | 110 | 230m | 46899 Mi + 100 Mi | 63.77 | 1,906.10 GiB | 97.6% |
| GKE 1.37+ (Container-Optimized OS) 15 MiB × max pods + 500 MiB, CPU capped at 1 vCPU | 110 | 230m | 2150 Mi + 100 Mi | 63.77 | 1,949.80 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 m3-ultramem-64
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 | $83.11 |
| 250m / 256 Mi | 107 | 107 | $83.11 |
| 250m / 512 Mi | 107 | 107 | $83.11 |
| 500m / 512 Mi | 107 | 107 | $83.11 |
| 500m / 1024 Mi | 107 | 107 | $83.11 |
| 1000m / 1024 Mi | 63 | 63 | $141.16 |
| 1000m / 2048 Mi | 63 | 63 | $141.16 |
| 2000m / 4096 Mi | 31 | 31 | $286.88 |
| 4000m / 8192 Mi | 15 | 15 | $592.88 |
Similar GKE node types
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| n1-ultramem-80 | 80 | 1922 | 79.73 | 1,876.70 GiB | 110 | $7,790.85 | $4.15 |
| m1-ultramem-80 | 80 | 1922 | 79.73 | 1,876.70 GiB | 110 | $9,187.93 | $4.90 |
| m4-ultramem-56 | 56 | 1488 | 55.79 | 1,451.38 GiB | 110 | $5,711.34 | $3.94 |
| m4n-ultramem-56 | 56 | 1488 | 55.79 | 1,451.38 GiB | 110 | $10,280.41 | $7.08 |
| m4-hypermem-64 | 64 | 992 | 63.77 | 965.30 GiB | 110 | $4,163.38 | $4.31 |
| m4n-hypermem-64 | 64 | 992 | 63.77 | 965.30 GiB | 110 | $7,494.08 | $7.76 |
| m3-megamem-64 | 64 | 976 | 63.77 | 949.62 GiB | 110 | $5,259.50 | $5.54 |
| m3-megamem-128 | 128 | 1952 | 127.61 | 1,906.10 GiB | 110 | $10,519.01 | $5.52 |
Closest match on other clouds
| Type | vCPU | GiB | Alloc CPU | Alloc memory | Max pods | $/month | $ per alloc GiB |
|---|---|---|---|---|---|---|---|
| x1e.16xlarge | 64 | 1952 | 63.77 | 1,949.14 GiB | 234 | $9,741.12 | $5.00 |
| x2iedn.16xlarge | 64 | 2048 | 63.77 | 2,039.74 GiB | 737 | $9,736.74 | $4.77 |
| m6432bds1v3 | 64 | 1397 | 63.26 | 1,391.97 GiB | 250 | $16,775.40 | $12.05 |
| m64bds1v3 | 64 | 1397 | 63.26 | 1,391.97 GiB | 250 | $16,775.40 | $12.05 |
| m96s2v3 | 96 | 1946 | 94.94 | 1,940.97 GiB | 250 | $9,088.50 | $4.68 |
| m96ds2v3 | 96 | 1946 | 94.94 | 1,940.97 GiB | 250 | $9,657.90 | $4.98 |