m6id.24xlarge on Databricks
m6id.24xlarge is a general purpose AWS worker type with 96 vCPUs and 384 GB of memory. On Databricks it uses 18.24 DBUs an hour, so 1 node on Jobs Compute in US East (N. Virginia) costs $8.43 an hour: $2.74 for the DBUs and $5.70 to AWS for the machine.
AWS · General purpose · m6id
m6id.24xlarge at a glance
- vCPUs
- 96
- Memory
- 384 GB
- Own SSD
- 5,700 GB
- DBUs an hour
- 18.24
- Classic compute
- With Photon
- 52.9
- DBUs an hour, Jobs Compute
- 1 node an hour
- $8.43
- Jobs Compute, US East (N. Virginia)
1 node an hour on each kind of compute
In US East (N. Virginia) on the Premium plan. Databricks charges the DBUs, AWS the machine. A cluster pays this for every worker and for the driver.
| Compute | DBUs an hour | A DBU | Databricks | Machine | An hour |
|---|---|---|---|---|---|
| Jobs Compute | 18.24 | $0.15 | $2.74 | $5.70 | $8.43 |
| Jobs Compute + Photon | 52.9 | $0.15 | $7.93 | $5.70 | $13.63 |
| All-Purpose Compute | 18.24 | $0.55 | $10.03 | $5.70 | $15.73 |
| All-Purpose Compute + Photon | 36.48 | $0.55 | $20.06 | $5.70 | $25.76 |
| Lakeflow pipelines Core | 18.24 | $0.20 | $3.65 | $5.70 | $9.34 |
| Lakeflow pipelines Core + Photon | 52.9 | $0.20 | $10.58 | $5.70 | $16.27 |
| Lakeflow pipelines Pro | 18.24 | $0.25 | $4.56 | $5.70 | $10.26 |
| Lakeflow pipelines Pro + Photon | 52.9 | $0.25 | $13.22 | $5.70 | $18.92 |
| Lakeflow pipelines Advanced | 18.24 | $0.36 | $6.57 | $5.70 | $12.26 |
| Lakeflow pipelines Advanced + Photon | 52.9 | $0.36 | $19.04 | $5.70 | $24.74 |
In each region
1 node an hour on Jobs Compute, Premium, cheapest first.
| Region | Machine | An hour |
|---|---|---|
| US East (N. Virginia) | $5.70 | $8.43 |
| US East (Ohio) | $5.70 | $8.43 |
| US West (Oregon) | $5.70 | $8.43 |
| Asia Pacific (Mumbai) | $6.15 | $8.88 |
| Canada (Central) | $6.35 | $9.09 |
| Europe (Ireland) | $6.35 | $9.09 |
| Europe (London) | $6.60 | $9.34 |
| US West (N. California) | $6.70 | $9.44 |
| Europe (Frankfurt) | $6.85 | $9.59 |
| Asia Pacific (Seoul) | $7.01 | $9.74 |
| Asia Pacific (Singapore) | $7.11 | $9.84 |
| Asia Pacific (Sydney) | $7.16 | $9.89 |
| Asia Pacific (Tokyo) | $7.36 | $10.09 |
| South America (São Paulo) | $9.07 | $11.81 |
Other general purpose sizes
- m6id.xlarge4 vCPUs, 16 GB, 0.76 DBUs an hour
- m6id.2xlarge8 vCPUs, 32 GB, 1.52 DBUs an hour
- m6id.4xlarge16 vCPUs, 64 GB, 3.04 DBUs an hour
- m6id.8xlarge32 vCPUs, 128 GB, 6.08 DBUs an hour
- m6id.12xlarge48 vCPUs, 192 GB, 9.12 DBUs an hour
- m6id.16xlarge64 vCPUs, 256 GB, 12.16 DBUs an hour
- m6id.24xlarge96 vCPUs, 384 GB, 18.24 DBUs an hour
- m6id.32xlarge128 vCPUs, 512 GB, 24.32 DBUs an hour
Questions
- How many DBUs does m6id.24xlarge use on Databricks?
- m6id.24xlarge uses 18.24 DBUs an hour on Databricks classic compute without Photon. With Photon it uses 52.9 DBUs an hour on Jobs Compute and 36.48 on All-Purpose Compute, as Databricks' price list gives them.
- How much does m6id.24xlarge cost an hour on Databricks?
- In US East (N. Virginia), 1 m6id.24xlarge node on Jobs Compute costs $8.43 an hour: $2.74 to Databricks for 18.24 DBUs at $0.15 each, and $5.70 to AWS for the machine. On All-Purpose Compute the same node costs $15.73 an hour, and on Lakeflow pipelines Core $9.34. A cluster pays this for every worker and for the driver.
- Does Photon cost more on m6id.24xlarge?
- Yes, by the hour: with Photon, 1 m6id.24xlarge node on Jobs Compute in US East (N. Virginia) costs $13.63 an hour instead of $8.43, because it uses 52.9 DBUs an hour instead of 18.24. It saves money only when the job finishes fast enough to make up the difference.
- Which region is cheapest for m6id.24xlarge on Databricks?
- US East (N. Virginia), at $8.43 an hour for 1 node on Jobs Compute, against $11.81 in South America (São Paulo), the dearest of the 14 regions listed here that offer it. Databricks charges the same for a DBU of Jobs Compute in every AWS region listed, so the gap is the machine. Europe (Paris) and Asia Pacific (Jakarta) do not offer it.
- What is the same size as m6id.24xlarge on the other clouds?
- The nearest worker types of the same kind are Standard_D96ds_v5 on Azure and n2-standard-80 on Google Cloud. Each cloud gives its own DBUs an hour for its types, so compare what a cluster costs in the calculator, not the DBUs alone.
- What is the next size up or down from m6id.24xlarge?
- One size down is m6id.16xlarge, at 12.16 DBUs an hour; one size up is m6id.32xlarge, at 24.32. Within a family, DBUs and the machine price grow with the vCPUs, so 2 nodes of a size cost about what 1 node of twice the size costs.
Sources
Where each fact comes from
- 1Databricks: pricing calculator and instance types (the DBUs an hour of each worker type, with and without Photon)
- 2Databricks: pricing on AWS (the price of a DBU for each kind of compute and plan)
- 3AWS: Amazon EC2 on-demand pricing
- 4Databricks: compute configuration reference (the driver is the workers' type by default)
- 5Databricks: Photon
- 6AWS: m6id sizes