Standard_D32ds_v5 on Databricks
Standard_D32ds_v5 is a general purpose Azure worker type with 32 vCPUs and 128 GB of memory. On Databricks it uses 8 DBUs an hour, so 1 node on Jobs Compute in East US costs $4.21 an hour: $2.40 for the DBUs and $1.81 to Azure for the machine.
Azure · General purpose · Ddsv5
Standard_D32ds_v5 at a glance
- vCPUs
- 32
- Memory
- 128 GB
- Own SSD
- 1,200 GB
- DBUs an hour
- 8
- Classic compute
- With Photon
- 20
- DBUs an hour, Jobs Compute
- 1 node an hour
- $4.21
- Jobs Compute, East US
1 node an hour on each kind of compute
In East US on the Premium plan. Databricks charges the DBUs, Azure 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 | 8 | $0.30 | $2.40 | $1.81 | $4.21 |
| Jobs Compute + Photon | 20 | $0.30 | $6.00 | $1.81 | $7.81 |
| All-Purpose Compute | 8 | $0.55 | $4.40 | $1.81 | $6.21 |
| All-Purpose Compute + Photon | 16 | $0.55 | $8.80 | $1.81 | $10.61 |
| Lakeflow pipelines Core | 8 | $0.30 | $2.40 | $1.81 | $4.21 |
| Lakeflow pipelines Pro | 8 | $0.38 | $3.04 | $1.81 | $4.85 |
| Lakeflow pipelines Advanced | 8 | $0.54 | $4.32 | $1.81 | $6.13 |
In each region
1 node an hour on Jobs Compute, Premium, cheapest first.
| Region | Machine | An hour |
|---|---|---|
| East US | $1.81 | $4.21 |
| East US 2 | $1.81 | $4.21 |
| West US 2 | $1.81 | $4.21 |
| West US 3 | $1.81 | $4.21 |
| Sweden Central | $1.92 | $4.32 |
| Central India | $1.95 | $4.35 |
| Canada Central | $2.02 | $4.42 |
| North Europe | $2.02 | $4.42 |
| Central US | $2.04 | $4.44 |
| UK South | $2.10 | $4.50 |
| France Central | $2.11 | $4.51 |
| West Europe | $2.18 | $4.58 |
| Germany West Central | $2.18 | $4.58 |
| Southeast Asia | $2.26 | $4.66 |
| Australia East | $2.27 | $4.67 |
| Japan East | $2.34 | $4.74 |
| Brazil South | $2.88 | $5.28 |
Other general purpose sizes
- Standard_D4ds_v54 vCPUs, 16 GB, 1 DBUs an hour
- Standard_D8ds_v58 vCPUs, 32 GB, 2 DBUs an hour
- Standard_D16ds_v516 vCPUs, 64 GB, 4 DBUs an hour
- Standard_D32ds_v532 vCPUs, 128 GB, 8 DBUs an hour
- Standard_D48ds_v548 vCPUs, 192 GB, 12 DBUs an hour
- Standard_D64ds_v564 vCPUs, 256 GB, 16 DBUs an hour
- Standard_D96ds_v596 vCPUs, 384 GB, 24 DBUs an hour
Questions
- How many DBUs does Standard_D32ds_v5 use on Databricks?
- Standard_D32ds_v5 uses 8 DBUs an hour on Databricks classic compute without Photon. With Photon it uses 20 DBUs an hour on Jobs Compute and 16 on All-Purpose Compute, as Databricks' price list gives them.
- How much does Standard_D32ds_v5 cost an hour on Databricks?
- In East US, 1 Standard_D32ds_v5 node on Jobs Compute costs $4.21 an hour: $2.40 to Databricks for 8 DBUs at $0.30 each, and $1.81 to Azure for the machine. On All-Purpose Compute the same node costs $6.21 an hour, and on Lakeflow pipelines Core $4.21. A cluster pays this for every worker and for the driver.
- Does Photon cost more on Standard_D32ds_v5?
- Yes, by the hour: with Photon, 1 Standard_D32ds_v5 node on Jobs Compute in East US costs $7.81 an hour instead of $4.21, because it uses 20 DBUs an hour instead of 8. It saves money only when the job finishes fast enough to make up the difference.
- Which region is cheapest for Standard_D32ds_v5 on Databricks?
- East US, at $4.21 an hour for 1 node on Jobs Compute, against $5.28 in Brazil South, the dearest of the 17 regions listed here that offer it. Databricks charges the same for a DBU of Jobs Compute in every Azure region listed, so the gap is the machine.
- What is the same size as Standard_D32ds_v5 on the other clouds?
- The nearest worker types of the same kind are m6id.8xlarge on AWS and n2-standard-32 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 Standard_D32ds_v5?
- One size down is Standard_D16ds_v5, at 4 DBUs an hour; one size up is Standard_D48ds_v5, at 12. 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