Billing
Puhti compute billing
Puhti is a heterogeneous system with CPU, GPU and IO nodes and nodes with varying amounts of memory. Moreover, it is possible to use only a fraction of a node and its various resources. The billing scheme charges billing units (BU) in a fair way based on the reserved amount of node resources.
In the billing scheme, the BU consumption rate of a compute job depends linearly on the number of requested cores, the number of requested GPUs and the amount of requested memory. More precisely:
- Each reserved core consumes 1 BU per hour.
- Each GiB of reserved memory consumes 0.1 BU per hour.
- Each reserved GiB of NVMe disk (if available) consumes 0.006 BU per hour.
- Each reserved GPU consumes 60 BU per hour.
The total BU consumption per hour is the sum of the above terms:
Total BU = ( NCores * 1 + MemGiBs * 0.1 + NVMeGiBs * 0.006 + NGPUs * 60 ) * Walltime hours
Mahti compute billing
In contrast to Puhti, in Mahti resources are used and billed per node in all normal partitions. In the interactive partition, which can be used for interactive work, as well as for small scale pre- and postprocessing, the usage is billed per CPU core. The GPU nodes in Mahti are billed according to the number of reserved GPUs. Memory is not billed separately.
- Each reserved CPU node consumes 100 BU per hour.
- In interactive partition each core consumes 1 BU per hour.
- Each reserved A100 GPU consumes 100 BU per hour, or in total 400 BU per hour for a full GPU node.
- A slice of an A100 GPU with one seventh of an A100 (a100_1g.5gb) consumes 15 BU per hour
Scratch disk billing
Puhti and Mahti have the same billing for scratch storage. Usage up to 1 TiB is free of charge.
- Excess usage beyond 1 TiB is billed: 1 TiB consumes 5 BU per hour.
ProjAppl disk billing
Puhti and Mahti have the same billing for scratch storage. Usage up to 50 GiB is free of charge.
- Excess usage beyond 50 GiB is billed: 1 TiB consumes 5 BU per hour.