# ComputeQuotaResourceConfigProperty

Per-instance-type compute quota element.

Reused under ComputeQuotaConfig.ComputeQuotaResources and ResourceSharingConfig.AbsoluteBorrowLimits.

- Source: Generated code
- Type last changed: v2.276.0 · 10 Nov 2026

## Example

```ts
// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import { aws_sagemaker as sagemaker } from 'aws-cdk-lib';
const computeQuotaResourceConfigProperty: sagemaker.CfnComputeQuota.ComputeQuotaResourceConfigProperty = {
  instanceType: 'instanceType',

  // the properties below are optional
  acceleratorPartition: {
    count: 123,
    type: 'type',
  },
  accelerators: 123,
  count: 123,
  memoryInGiB: 123,
  vCpu: 123,
};
```

## Properties

- `acceleratorPartition`: [IResolvable](/cdk/api/v2/docs/aws-cdk-lib.IResolvable.html) | [CfnComputeQuota.AcceleratorPartitionConfigProperty](/cdk/api/v2/docs/aws-cdk-lib.aws_sagemaker.CfnComputeQuota.AcceleratorPartitionConfigProperty.html) — Multi-instance GPU (MIG) partition profile and count for fractional accelerator allocation.

- `accelerators`: number — The number of accelerators (e.g., GPUs) of this type allocated to the team. SageMaker auto-fills VCpu / MemoryInGiB ratio-proportionally if only this is set.

- `count`: number — The number of instances of this type allocated to the team.

- `instanceType`: string (required) — The cluster instance type the quota applies to (e.g., 'ml.m5.xlarge', 'ml.p5.48xlarge'). Validated by the service against the supported HyperPod instance types.

- `memoryInGiB`: number — The memory in GiB allocated to the team.

  Floating-point. SageMaker auto-fills Accelerators / VCpu ratio-proportionally if only this is set.

- `vCpu`: number — The number of vCPUs allocated to the team.

  Floating-point. SageMaker auto-fills Accelerators / MemoryInGiB ratio-proportionally if only this is set.