Multi-layer QOS management in a distributed computing environment
A method for multi-layer quality of service (QoS) management in a distributed computing environment includes receiving a workload and identifying a workload QoS class for the workload. The workload is scheduled to be run compute nodes. The workload QoS class is translated to a plurality of different resource layer QoS classes, and the plurality of QoS classes are communicated to a workload execution manager of the compute nodes. The workload execution manager extends the plurality of different resource layer QoS classes to respective resource layers involved in the running of the workload.
1. A method, comprising: receiving a workload and identifying a workload quality of service (QoS) class for the workload; scheduling running of the workload on one or more compute nodes; translating the workload QoS class to a plurality of different resource layer QoS classes; communicating the plurality of QoS classes to a workload execution manager of the compute nodes; and extending the plurality of different resource layer QoS classes to respective resource layers involved in the running of the workload.
2. The method of claim 1 , further comprising identifying a storage level QoS class from the plurality of different resource layer QoS classes.
3. The method of claim 1 , further comprising identifying a network level QoS class from the plurality of different resource layer QoS classes.
4. The method of claim 1 , further comprising: identifying a storage level QoS class from the plurality of different resource layer QoS classes; and communicating the storage level QoS class to a local storage manager of a respective compute node, the local storage manager managing a storage resource local to the compute node according to the storage level QoS class.
5. The method of claim 1 , further comprising: identifying a storage level QoS class and a network level QoS class from the plurality of different resource layer QoS classes; and communicating the storage level QoS class and the network level QoS class to a local storage manager of a respective compute node, the local storage manager managing a storage resource local to the compute node according to the storage level QoS class and the network level QoS class.
6. The method of claim 1 , further comprising: identifying a storage level QoS class from the plurality of different resource layer QoS classes; and communicating the storage level QoS class to a distributed storage manager, the distributed storage manager managing non-local, distributed storage resources relative to the compute node according to the storage level QoS class.
7. The method of claim 1 , further comprising: identifying a network level QoS class from the plurality of different resource layer QoS classes; and communicating the network level QoS class to a distributed storage manager, the distributed storage manager managing non-local, distributed storage resources relative to the compute node according to the network level QoS class.
8. The method of claim 1 , further comprising: checking QoS requirements for running and pending workloads; and notifying a storage manager of a quantity of running and pending workloads associated with a particular QoS class.
9. The method of claim 8 , further comprising adjusting reserved storage resource allocations based on workload scheduling policies and resource demands as indicated by the workload scheduler according to workload scheduling policies and the quantity of running and pending workloads associated with the particular QoS class.
10. The method of claim 1 , further comprising: identifying a storage level QoS class and a network level QoS class from the plurality of different resource layer QoS classes; and communicating the storage level QoS class and the network level QoS class to a local storage manager of a respective compute node, the local storage manager assigning a storage allocation and a bandwidth allocation based on the storage level QoS class and the network level QoS class.