[MESOS-5377] Improve DRF behavior with scarce resources.

• Type: Epic
• Status: Accepted
• Priority: Major
• Resolution: Unresolved
• Component/s: allocation
• Epic Name: Scarce Resources
• Assignee: Guangya Liu
• Reporter: Benjamin Mahler
• Shepherd: Benjamin Mahler
• Created: 13/May/16 01:16
• Updated: 01/Jul/16 01:15

Description

The allocator currently uses the notion of Weighted Dominant Resource Fairness (WDRF) to establish a linear notion of fairness across allocation roles.

DRF behaves well for resources that are present within each machine in a cluster (e.g. CPUs, memory, disk). However, some resources (e.g. GPUs) are only present on a subset of machines in the cluster.

Consider the behavior when there are the following agents in a cluster:

1000 agents with (cpus:4,mem:1024,disk:1024)
1 agent with (gpus:1,cpus:4,mem:1024,disk:1024)

If a role wishes to use both GPU and non-GPU resources for tasks, consuming 1 GPU will lead DRF to consider the role to have a 100% share of the cluster, since it consumes 100% of the GPUs in the cluster. This framework will then not receive any other offers.

Among possible improvements, fairness can have understanding of resource packages. In a sense there is 1 GPU package that is competed on and 1000 non-GPU packages competed on, and ideally a role's consumption of the single GPU package does not have a large effect on the role's access to the other 1000 non-GPU packages.

In the interim, we should consider having a recommended way to deal with scarce resources in the current model.

Issue Links

• is related to: MESOS-5758 Add ability to exclude resources from fair sharing.
• is related to: MESOS-5634 Add Framework Capability for GPU_RESOURCES
• relates to: MESOS-4923 Treat revocable resources as a separate pool when considering fairness
• relates to: MESOS-4424 Initial support for GPU resources.