Neil Conway created MESOS-5741:
----------------------------------
Summary: Quota + reserved resources is overly conservative
Key: MESOS-5741
URL: https://issues.apache.org/jira/browse/MESOS-5741
Project: Mesos
Issue Type: Bug
Reporter: Neil Conway
Consider this scenario:
* Cluster has 10 CPUs total. 8 CPUs are reserved for role X, 2 CPUs are
unreserved.
* Role X has a quota for 4 CPUs, but has only been allocated 2 CPUs (e.g.,
because it has declined an offer for the other 2 CPUs). The CPUs it has been
allocated come from the reserved resources, so there are 6 reserved CPUs and 2
unreserved CPUs available.
* That means 6 CPUs should be offered as non-quota resources. However, which 6
CPUs should be offered -- the 6 reserved CPUs, or 4 reserved CPUs and 2
unreserved CPUs?
The current quota allocation logic appears to always offer the 6 reserved CPUs.
This is unfortunate, because frameworks in other roles won't be able to use
those resources. The reason for this behavior is:
{code}
Resources remainingClusterResources = roleSorter->totalScalarQuantities();
foreachkey (const string& role, activeRoles) {
remainingClusterResources -= roleSorter->allocationScalarQuantities(role);
}
{code}
{{remainingClusterResources}} may have a {{role}} set (although dynamically
reserved resources will have been converted into effectively static
reservations).
{code}
Resources unallocatedQuotaResources;
foreachpair (const string& name, const Quota& quota, quotas) {
// Compute the amount of quota that the role does not have allocated.
//
// NOTE: Revocable resources are excluded in `quotaRoleSorter`.
// NOTE: Only scalars are considered for quota.
Resources allocated = getQuotaRoleAllocatedResources(name);
const Resources required = quota.info.guarantee();
unallocatedQuotaResources += (required - allocated);
}
{code}
{{unallocatedQuotaResources}} will *not* have {{role}} set, per the
implementation of {{getQuotaRoleAllocatedResources}}.
{code}
remainingClusterResources -= unallocatedQuotaResources;
{code}
This means that *only* unreserved resources will be subtracted from
{{remainingClusterResources}}. At best this is sub-optimal, because it seems
better to lay-away resources for role X that are already reserved for X. I
don't *think* it will result in violating quota guarantees, though.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
