openstack/nova
Code Issues Proposed changes

placement: Object changes for granular

Browse Source 
To implement the merge-and-filter algorithm for granular resource
requests later in this code series, a handful of object changes are
needed.  This change set just makes those alterations.

Change-Id: I446f7534e7f3b27f71683283ce4a8fd361fc05a7
blueprint: granular-resource-requests
This commit is contained in:
Eric Fried 2017-11-04 09:05:43 -05:00
parent 71de700c8f
commit 7f996eb2c4
1 changed files with 59 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

78
nova/api/openstack/placement/objects/resource_provider.py
View File

@ -2769,6 +2769,15 @@ class AllocationRequestResource(base.VersionedObject):
class AllocationRequest(base.VersionedObject):
fields = {
# UUID of (the root of the tree including) the non-sharing resource
# provider associated with this AllocationRequest. Internal use only,
# not included when the object is serialized for output.
'anchor_root_provider_uuid': fields.UUIDField(),
# Whether all AllocationRequestResources in this AllocationRequest are
# required to be satisfied by the same provider (based on the
# corresponding RequestGroup's use_same_provider attribute). Internal
# use only, not included when the object is serialized for output.
'use_same_provider': fields.BooleanField(),
'resource_requests': fields.ListOfObjectsField(
'AllocationRequestResource'
),
@ -2782,6 +2791,9 @@ class ProviderSummaryResource(base.VersionedObject):
'resource_class': rc_fields.ResourceClassField(read_only=True),
'capacity': fields.NonNegativeIntegerField(),
'used': fields.NonNegativeIntegerField(),
# Internal use only; not included when the object is serialized for
# output.
'max_unit': fields.NonNegativeIntegerField(),
}
@ -2816,6 +2828,7 @@ def _get_usages_by_provider_and_rc(ctx, rp_ids, rc_ids):
# , inv.total
# , inv.reserved
# , inv.allocation_ratio
# , inv.max_unit
# , usage.used
# FROM resource_providers AS rp
# JOIN inventories AS inv
@ -2870,6 +2883,7 @@ def _get_usages_by_provider_and_rc(ctx, rp_ids, rc_ids):
inv.c.total,
inv.c.reserved,
inv.c.allocation_ratio,
inv.c.max_unit,
usage.c.used,
]).select_from(usage_join).where(
sa.and_(rpt.c.id.in_(rp_ids),
@ -3348,9 +3362,9 @@ def _build_provider_summaries(context, usages, prov_traits):
if not summary:
summary = ProviderSummary(
context,
resource_provider=ResourceProvider(
resource_provider=ResourceProvider.get_by_uuid(
context,
uuid=rp_uuid,
rp_uuid,
),
resources=[],
)
@ -3362,6 +3376,7 @@ def _build_provider_summaries(context, usages, prov_traits):
resource_class=rc_name,
capacity=cap,
used=used,
max_unit=usage['max_unit'],
)
summary.resources.append(rpsr)
summary.traits = [Trait(context, name=tname) for tname in traits]
@ -3419,37 +3434,46 @@ def _shared_allocation_request_resources(ctx, ns_rp_id, requested_resources,
return res_requests
def _allocation_request_for_provider(ctx, requested_resources, rp_uuid):
def _allocation_request_for_provider(ctx, requested_resources, provider):
"""Returns an AllocationRequest object containing AllocationRequestResource
objects for each resource class in the supplied requested resources dict.
:param ctx: nova.context.RequestContext object
:param requested_resources: dict, keyed by resource class ID, of amounts
being requested for that resource class
:param rp_uuid: UUID of the resource provider supplying the resources
:param provider: ResourceProvider object representing the provider of the
resources.
"""
resource_requests = [
AllocationRequestResource(
ctx, resource_provider=ResourceProvider(ctx, uuid=rp_uuid),
ctx, resource_provider=provider,
resource_class=_RC_CACHE.string_from_id(rc_id),
amount=amount,
) for rc_id, amount in requested_resources.items()
]
return AllocationRequest(ctx, resource_requests=resource_requests)
# NOTE(efried): This method only produces an AllocationRequest with its
# anchor in its own tree. If the provider is a sharing provider, the
# caller needs to identify the other anchors with which it might be
# associated.
return AllocationRequest(
ctx, resource_requests=resource_requests,
anchor_root_provider_uuid=provider.root_provider_uuid)
def _alloc_candidates_no_shared(ctx, requested_resources, rps):
def _alloc_candidates_single_provider(ctx, requested_resources, rps):
"""Returns a tuple of (allocation requests, provider summaries) for a
supplied set of requested resource amounts and resource providers. The
supplied resource providers have capacity to satisfy ALL of the resources
in the requested resources as well as ALL required traits that were
requested by the user.
This is an optimized code path for the common scenario when no sharing
providers exist in the system for any requested resource. In this scenario,
we can more efficiently build the list of AllocationRequest and
ProviderSummary objects due to not having to determine requests for some
shared and some non-shared resources.
This is used in two circumstances:
- To get results for a RequestGroup with use_same_provider=True.
- As an optimization when no sharing providers satisfy any of the requested
resources, and nested providers are not in play.
In these scenarios, we can more efficiently build the list of
AllocationRequest and ProviderSummary objects due to not having to
determine requests across multiple providers.
:param ctx: nova.context.RequestContext object
:param requested_resources: dict, keyed by resource class ID, of amounts
@ -3478,10 +3502,20 @@ def _alloc_candidates_no_shared(ctx, requested_resources, rps):
alloc_requests = []
for rp_id in rp_ids:
rp_summary = summaries[rp_id]
rp_uuid = rp_summary.resource_provider.uuid
req_obj = _allocation_request_for_provider(ctx, requested_resources,
rp_uuid)
req_obj = _allocation_request_for_provider(
ctx, requested_resources, rp_summary.resource_provider)
alloc_requests.append(req_obj)
# If this is a sharing provider, we have to include an extra
# AllocationRequest for every possible anchor.
if os_traits.MISC_SHARES_VIA_AGGREGATE in rp_summary.traits:
for anchor in _anchors_for_sharing_provider(
ctx, rp_summary.resource_provider.id):
# We already added self
if anchor == rp_summary.resource_provider.root_provider_uuid:
continue
req_obj = copy.deepcopy(req_obj)
req_obj.anchor_root_provider_uuid = anchor
alloc_requests.append(req_obj)
return alloc_requests, list(summaries.values())
@ -3573,6 +3607,7 @@ def _alloc_candidates_with_shared(ctx, requested_resources, required_traits,
# provider involved in the request
ns_rp_summary = summaries[ns_rp_id]
ns_rp_uuid = ns_rp_summary.resource_provider.uuid
anchor_rp_uuid = ns_rp_summary.resource_provider.root_provider_uuid
ns_resource_class_names = ns_rp_summary.resource_class_names
ns_resources = set(
rc_id for rc_id in requested_resources
@ -3667,7 +3702,8 @@ def _alloc_candidates_with_shared(ctx, requested_resources, required_traits,
continue
alloc_prov_ids.append(all_prov_ids)
req = AllocationRequest(ctx, resource_requests=list(res_requests))
req = AllocationRequest(ctx, resource_requests=list(res_requests),
anchor_root_provider_uuid=anchor_rp_uuid)
alloc_requests.append(req)
# The process above may have removed some previously-identified resource
@ -3791,7 +3827,11 @@ class AllocationCandidates(base.VersionedObject):
)
@staticmethod
@db_api.api_context_manager.reader
# TODO(efried): This is only a writer context because it accesses the
# resource_providers table via ResourceProvider.get_by_uuid, which does
# data migration to populate the root_provider_uuid. Change this back to a
# reader when that migration is no longer happening.
@db_api.api_context_manager.writer
def _get_by_requests(context, requests, limit=None):
# We first get the list of "root providers" that either have the
# requested resources or are associated with the providers that
@ -3856,8 +3896,8 @@ class AllocationCandidates(base.VersionedObject):
required_trait_map,
forbidden_trait_map,
member_of)
alloc_request_objs, summary_objs = _alloc_candidates_no_shared(
context, resources, rp_ids)
alloc_request_objs, summary_objs = (
_alloc_candidates_single_provider(context, resources, rp_ids))
else:
if required_trait_map:
# TODO(cdent): Now that there is also a forbidden_trait_map