Cache-assisted service contact instance selection for computing-aware traffic steering networks
In a computing-aware traffic steering (CATS) network, such as those conforming to an IETF proposal, a service contact instance for a requested service is selected based on at least cache metrics associated with data cached at one or more egress nodes of the CATS network. In some embodiments, the selection is also based on compute and network metrics. In one implementation, the selection is based only on cache metrics unless that selection is too costly in terms of compute and/or network load. In that case, the selection is based on the compute and network metrics. In this way, cache metrics are prioritized over compute and network metrics as long as the cache-based selection is not too costly.
1 . A method for determining a selected service contact instance for a requested service in a computing-aware traffic steering (CATS) network, the method comprising an ingress node of the CATS network:
receiving cache metrics for one or more possible service contact instances for the requested service, wherein the cache metrics are different from compute and network metrics;
determining the selected service contact instance for the requested service based on the cache metrics; and
implementing the requested service based on the selected service contact instance.
2 . The method of claim 1 , further comprising the ingress node:
receiving compute metrics for the one or more possible service contact instances; and
receiving network metrics for the one or more possible service contact instances, wherein the ingress node determines the selected service contact instance for the requested service based on the compute, network, and cache metrics.
3 . The method of claim 2 , wherein determining the selected service contact instance comprises:
selecting a cache-metric candidate service contact instance based on the cache metrics;
determining if the cache-metric candidate service contact instance satisfies at least one of (i) one or more compute-metric threshold tests and (ii) one or more network-metric threshold tests;
if so, then determining the cache-metric candidate service contact instance to be the selected service contact instance; and
if not, then determining a compute-and-network-metrics candidate service contact instance to be the selected service contact instance.
4 . The method of claim 3 , wherein:
the cache-metric candidate service contact instance is determined to be the selected service contact instance if the cache-metric candidate service contact instance is determined to satisfy both (i) the one or more compute-metric threshold tests and (ii) the one or more network-metric threshold tests; and
the compute-and-network-metrics candidate service contact instance is determined to be the selected service contact instance if the cache-metric candidate service contact instance is determined to fail at least one compute-metric threshold test or network-metric threshold test.
5 . The method of claim 3 , wherein the compute-and-network-metrics candidate service contact instance is selected based on the compute and network metrics, but not on the cache metrics.
6 . The method of claim 3 , wherein the compute-and-network-metrics candidate service contact instance is selected before determining whether the cache-metric candidate service contact instance fails at least one of the compute-metric threshold test and the network-metric threshold test.
7 . The method of claim 3 , wherein the compute-and-network-metrics candidate service contact instance is selected after determining that the cache-metric candidate service contact instance fails at least one of the compute-metric threshold test and the network-metric threshold test.
8 . The method of claim 2 , wherein:
the compute metrics and the cache metrics are received from one or more egress nodes of the CATS network; and
the network metrics are received from a network metric agent of the CATS network.
9 . An apparatus comprising:
at least one processor; and
at least one memory storing instructions that, upon being executed by the at least one processor, cause the apparatus at least to:
receive cache metrics for one or more possible service contact instances for the requested service, wherein the cache metrics are different from compute and network metrics;
determine the selected service contact instance for the requested service based on the cache metrics; and
implement the requested service based on the selected service contact instance.
10 . The apparatus of claim 9 , wherein the apparatus is further adapted to:
receive compute metrics for the one or more possible service contact instances; and
receive network metrics for the one or more possible service contact instances, wherein the apparatus is adapted to determine the selected service contact instance for the requested service based on the compute, network, and cache metrics.
11 . The apparatus of claim 10 , wherein the apparatus is adapted to determine the selected service contact instance by:
selecting a cache-metric candidate service contact instance based on the cache metrics;
determining if the cache-metric candidate service contact instance satisfies at least one of (i) one or more compute-metric threshold tests and (ii) one or more network-metric threshold tests;
if so, then determining the cache-metric candidate service contact instance to be the selected service contact instance; and
if not, then determining a compute-and-network-metrics candidate service contact instance to be the selected service contact instance.
12 . The apparatus of claim 11 , wherein:
the apparatus is adapted to determine the cache-metric candidate service contact instance to be the selected service contact instance if the cache-metric candidate service contact instance is determined to satisfy both (i) the one or more compute-metric threshold tests and (ii) the one or more network-metric threshold tests; and
the apparatus is adapted to determine the compute-and-network-metrics candidate service contact instance to be the selected service contact instance if the cache-metric candidate service contact instance is determined to fail at least one compute-metric threshold test or network-metric threshold test.
13 . The apparatus of claim 11 , wherein the apparatus is adapted to select the compute-and-network-metrics candidate service contact instance based on the compute and network metrics, but not on the cache metrics.
14 . The apparatus of claim 11 , wherein the apparatus is adapted to select the compute-and-network-metrics candidate service contact instance before determining whether the cache-metric candidate service contact instance fails at least one of the compute-metric threshold test and the network-metric threshold test.
15 . The apparatus of claim 11 , wherein the apparatus is adapted to select the compute-and-network-metrics candidate service contact instance after determining that the cache-metric candidate service contact instance fails at least one of the compute-metric threshold test and the network-metric threshold test.
16 . The apparatus of claim 10 , wherein the apparatus is adapted to:
receive the compute metrics and the cache metrics from one or more egress nodes of the CATS network; and
receive the network metrics from a network metric agent of the CATS network.
17 . The apparatus of claim 9 , wherein the apparatus is an ingress node of a CATS network.