Vector-based anomaly detection

1. A hybrid-fabric apparatus for detecting anomalous behavior of a network fabric comprising a plurality of network nodes, the hybrid-fabric apparatus comprising:

at least one memory configured to at least store a plurality of behavior metrics;

at least one processor coupled to the at least one memory; and

software code configured to use the at least one processor, access the at least one memory, and cause the apparatus to at least:

disaggregate a set of anomaly detection criteria into a plurality of anomaly criterion to be distributed among the plurality of network nodes, the set of anomaly detection criteria characterizing a variation from a baseline vector corresponding to nominal traffic flow through the network fabric, and the plurality of anomaly criterion comprising a function of a measured vector of behavior metrics comprising a threshold;

aggregate anomaly criterion statuses calculated by at least some of the plurality of network nodes to detect anomalous behavior, each anomaly criterion status being calculated with respect to a network node as a function of the network node's anomaly criterion and the measured behavior vector of behavior metrics; and

initiate a notification regarding the anomalous behavior.

2. The hybrid-fabric apparatus of claim 1 , wherein the baseline vector comprises at least two correlated behavior metrics.

3. The hybrid-fabric apparatus of claim 1 , further caused to at least establish the anomaly detection criteria by at least:

simulating an anomalous behavior within the network fabric while the network fabric is active to derive the variation from the baseline vector.

4. The hybrid-fabric apparatus of claim 1 , further caused to at least establish the anomaly detection criteria by at least:

modeling an anomalous behavior within the network fabric while the network fabric is active to derive the variation from the baseline vector by at least constructing a logical representation of the network fabric by at least using nodes of the network fabric.

5. The hybrid-fabric apparatus of claim 1 , further caused to at least establish the anomaly detection criteria by at least:

modeling the anomalous behavior by at least running a live drill.

6. The hybrid-fabric apparatus of claim 1 , wherein at least some of the plurality of network nodes calculate their anomaly criterion statuses as a function of a trend of the measured vector of the behavior metrics.

7. The hybrid-fabric apparatus of claim 1 , further caused to at least generate a leading indicator of a likelihood that anomalous behavior is about to occur as a function of the aggregated anomaly criterion statuses.

8. The hybrid-fabric apparatus of claim 7 , further caused to at least generate the leading indicator by at least calculating a likelihood of the anomalous behavior occurring while the anomaly detection criteria remains unsatisfied.

9. The hybrid-fabric apparatus of claim 1 , further caused to at least identify an anomaly type of the anomalous behavior based on the anomaly criterion statuses.

10. The hybrid-fabric apparatus of claim 8 , further caused to at least automatically respond to the anomalous behavior according to a prior defined action based at least in part on the anomaly type.

11. The hybrid-fabric apparatus of claim 1 , further caused to at least migrate anomalous traffic to a monitored data channel within the network fabric.

12. The hybrid-fabric apparatus of claim 1 , further caused to at least update the anomaly detection criteria according to a known change in the network fabric and sending the receiving nodes correspondingly updated anomaly criterion.

13. The hybrid-fabric apparatus of claim 12 , wherein the known change comprises an expected behavior change vector reflecting expected behavior changes due to deployment of an application within the network fabric.

14. The hybrid-fabric apparatus of claim 1 , further caused to at least store a history of the anomaly criterion statuses in the at least one memory.