Granted Patent B2

US 9,838,475 · App. 14/246,224 · Granted Dec 5, 2017

Inventors: Eran Brown (Givatayim, IL); Jacob Broido (Tel-Aviv, IL); Anat Rapaport (Hod Hasharon, IL)

Assignee: INFINIDAT LTD.

H04L67/1097H04L41/065H04L41/0677H04L43/0817H04L67/1034H04L69/40

Abstract

Claims (41)

1. A method for connectivity analysis by a mass storage system that is coupled to one or more clusters of host computers, the method comprises:

obtaining, by the mass storage system, hierarchical cluster mapping information that includes: (i) a mapping of host computers to a cluster that maps host computers to the one or more clusters, and (ii) a mapping of host ports to a host computer that maps host ports to host computers;

receiving, from at least one intermediate device of a network that couples the one or more clusters to the mass storage system, host port state information indicative of a state of at least one host port;

determining, based on the hierarchical cluster mapping information, an identity of at least one entity out of: at least one host computer that includes the at least one host port; and at least one cluster that includes the at least one host computer;

estimating, by the mass storage system, a state of at least one entity out of (a) the at least one host computer, and (b) the at least one cluster, so as to provide an estimated state of the at least one entity; wherein the estimating is based on the hierarchical cluster mapping information and the host port state information;

determining whether to generate an alert, by the mass storage system, in response to the estimated state of the at least one entity;

if determining to generate an alert then generating a first alert;

estimating, by the mass storage system, a state of at least one path out of a group of paths that couple the mass storage system to the one or more clusters, based on traffic statistics related to traffic conveyed over the at least one path; and

generating a second alert if the traffic statistics are indicative of a deviation in a traffic pattern from an allowable traffic pattern.

2. The method according to claim 1 wherein the first alert comprises at least one out of the identity of the at least one entity and the estimated state of the at least one entity.

3. The method according to claim 1 wherein the determining of whether to generate an alert, by the mass storage system, is in response to at least the estimated state of the at least one entity.

4. The method according to claim 1 wherein the mass storage system is coupled to multiple clusters, wherein the host port state information refers to a host computer that belongs to a certain cluster of the multiple clusters and the method comprises determining an identity of the certain cluster.

5. The method according to claim 1 wherein the host port state information is indicative of whether a certain host port is connected or disconnected.

6. The method according to claim 1 wherein the at least one intermediate device is at least one fibre channel switch and wherein the host port state information is embedded in messages sent to the mass storage system from the at least one fibre channel switch.

7. The method according to claim 1 comprising generating the second alert if the traffic statistics are indicative of a non-allowable unbalanced traffic.

8. The method according to claim 1 wherein the estimating of the state of the at least one path is further based on traffic rules applied by the network.

9. The method according to claim 1 wherein the estimating comprises estimating a severity of the estimated state.

10. The method according to claim 1 wherein the estimated state is indicative of whether all host ports of a certain host computer have potentially failed.

11. The method according to claim 1 wherein the estimated state is indicative of whether all host ports of a certain cluster have potentially failed.

12. The method according to claim 1 wherein the estimated state is indicative of whether only a part of all host ports of a certain cluster have potentially failed.

13. The method according to claim 1 wherein the estimated state is indicative of whether a host computer has potentially failed.

14. The method according to claim 1 wherein the estimated state is indicative of whether a certain cluster has potentially failed.

15. The method according to claim 1 comprising allocating logical volumes to host computers of the cluster.

16. A non-transitory computer readable medium that stores instructions that once executed by a mass storage system that is coupled to one or more clusters of host computers cause the mass storage system to perform the steps of:

obtaining, by the mass storage system, hierarchical cluster mapping information that includes: (i) a mapping of host computers to a cluster, that maps host computers to the one or more clusters, and (ii) a mapping of host ports to a host computer that maps host ports to host computers;

receiving, from at least one intermediate device of a network that couples the one or more clusters to the mass storage system, host port state information indicative of a state of at least one host port;

determining, based on the hierarchical cluster mapping information, an identity of at least one entity out of: at least one host computer that includes the at least one host port; and at least one cluster that includes the at least one host computer;

estimating, by the mass storage system, a state of at least one entity out of (a) the at least one host computer, and (b) the at least one cluster, so as to provide an estimated state of the at least one entity; wherein the estimating is based on the hierarchical cluster mapping information and the host port state information;

determining whether to generate an alert, by the mass storage system, in response to the estimated state of the at least one entity;

if determining to generate an alert then generating a first alert;

estimating, by the mass storage system, a state of at least one path out of a group of paths that couple the mass storage system to the one or more clusters, based on traffic statistics related to traffic conveyed over the at least one path; and

generating a second alert if the traffic statistics are indicative of a deviation in a traffic pattern from an allowable traffic pattern.

17. A mass storage system that is coupled to one or more clusters of host computers and comprises a storage module and a processor;

wherein the storage module is arranged to store hierarchical cluster mapping information that includes: (i) a mapping of host computers to a cluster that maps host computers to the one or more clusters; and (ii) a mapping of host ports to a host computer that maps host ports to host computers;

wherein the processor is arranged to:

(i) receive, from at least one intermediate device of a network that couples the one or more clusters to the mass storage system, host port state information indicative of a state of at least one host port; and determine, based on the hierarchical cluster mapping information, an identity of at least one entity out of: at least one host computer that includes the at least one host port; and at least one cluster that includes the at least one host computer;

(ii) estimate a state of at least one entity out of (a) the at least one host computer, and (b) the at least one cluster, so as to provide an estimated state of the at least one entity; wherein the estimating is based on the hierarchical cluster mapping information and the host port state information;

(iii) determine whether to generate an alert in response to the estimated state of the at least one entity;

(iv) if generating an alert is determined then generate a first alert;

(v) estimate a state of at least one path out of a group of paths that couple the mass storage system to the one or more clusters, based on traffic statistics related to traffic conveyed over the at least one path; and

(vi) generate a second alert if the traffic statistics are indicative of a deviation in a traffic pattern from an allowable traffic pattern.

Assignments (8)

Continuity (1)