Methods and apparatus for generating causality matrix and impacts using graph processing
Methods and apparatus for generating a causality matrix using vertex-centric processing framework to be used by a codebook correlation engine to determine a set of problems to explain active symptoms in a system. Methods and apparatus for calculating impacts of problems using vertex-centric processing framework.
1. A method, comprising:
reading a domain model of a system, the domain model corresponding to a plurality of classes, wherein each of the classes correspond to a respective managed object in the system and have a plurality of attributes, the attribute being one of relationship attribute, external attribute, and computed attribute;
optimizing the domain model;
generating a semantic model as an instantiation of the domain model, the semantic model comprising a plurality of nodes, wherein each of the nodes corresponds to a respective one of the classes of the domain model;
importing topology information based on the semantic model and relationship among the plurality of classes, wherein vertices in the topology correspond to a respective one of the classes of the domain model;
generating initial messages for each vertex, wherein the initial messages correspond to updates in the external attributes of the classes;
for each vertex in the topology,
a) determining a semantic model that corresponds to the vertex;
b) processing messages in a processing queue of the vertex, the processing of any of the messages including identifying an update expression associated with the message, identifying a dependent expression that is associated with the update expression, detecting whether the update expression and the dependent expression belong to the same class, and re-evaluating the dependent expression when the update expression and the dependent expression belong to the same class;
c) sending dependent messages generated during step b) to the one or more target vertices corresponding to each of the dependent messages; repeating steps a)-c) until no more messages are generated; and
processing updates of the attributes caused by the processing of the messages.
2. The method according to claim 1 , further comprising:
processing impact information;
generating a semantic model including impact expressions as an instantiation of the domain model;
storing the impact information for problems in the system; and
updating impacts for each of the problems based upon messages containing the updates of the attributes.
3. The method according to claim 2 , wherein a first one of the problems includes a storage device being unavailable, further comprising propagating a message for the storage device being unavailable to indicate a potential data loss.
4. The method according to claim 1 , the processing messages further comprising:
determining a semantic model node corresponding the message being processed; and
evaluating and updating a value of the message.
5. The method according to claim 1 , wherein the plurality of classes includes storage devices, volumes, and storage pools.
6. The method according to claim 1 , further comprising maintaining an unavailable device count.
7. A system, comprising:
a processor and a memory configured to:
process a domain model of a system, the domain model corresponding to a plurality of classes, wherein each of the classes correspond to a respective managed object in the system and have a plurality of attributes, the attribute being one of relationship attribute, external attribute, and computed attribute;
generate a semantic model as an instantiation of the domain model, the semantic model comprising a plurality of nodes, wherein each of the nodes corresponds to a respective one of the classes of the domain model;
import topology information based on the semantic model and relationship among the plurality of classes, wherein vertices in the topology correspond to a respective one of the classes of the domain model;
generate initial messages for each vertex, wherein the initial messages correspond to updates in the external attributes of the classes;
for each vertex in the topology,
a) determine a semantic model node that corresponds to the vertex;
b) processing messages in a processing queue of the vertex, the processing of any of the messages including identifying an update expression associated with the message, identifying a dependent expression that is associated with the update expression, detecting whether the update expression and the dependent expression belong to the same class, and re-evaluating the dependent expression when the update expression and the dependent expression belong to the same class;
c) send dependent messages generated during step b) to the one or more target vertices corresponding to each of the dependent messages;
repeating steps a)-c) until no more message is generated; and
process updates of the attributes caused by the processing of the messages; and
maintain an unavailable device count.
8. The system according to claim 7 , the processor and a memory is further configured to:
process impact information;
generate an impact semantic model including impact expressions as an instantiation of the domain model;
store the impact information for problems in the system; and
update impacts for each of the problems based upon messages containing the updates of the attributes.
9. The system according to claim 8 , wherein a first one of the problems includes a storage device being unavailable, the processor and a memory is further configured to propagate a message for the storage device being unavailable to indicate a potential data loss.
10. The system according to claim 7 , the step to process messages further comprising:
determining a semantic model node corresponding the message being processed; and
evaluating and updating a value of the message.
11. The system according to claim 7 , wherein the system comprises a data center environment having elastic converged storage.
12. The system according to claim 7 , wherein the plurality of classes includes storage devices, volumes, and storage pools.
13. The system according to claim 7 , wherein the plurality of classes includes storage devices, volumes, and storage pools.
14. An article, comprising:
a non-transitory storage medium having stored instructions that enable a machine to:
process a domain model of a system, the domain model corresponding to a plurality of classes, wherein each of the classes correspond to a respective managed object in the system and have a plurality of attributes, the attribute being one of relationship attribute, external attribute, and computed attribute;
generate a semantic model as an instantiation of the domain model, the semantic model comprising a plurality of nodes, wherein each of the nodes corresponds to a respective one of the classes of the domain model;
import topology information based on the semantic model and relationship among the plurality of classes, wherein vertices in the topology correspond to a respective one of the classes of the domain model;
generate initial messages for each vertex, wherein the initial messages correspond to updates in the external attributes of the classes;
for each vertex in the topology,
a) determine a semantic model node that corresponds to the vertex;
b) processing messages in a processing queue of the vertex, the processing of any of the messages including identifying an update expression associated with the message, identifying a dependent expression that is associated with the update expression, detecting whether the update expression and the dependent expression belong to the same class, and re-evaluating the dependent expression when the update expression and the dependent expression belong to the same class;
c) send dependent messages generated during step b) to the one or more target vertices corresponding to each of the dependent messages;
repeating steps a)-c) until no more message is generated;
process updates of the attributes caused by the processing of the messages;
maintain an unavailable device count;
process impact information;
generate an impact semantic model including impact expressions as an instantiation of the domain model;
store the impact information for problems in the system; and
update impacts for each of the problems based upon messages containing the updates of the attributes.
15. The article according to claim 14 , wherein a first one of the problems includes a storage device being unavailable, and the instructions further enable the machine to propagate a message for the storage device being unavailable to indicate a potential data loss.
16. The article according to claim 14 , the step to process messages further comprising:
determining a semantic model node corresponding the message being processed; and
evaluating and updating a value of the message.
17. The article according to claim 14 , wherein the system comprises a data center environment having elastic converged storage.