Distributed computation system for servicing queries using revisions maps

1. A method for servicing query requests, comprising:

determining, by a first query node and in response to a first query request, a user data sub-region and a target sub-region associated with the first query request,

wherein determining the user data sub-region associated with the query request comprising using a data flow graph,

wherein the data flow graph is associated with a data model and specifies relationships between a plurality of sub-regions in the data model, and

wherein the user data sub-region is one of the plurality of sub-regions;

issuing a revisions map request to a database management system;

obtaining, in response to the revisions map request, a revisions map, wherein the revisions map specifies a version of user data associated with the user data sub-region to use to service the query request, wherein the version of user data is one of a plurality of versions of user data;

identifying, using the data flow graph, a first input sub-region of the target sub-region;

identifying a second query node associated with the first input sub-region, wherein the second query node is a physical computing device or a logical computing device;

issuing a second query request to the second query node to obtain data associated with the first input sub-region;

receiving, by the second query node, the second query request, wherein the second query request comprises the revisions map;

making a first determination that data stored in a cache on the second query node is associated with the first input sub-region and is not associated with the version of the user data specified in the revisions man;

based on the first determination, making a second determination that the data associated with the first input sub-region is the user data;

based on the second determination, obtaining the user data associated with the first input sub-region from a persistent storage;

sending the second query result to the first query node, wherein the second query result comprises the user data obtained from the persistent storage;

receiving, in response to the second query request, a second query result from the second query node, wherein the second query result is based on the version of the user data;

generating a first query result for the target sub-region using at least the second query result; and

providing the first query result to an entity associated with the issuance of the first query request.

2. The method of claim 1 , further comprising:

receiving, by the second query node, the second query request, wherein the second query request comprises the revisions map;

determining that data stored in a cache on the second query node is associated with the first input sub-region and with the version of the user data specified in the revisions map; and

based on the determination, sending the second query result to the first query node, wherein the second query result comprises at least a portion of the data stored in the cache.

3. The method of claim 2 , wherein the data stored in the cache is derived data.

4. The method of claim 1 , further comprising:

receiving, by the second query node, the second query request;

identifying, using the data flow graph, a second input sub-region of the first input sub-region;

identifying a third query node associated with the second input sub-region;

issuing a third query request to the third query node to obtain data associated with the second input sub-region;

receiving a third query response from the third query node;

determining a value for at least one cell in the first input sub-region using the third query response; and

sending the second query result to the first query node, wherein the second query result comprises the value.

5. The method of claim 1 , wherein the version of user data is a most recent version of user data.

6. The method of claim 1 , wherein identifying the second query node associated with the first input sub-region comprises:

determining a distribution key using an outer dimension associated with the first input sub-region;

identifying the second query node using the distribution key and a node distribution key list.

7. The method of claim 6 , wherein the distribution key is a value generated using consistent hashing.

8. The method of claim 6 , wherein the second query node is a lead query node and wherein the second query node distributes updates to the node distribution key list to the first query node.

9. The method of claim 1 , wherein the entity is one selected from a group consisting of a client and a subscription engine.

10. The method of claim 1 , wherein the first query request is issued in response to a subscription engine receiving a notification that user data associated with the target sub-region has changed.

11. The method of claim 1 , wherein the data flow graph comprises at least one cycle.

12. The method of claim 1 , wherein the first query node comprises the target sub-region and a second sub-region, wherein each cell in the target sub-region and the second sub-region are associated with the same outer dimension.

13. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for servicing query requests, the method comprising:

determining, by a first query node and in response to a first query request, a user data sub-region and a target sub-region associated with the first query request,

wherein determining the user data sub-region associated with the query request comprising using a data flow graph,

wherein the data flow graph is associated with a data model and specifies relationships between a plurality of sub-regions in the data model, and

wherein the user data sub-region is one of the plurality of sub-regions;

issuing a revisions map request to a database management system;

obtaining, in response to the revisions map request, a revisions map, wherein the revisions map specifies a version of user data associated with the user data sub-region to use to service the query request, wherein the version of user data is one of a plurality of versions of user data;

identifying, using the data flow graph, a first input sub-region of the target sub-region;

identifying a second query node associated with the first input sub-region;

issuing a second query request to the second query node to obtain data associated with the first input sub-region, wherein the second query node is a physical computing device or a logical computing device;

receiving, by the second query node, the second query request, wherein the second query request comprises the revisions map;

making a first determination that data stored in a cache on the second query node is associated with the first input sub-region and is not associated with the version of the user data specified in the revisions map;

based on the first determination, making a second determination that the data associated with the first input sub-region is the user data;

based on the second determination, obtaining the user data associated with the first input sub-region from a persistent storage;

sending the second query result to a first query node, wherein the second query result comprises the user data obtained from the persistent storage;

receiving, in response to the second query request, a second query result from the second query node, wherein the second query result is based on the version of the user data;

generating a first query result for the target sub-region using at least the second query result; and

providing the first query result to an entity associated with the issuance of the first query request.

14. The non-transitory computer readable medium of claim 13 , wherein the method further comprises:

receiving, by the second query node, the second query request, wherein the second query request comprises the revisions map;

determining that data stored in a cache on the second query node is associated with the first input sub-region and with the version of the user data specified in the revisions map; and

based on the determination, sending the second query result to the first query node, wherein the second query result comprises at least a portion of the data stored in the cache.

15. The non-transitory computer readable medium of claim 13 , wherein identifying the second query node associated with the first input sub-region comprises:

determining a distribution key using an outer dimension associated with the first input sub-region; and

identifying the second query node using the distribution key and a node distribution key list,

wherein the distribution key is a value generated using consistent hashing.