GEOGRAPHICALLY PARTITIONED ONLNE SEARCH SYSTEM

• US 20130073581A1
• Filed: 09/15/2011
• Published: 03/21/2013
• Est. Priority Date: 09/15/2011
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Sandholm, Thomas E.
• Granted Patent: US 9,396,275 B2
• Time in Patent Office: Days
• US Class Current: 707/769
• CPC Class Codes: G06F 16/2471 Distributed queries
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Abstract

In one example, a memory includes a data structure organized to include information resident in a number of databases geographically-partitioned according to a number of shards. The data structure includes a geographically-hashed look up table containing a number of gee-hash ranges associated with the shards and is organized to enable instructions to execute and access the ranges according to a number of characters in a generated geo-hash value. The data structure also includes a plurality of user data object tables that contain respective user data attributes and link to a particular server node and an associated geo-partitioned database.

Claims

1. A computer implemented method of interacting with a data structure to exchange mobile resource information between a client network node and an online search system, the method comprising:

providing a number of databases geographically-partitioned according to a number of shards as part of the resource search system;

generating a geographically based hash (geo-hash) value from a hashing algorithm based on geographic location information received from the client network node;

retrieving, from a shard location table, a database server identifier associated with the geo-hash value;

accessing the identified database server; and

querying the identified database server for mobile resource information in association with the geographic location information.

2. The computer implemented method of claim 1, wherein providing the number of databases geographically-partitioned according to the number of shards includes providing the number of databases partitioned into geographic areas for shards such that search queries limited to thirty miles from a particular geographic location can touch between one to four shards.

3. The computer implemented method of claim 1, wherein providing the number of databases geographically-partitioned according to the number of shards includes providing the number of databases partitioned according to the first three characters of the generated geo-hash value.

4. The computer implemented method of claim 1, wherein querying the identified database server for mobile resource information includes querying the identified database server using the first three characters of the generated geo-hash value.

5. The computer implemented method of claim 1, wherein querying the identified database server for mobile resource information includes querying according to a URL for one of:

a popularity ranking, a distance ranking, and a similarity ranking.

6. The computer implemented method of claim 5, wherein the method further includes aggregating query results based on a number of views, a set of tags, and a number of geographic locations, a set of titles from a shard table in the identified database server.

7. The computer implemented method of claim 1, wherein the method further includes filtering query results according to a tag identifier, a user identifier, a referenced URL identifier, a title, and a regular expression identifier.

8. A memory including a data structure, the data structure organized to include information resident in a number of databases geographically-partitioned (geo-partitioned) according to a number of shards, the data structure including:

a geographically-hashed (geo-hash) look up table containing a number of geo-hash ranges associated with respective database server information, wherein the geo-partitioned databases are partitioned into geographic areas for shards such that search queries limited to a particular geographic location can touch between one to four shards, the geo-hash look up table to receive a function call including a geographic location identifier and a client network node identifier parameter to the function call and link the client network node to a particular server node and an associated geo-partitioned database containing the geo-hash ranges in a shard data object table; and

a plurality of user data object tables, each user data object table containing respective user data attributes including user recommendation credit data attributes associated with a particular user, the plurality of user data object tables to link to the particular server node and the associated geo-partitioned database.

9. The memory of claim 8, wherein the data structure is organized to enable instructions to execute and access the particular server node and associated geo-partitioned database based on a first three character values of a geo-hash value.

10. The memory of claim 8, wherein the data structure is organized to enable instructions to execute and query the geo-hash look up table according to a URL for one of:

a popularity ranking, a distance, and a similarity ranking.

11. The memory of claim 8, wherein the data structure is organized to enable instructions to execute and link the particular user to the particular server node and the particular geo-partitioned database and index, by URL, a comment from the particular user, regarding a mobile print location (MPL), based on the geographic location identifier, a number of views, a user identifier, a tag identifier, and a referenced URL.

12. The memory of claim 8, wherein each shard data object table contains a number of data attributes including:

a shard data attribute;

a user data attribute;

a geographic location data attribute; and

a URL data attribute.

13. A memory including a data structure, the data structure organized to include information resident in a number of databases geographically-partitioned (geo-partitioned) according to a number of shards, the data structure including:

a geographically-hashed (geo-hash) look up table containing a number of geo-hash ranges associated with respective database server information, wherein the geo-partitioned databases are partitioned into geographic areas for shards such that search queries limited to a particular geographic location can touch between one to four shards, the geo-hash look up table linking the client network node to a database server containing the geo-hash ranges in a shard data object table containing a tag attribute, a geographic location attribute, and a title attribute; and

a plurality of user data object tables, each user data object table containing respective user data attributes including user recommendation credit data attributes associated with a particular user, the plurality of user data object tables to link to the database server.

14. The memory of claim 13, wherein the shard data object table includes:

a bid data attribute;

a referenced URL data attribute; and

a view attribute.

15. The memory of claim 13, wherein the data structure includes:

a note object table logically linked to the particular user;

a reward object table physically linked to the particular user;

a friend object table physically linked to the particular user;

an advertisement object table logically and physically linked to the shard object table;

a URL object table physically linked to the advertisement object table; and

an advertisement log object table logically linked to the advertisement object table.