Cloud computing environment with replication system configured to reduce latency of data read access

An article of manufacture comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code is executable by one or more processors to perform a method:

• configuring a data replication system to execute in a plurality of sites of a computing system, wherein the plurality of sites comprises a primary site and at least one replica site, wherein the primary site comprises a primary storage to store primary data of a given application, and wherein the at least one replica site comprises a replica storage to store a replica of the primary data;
• configuring a routing decision system to execute in a distributed manner in each site of the plurality of sites of the computing system, and to communicate with the data replication system to acquire information which is maintained by the routing decision system in a database of routing decision data to coordinate routing of data access requests, wherein at least a portion of the database of routing decision data is maintained in a distributed manner in each site of the plurality of sites of the computing system;
• receiving, by the routing decision system, a data access request from the given application executing on a client device in a geographical location;
• responsive to the data access request comprising a data write request, the routing decision system routing the data write request to the primary site to write data to the primary storage;
• responsive to the data access request comprising a data read request, the routing decision system determining which site of the plurality of sites of the computing system for handling the data read request, wherein determining comprises utilizing the database of routing decision data to determine which given site among the plurality of sites including the primary site and the at least one replica site provides a lowest read access latency for accessing data from the given site to respond to the data read request of the given application based, at least in part, on which site among the plurality of sites resides in a geographical location which is closest in physical distance to the geographical location of the client device which executes the given application;
• routing, by the routing decision system, the data read request to the at least one replica site, in response to determining that the at least one replica site provides a lower read access latency for accessing data from the replica storage as compared to accessing data from the primary storage of the primary site based, at least in part, on a determination that the geographical location of the at least one replica site is closest in physical distance to the geographical location of the client device which executes the given application;
• and
• routing, by the routing decision system, the data read request to the primary site, in response to determining that the primary site provides a lower read access latency for accessing data from the primary storage as compared to accessing data from the replica storage of the at least one replica site based, at least in part, on a determination that the geographical location of the primary site is closest in physical distance to the geographical location of the client device which executes the given application;
• wherein configuring the routing decision system to execute in a distributed manner in each site of the plurality of sites of the computing system comprises running an application plug-in, which implements functions of the routing decision system, on at least one host server of each site of the plurality of sites of the computing system.

A computing system, comprising:

• a data replication system configured to execute in a plurality of sites of the computing system, wherein the plurality of sites comprises a primary site and at least one replica site, wherein the primary site comprises a primary storage to store primary data of a given application, and wherein the at least one replica site comprises a replica storage to store a replica of the primary data;
• and
• a routing decision system which is configured to execute in a distributed manner in each site of the plurality of sites of the computing system, and to communicate with the data replication system to acquire information which is maintained by the routing decision system in a database of routing decision data to coordinate routing of data access requests, wherein at least a portion of the database of routing decision data is maintained in a distributed manner in each site of the plurality of sites of the computing system, and wherein the routing decision system is configured to execute a process comprising:
• receiving a data access request from the given application executing on a client device in a geographical location;
• responsive to the data access request comprising a data write request, routing the data write request to the primary site to write data to the primary storage;
• responsive to the data access request comprising a data read request, determining which site of the plurality of sites of the computing system for handling the data read request, wherein determining comprises utilizing a database of routing decision data to determine which given site among the plurality of sites including the primary site and the at least one replica site provides a lowest read access latency for accessing data from the given site to respond to the data read request of the given application based, at least in part, on which site among the plurality of sites resides in a geographical location which is closest in physical distance to the geographical location of the client device which executes the given application;
• routing the data read request to the at least one replica site, in response to determining that the at least one replica site provides a lower read access latency for accessing data from the replica storage as compared to accessing data from the primary storage of the primary site based, at least in part, on a determination that the geographical location of the at least one replica site is closest in physical distance to the geographical location of the client device which executes the given application;
• and
• routing the data read request to the primary site, in response to determining that the primary site provides a lower read access latency for accessing data from the primary storage as compared to accessing data from the replica storage of the at least one replica site based, at least in part, on a determination that the geographical location of the primary site is closest in physical distance to the geographical location of the client device which executes the given application;
• wherein the routing decision system is configured to execute in a distributed manner in each site of the plurality of sites of the computing system by running an application plug-in, which implements functions of the routing decision system, on at least one host server of each site of the plurality of sites of the computing system.