Data replication optimization for online social network cloud computing
Systems and methods are provided for optimizing data storage allocation and replication of user data in computing systems. For example, a method includes determining a ranking of each data center in the computing system for a given user of the computing system, and selecting a first data center to store a primary copy of user data, and a second data center to store a replica copy of the user data, based at least in part on the determined rankings of the data centers for the given user. The data centers are ranked using a ranking function which includes an access ratio that is determined as ratio of (i) a number of times that each of a total number of users in the given data center have accessed shared data of the given user to (ii) a number of data uploads that the given user has made to the computing system.
1. A method, comprising:
storing user data in a plurality of data centers of a computing system, wherein the user data comprises shared data that is shared between users of the computing system; and
performing a storage allocation and replication process to optimize storage of the user data in the data centers of the computing system, wherein performing a storage allocation and replication process comprises:
determining a ranking of each data center in the computing system for at least one user of the computing system, wherein determining a ranking comprises ranking a given data center as a function of access ratio values of friends of the at least one user, whose data is stored in the given data center, wherein an access ratio value of a given friend of the at least one user comprises a ratio which is based at least in part on (i) a number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user and (ii) a number of data uploads of shared data that the at least one user has made to the computing system; and
selecting a first data center to store a primary copy of user data of the at least one user, and selecting a second data center to store a replica copy of the user data of the at least one user, based at least in part on the determined rankings of the data centers for the at least one user, to reduce an average access delay experienced by the friends of the at least one user in accessing the shared data of the at least one user;
wherein one or more of the storing and performing steps are performed under control of at least one processing device.
2. The method of claim 1 , wherein selecting a first data center comprises selecting a highest ranked data center to store the primary copy of the user data, and wherein selecting a second data center comprises selected a next highest ranked data center to store the replica copy of the user data of the at least one user.
3. The method of claim 1 , further comprising migrating the primary copy of the user data of the at least one user to the selected first data center of the computing system.
4. The method of claim 1 , wherein the computing system comprises a social network.
5. The method of claim 1 , wherein ranking a given data center further comprises dividing said access ratio by a total number of users having user data stored in the given data center.
6. The method of claim 1 , wherein said number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user is based on a number of times that said shared data was accessed by the given friend in a given period of time prior to performing the storage allocation and replication process.
7. The method of claim 1 , wherein storing user data in a plurality of data centers of a computing system comprises initially storing the user data of a new user in one of the data centers that is closest to the new user.
8. 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 implement a method comprising:
storing user data in a plurality of data centers of a computing system, wherein the user data comprises shared data that is shared between users of the computing system; and
performing a storage allocation and replication process to optimize storage of the user data in the data centers of the computing system, wherein performing a storage allocation and replication process comprises:
determining a ranking of each data center in the computing system for at least one user of the computing system, wherein determining a ranking comprises ranking a given data center as a function of access ratio values of friends of the at least one user, whose data is stored in the given data center, wherein an access ratio value of a given friend of the at least one user comprises a ratio which is based at least in part on (i) a number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user and (ii) a number of data uploads of shared data that the at least one user has made to the computing system; and
selecting a first data center to store a primary copy of user data of the at least one user, and selecting a second data center to store a replica copy of the user data of the at least one user, based at least in part on the determined rankings of the data centers for the at least one user, to reduce an average access delay experienced by the friends of the at least one user in accessing the shared data of the at least one user;
wherein one or more of the storing and performing steps are performed under control of at least one processing device.
9. The article of manufacture of claim 8 , wherein selecting a first data center comprises selecting a highest ranked data center to store the primary copy of the user data, and wherein selecting a second data center comprises selected a next highest ranked data center to store the replica copy of the user data of the at least one user.
10. The article of manufacture of claim 8 , further comprising migrating the primary copy of the user data of the at least one user to the selected first data center of the computing system.
11. The article of manufacture of claim 8 , wherein the computing system comprises a social network.
12. The article of manufacture of claim 8 , wherein ranking a given data center further comprises dividing said access ratio by a total number of users having user data stored in the given data center.
13. The article of manufacture of claim 8 , wherein said number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user is based on a number of times that said shared data was accessed by the given friend in a given period of time prior to performing the storage allocation and replication process.
14. The article of manufacture of claim 8 , wherein storing user data in a plurality of data centers of a computing system comprises initially storing the user data of a new user in one of the data centers that is closest to the new user.
15. An apparatus of a computing system, the computing system comprising a plurality of data centers for storing user data, wherein the user data comprises shared data that is shared between users of the computing system, the apparatus comprising:
a processor; and
a memory coupled to the processor, wherein the memory comprises program instructions that are executed by the processor to perform a storage allocation and replication process to optimize storage of the user data in the data centers of the computing system, wherein the storage allocation and replication process comprises:
determining a ranking of each data center in the computing system for at least one user of the computing system, wherein determining a ranking comprises ranking a given data center as a function of access ratio values of friends of the at least one user, whose data is stored in the given data center, wherein an access ratio value of a given friend of the at least one user comprises a ratio which is based at least in part on (i) a number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user and (ii) a number of data uploads of shared data that the at least one user has made to the computing system; and
selecting a first data center to store a primary copy of user data of the at least one user, and selecting a second data center to store a replica copy of the user data of the at least one user, based at least in part on the determined rankings of the data centers for the at least one user, to reduce an average access delay experienced by the friends of the at least one user in accessing the shared data of the at least one user.
16. The apparatus of claim 15 , wherein selecting a first data center comprises selecting a highest ranked data center to store the primary copy of the user data, and wherein selecting a second data center comprises selected a next highest ranked data center to store the replica copy of the user data of the at least one user.
17. The apparatus of claim 15 , wherein the memory further comprises program instructions that are executed by the processor to migrate the primary copy of the user data of the at least one user to the selected first data center of the computing system.
18. The apparatus of claim 15 , wherein the computing system comprises a social network.
19. The apparatus of claim 15 , wherein ranking a given data center further comprises dividing said access ratio by a total number of users having user data stored in the given data center.
20. The apparatus of claim 15 , wherein said number of times that the given friend whose data is stored in the given data center has accessed the shared data of the at least one user is based on a number of times that said shared data was accessed by the given friend in a given period of time prior to performing the storage allocation and replication process.