DISTRIBUTING USER INFORMATION ACROSS REPLICATED SERVERS

摘要

Online systems store information describing a large number of users in order to process requests accessing the user information. The user information is distributed across multiple servers. The distribution is performed so that the information is available even if one or more servers fail. The user information is distributed across a first set of servers and a second copy of the user information is distributed across a second set of servers. The user information from each server of the first set is uniformly distributed across multiple servers from the second set, for example, using random distribution, round robin strategy, or any other strategy that uniformly distributes the information across a given set of processors. Requests previously directed to a failed server are redistributed across multiple servers thereby load balancing the processing of these requests.

主权项

1. A computer-implemented method comprising:
receiving, by an online system, user profiles describing users of a system; storing a first copy of user profiles in a first plurality of servers, such that a set of user profiles is mapped to each server of the first plurality of servers; for each server from the first plurality, the server storing a set of user profiles:
mapping the set of user profiles to a second plurality of servers, such that a subset of user profiles from the set is mapped to each server of the second plurality;storing a second copy of the set of user profiles in the second plurality of servers according to the mapping; receiving information describing failure of a server from the first plurality; receiving requests for information stored on the failed server, each request identifying a user profile stored on the failed server; and for each request identifying a user profile, identifying a server storing a second copy of the user profile and redirecting the request to the identified server.