Organizing data in a distributed storage system

A distributed storage system, comprising:

• a plurality of zones, each zone including a plurality of computer servers and each computer server including one or more processors and memory;
• a plurality of network links that connects two respective zones, wherein data within the distributed storage system is associated with a plurality of accounts and divided into a plurality of groups, each group including a plurality of splits and each split being associated with a respective account;
• and
• a plurality of program modules, wherein the plurality of program modules are stored in the memory of a respective computer server and to be executed by the one or more processors of the respective computer server, and the plurality of program modules include instructions for:
• dividing data associated with a first account and data associated with a second account into a first set of splits and a second set of splits, respectively, wherein the first account and the second account have different data replication policies;
• adding each of the first set of splits and the second set of splits to a respective group of splits, wherein there is no group including a member of the first set of splits and a member of the second set of splits;
• creating a predefined number of replicas for each group of splits that includes a member of the first set of splits and allocating the replicas within the distributed storage system in accordance with a data replication policy associated with the first account;
• and
• creating a predefined number of replicas for each group of splits that includes a member of the second set of splits and allocating the replicas within the distributed storage system in accordance with a data replication policy associated with the second account.

A non-transitory computer readable storage medium storing one or more program modules configured for execution by a computer server that includes one or more processors and memory and is part of a distributed storage system, wherein the distributed storage system is configured for dividing data associated with a plurality of accounts into a plurality of groups, each group including a plurality of splits and each split being associated with a respective account, the one or more programs comprising instructions for:

• dividing data associated with a first account and data associated with a second account into a first set of splits and a second set of splits, respectively, wherein the first account and the second account have different data replication policies;
• adding each of the first set of splits and the second set of splits to a respective group of splits, wherein there is no group including a member of the first set of splits and a member of the second set of splits;
• creating a predefined number of replicas for each group of splits that includes a member of the first set of splits and allocating the replicas within the distributed storage system in accordance with a data replication policy associated with the first account;
• and
• creating a predefined number of replicas for each group of splits that includes a member of the second set of splits and allocating the replicas within the distributed storage system in accordance with a data replication policy associated with the second account.

A computer-implemented method, comprising:

• at a computer server including one or more processors and memory:
• providing a distributed storage system, wherein data within the distributed storage system is associated with a plurality of accounts and divided into a plurality of groups, each group including a plurality of splits and each split being associated with a respective account;
• identifying a first split associated with an account, wherein the first split is a member of a first group of splits and the first split has a data range parameter indicating that more data can be added to the first split and a split size indicating an actual amount of data in the first split;
• in response to determining that the split size of the first split has reached a predefined limit:
• updating the data range parameter in the first split so that no additional data can be added to the first split;
• creating a second split for the account, wherein the second split includes a data range parameter indicating that more data can be added to the second split;
• adding the second split to a second group of splits;
• and
• replicating the second group of splits within the distributed storage system in accordance with a data replication policy associated with the account.