Techniques for capturing data sets

A computer-implemented method for capturing a representation of data, comprising:

• under control of one or more computer systems configured with executable instructions,
• receiving an instruction from a client to capture the representation of the data based at least in part on a capture policy received from the client;
• suspending, based at least in part on the instruction, acknowledgment of a data operation, the data operation including manipulation of a distributed data set;
• and
• enabling a device to capture at least a portion of the data based at least in response to the suspension of the acknowledgment of the data operation, the capture further based at least in part on the capture policy.

A system operable to direct capture of a representation of data, comprising:

• one or more processors;
• and
• memory including executable instructions that, when executed by the one or more processors, cause the system to:
• receive an instruction from a client device to capture the representation of the data at a point in time, the point in time based at least in part on a capture policy received from the client device;
• and
• implement a plurality of servers to perform data operations in a plurality of partitions;
• enable a storage interface to block acknowledgment of completion of an activity in connection with the data;
• and
• insert at least one token into a request stream from the storage interface to the plurality of servers, the at least one token being an indication to each server that receives the at least one token to perform a capture of a partition of the plurality of partitions based at least in part on the capture policy.