Data deletion in a distributed data storage system

A method for removing unused storage units, the method comprising:

• receiving, at a block server, a filter generated by a metadata server on at least a portion of block identifiers that identify storage units currently being referenced by a client address, wherein at least one of the storage units is referenced by multiple client addresses, and wherein the metadata server stores information on which block identifiers are referenced by which client addresses;
• using the filter to identify which storage units being stored on the block server are in use and not in use based upon whether block identifiers that exist on the block server are present in the filter, wherein the block server stores information on where a storage unit is stored on the block server for a block identifier;
• determining a storage unit is in use when a block identifier for that storage unit is found in the filter;
• determining a storage unit is not in use when a block identifier for that storage unit is not found in the filter;
• and
• deleting, from the block server, storage units not in use and associated block identifiers not present in the filter.

A system comprising:

• a metadata server configured to:
• construct a filter on at least a portion of block identifiers that identify storage units, wherein at least one of the storage units is referenced by multiple client addresses, and wherein the metadata server stores information on which block identifiers are referenced by which client addresses;
• a block server, wherein the block server is designated to store storage units associated with different ranges of block identifiers, and wherein the block server is configured to:
• receive the filter that includes block identifiers in the range associated with the block server;
• use the filter to identify which storage units being stored on the block server are in use and not in use based on whether block identifiers that exist on the block server are present in the filter, wherein the block server stores information on where a storage unit is stored on the block server for a block identifier and;
• determine a storage unit is in use when a block identifier for that storage unit is found in the filter;
• determine a storage unit is not in use when a block identifier for that storage unit is not found in the filter;
• and
• delete, from the block server, storage units determined to not bin in use and associated block identifiers not present in the filter, wherein the block server comprises a non-transitory storage medium.