Physical enumeration of a distributed namespace in a distributed file system

A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for recovery of a portion of a backup image, the method comprising:

• obtaining, by a garbage collection (GC) agent of a computing node in a distributed namespace (DNS) system, a request for processing a set of pages in a file system,
• wherein each page in the set of pages is associated with one of a plurality of snapshots,
• wherein each of the plurality of snapshots is either a dataless or dataful snapshot;
• in response to the request:
• performing a page processing on each page in the set of pages based on an association to a dataless or dataful snapshot of each page to obtain a subset of the pages;
• and
• performing, using a data manager of the computing node, a physical enumeration of the subset of the pages to obtain a managed storage of the DNS system,
• wherein the managed storage is cleaned based on data not associated with the subset of pages.

A system comprising:

• a computing node in a distributed namespace (DNS) system, operating on a processor;
• and
• memory comprising instructions, which when executed by the processor, perform a method comprising:
• obtaining, by a garbage collection (GC) agent of the computing node, a request for processing a set of pages in a file system,
• wherein each page in the set of pages is associated with one of a plurality of snapshots,
• wherein each of the plurality of snapshots is either a dataless or dataful snapshot;
• in response to the request:
• performing a page processing on each page in the set of pages based on an association to a dataless or dataful snapshot of each page to obtain a subset of the pages;
• and
• performing, using a data manager of the computing node, a physical enumeration of the subset of the pages to obtain a managed storage of the DNS system,
• wherein the managed storage is cleaned based on data not associated with the subset of pages.