Abstract: A system, method, and machine-readable storage medium for recovering data in a distributed storage system are provided. In some embodiments, the method includes identifying a failing storage device of a first storage node having an inaccessible data segment. When it is determined that the inaccessible data segment cannot be recovered using a first data protection scheme, a first chunk of data associated with the inaccessible data segment is identified and a group associated with the first chunk of data is identified. A second chunk of data associated with the group is selectively retrieved from a second storage node such that data associated with an accessible data segment of the first storage node is not retrieved. The inaccessible data segment is recovered by recovering the first chunk of data using a second data protection scheme and the second chunk of data.

Abstract: A system, method, and machine-readable storage medium for recovering data in a distributed storage system are provided. In some embodiments, the method includes identifying a failing storage device of a first storage node having an inaccessible data segment. When it is determined that the inaccessible data segment cannot be recovered using a first data protection scheme, a first chunk of data associated with the inaccessible data segment is identified and a group associated with the first chunk of data is identified. A second chunk of data associated with the group is selectively retrieved from a second storage node such that data associated with an accessible data segment of the first storage node is not retrieved. The inaccessible data segment is recovered by recovering the first chunk of data using a second data protection scheme and the second chunk of data.

Abstract: Technology is disclosed for subpartitioning a namespace region. In various embodiments, the technology creates at least two subpartitions from a partitioned namespace, wherein the partitioned namespace corresponds to at least two different name nodes of the large scale data storage service; and stores data corresponding to each subpartition as a separate file, e.g., so that it can be easily mounted by an operating system executed by a different computing device.

Abstract: Technology is disclosed for partitioning a namespace of a large scale data storage service. In various embodiments, the technology receives a file path; employs a hashing function to produce a hash value for the received file path; and identifies a name node handling the produced hash value, wherein the name node is configured to communicate with one or more data nodes for handling file requests on the file path.

Abstract: Technology is disclosed for partitioning a namespace of a large scale data storage service. In various embodiments, the technology receives a file path; employs a hashing function to produce a hash value for the received file path; and identifies a name node handling the produced hash value, wherein the name node is configured to communicate with one or more data nodes for handling file requests on the file path.

Abstract: Technology is disclosed for subpartitioning a namespace region. In various embodiments, the technology creates at least two subpartitions from a partitioned namespace, wherein the partitioned namespace corresponds to at least two different name nodes of the large scale data storage service; and stores data corresponding to each subpartition as a separate file, e.g., so that it can be easily mounted by an operating system executed by a different computing device.