Two level addressing in storage clusters
Digital objects are stored and accessed within a fixed content storage cluster by using a page mapping table and a pages index. A stream is read from the cluster by using a portion of its unique identifier as a key into the page mapping table. The page mapping table indicates a node holding a pages index indicating where the stream is stored. A stream is written by storing the stream on any suitable node and then updating a pages index stored within the cluster responsible for knowing the location of digital objects having unique identifiers that fall within a particular address range. The cluster recovers from a node failure by first replicating streams from the failed node and reallocating a page mapping table to create a new pages index. The remaining nodes send records of the unique identifiers corresponding to objects they hold to the new pages index.
1. A method of recovering from a node failure within a storage cluster, said method comprising:
detecting within said storage cluster that a first computer node out of a plurality of computer nodes is unavailable, said storage cluster including stored digital objects in persistent storage, each of said digital objects being identified by a unique identifier (UID), and wherein said unique identifiers having ranges of possible values, wherein a first UID range being mapped to said first computer node;
replacing said mapping of said first UID range to said first computer node with a mapping of said first UID range to a second computer node within said storage cluster different from said first computer node;
causing each computer node within said storage cluster other than said first computer node to determine a set of unique identifiers falling within said first UID range indicating digital objects stored upon said each computer node;
receiving, at said second computer node from said each computer node, said sets that identify said unique identifiers and said each computer nodes where the digital objects corresponding to said unique identifiers are stored; and
updating a disk index in RAM on said second computer node to indicate on which of said each computer nodes said unique identifiers are stored, said disk index including unique identifiers of digital objects stored on said second computer node and locations on a disk of said second computer node of said digital objects that are stored on said second computer node.
2. The method as recited in claim 1 wherein said storage cluster is a write-once read-many (WORM) fixed-content storage cluster.
3. The method as recited in claim 1 wherein said steps of claim 1 do not require a disk access.
4. The method as recited in claim 1 wherein said first UID range being mapped to said first computer node in a table of said storage cluster, and wherein said replacing occurs in said table.
5. A non-transitory computer-readable storage medium comprising computer code for recovering from a node failure in a storage cluster, said computer code of said non-transitory computer-readable storage medium executable by a machine to perform the following:
detecting within said storage cluster that a first computer node out of a plurality of computer nodes is unavailable, said storage cluster including stored digital objects in persistent storage, each of said digital objects being identified by a unique identifier (UID), and wherein said unique identifiers having ranges of possible values, wherein a first UID range being mapped to said first computer node;
replacing said mapping of said first UID range to said first computer node with a mapping of said first UID range to a second computer node within said storage cluster different from said first computer node;
causing each computer node within said storage cluster other than said first computer node to determine a set of unique identifiers falling within said first UID range indicating digital objects stored upon said each computer node;
receiving, at said second computer node from said each computer node, said sets that identify said unique identifiers and said each computer nodes where the digital objects corresponding to said unique identifiers are stored; and
updating a disk index in RAM on said second computer node to indicate on which of said each computer nodes said unique identifiers are stored, said disk index including unique identifiers of digital objects stored on said second computer node and locations on a disk of said second computer node of said digital objects that are stored on said second computer node.
6. The non-transitory computer-readable storage medium as recited in claim 5 wherein said storage cluster is a write-once read-many (WORM) fixed-content storage cluster.
7. The non-transitory computer-readable storage medium as recited in claim 5 wherein said steps of claim 5 do not require a disk access.
8. The non-transitory computer-readable storage medium as recited in claim 5 wherein said first UID range being mapped to said first computer node in a table of said storage cluster, and wherein said replacing occurs in said table.