Systems and methods for providing nonlinear journaling
View Patent ↗In one embodiment, systems and methods are provided for nonlinear journaling. In one embodiment, groups of data designated for storage in a data storage unit are journaled into persistent storage. In one embodiment, the journal data is recorded nonlinearly. In one embodiment, a linked data structure records data and data descriptors in persistent storage.
1. A system for atomically updating a journal for a data storage unit, comprising:
persistent memory; and
a program module configured to update a journal located in said persistent memory with atomic operations, wherein the updates to the journal include journal data that is too large to be recorded in a single atomic operation, wherein an atomic operation is an operation that either commits or aborts;
wherein the journal is not in an inconsistent state following a write failure to update the journal, wherein an inconsistent state is a state in which it is not known whether the data was actually written before the write failure.
2. The system of claim 1 , wherein the program module configured to update comprises being configured to update by linking at least one of the following into the journal: data and metadata.
3. The system of claim 1 , wherein the program module configured to update comprises being configured to update by unlinking at least one of the following from the journal: data and metadata.
4. The system of claim 1 , wherein the program module configured to update comprises being configured to modify metadata fields.
5. A computer implemented method of building atomically a journal for a data storage unit, comprising:
recording data in persistent memory, the data being too large to be recorded in a single atomic operation, wherein an atomic operation is an operation that either commits or aborts; and
building a journal in the persistent memory, the journal comprising the data, wherein the journal is built by linking into the journal at least one of the following: the data and metadata associated with the data;
wherein the journal is built with atomic operations such that the journal does not comprise only a portion of the data.
6. The system of claim 1 , wherein journaled data is designated to be written to a location on a storage device associated with the data storage unit, said data storage device being one of a plurality of storage devices associated with said data storage unit.
7. The system of claim 6 , wherein data that is journaled corresponds to a plurality of transactions, and the data to be written for each of the plurality of transactions is to be distributed among two or more of the plurality of storage devices.
8. The system of claim 6 , wherein at least one of data and metadata is unlinked in response to a communication from the storage device that the data has been stored at the location.
9. The system of claim 8 , wherein the communication indicates that a data cache on the storage device has been flushed.
10. The system of claim 6 , wherein at least one of data and metadata is unlinked based on a determination that the data is the least recently used data corresponding to the storage device.
11. The computer implemented method of claim 5 , wherein the journal is built using one of the following: linked list, two-dimensional linked list, linked list of multiple-dimensional linked lists, binary tree, and b tree.
12. The computer implemented method of claim 5 , further comprising unlinking at least one of the following from the journal: data and metadata.
13. The computer implemented method of claim 5 , wherein journaled data is designated to be written to a location on a storage device associated with the data storage unit.
14. The computer implemented method of claim 13 , further comprising unlinking data from the journal in response to a communication from the storage device that the data has been stored at the location.
15. A computer-readable storage medium having instructions stored thereon for implementing, when the instructions are executed, a journal for a data storage unit, the instructions comprising:
recording data in persistent memory, the data being too large to be recorded in a single atomic operation, wherein an atomic operation is an operation that either commits or aborts; and
linking into a journal in the persistent memory at least one of the following: the data and metadata associated with the data;
wherein the linking comprises an atomic operation such that the journal does not comprise only a portion of the data.
16. The computer-readable storage medium of claim 15 , wherein the journal includes one of the following: linked list, two-dimensional linked list, linked list of multiple-dimensional linked lists, binary tree, and b tree.
17. The computer-readable storage medium of claim 15 , the instructions further comprising unlinking at least one of the following from the journal: data and metadata.
18. The computer-readable storage medium of claim 15 , wherein journaled data is designated to be written to a location on a storage device associated with the data storage unit.
19. The computer-readable storage medium of claim 18 , the instructions further comprising unlinking data from the journal in response to a communication from the storage device that the data has been stored at the location.
20. A computer implemented method for journaling data to be written to a data storage unit, comprising:
storing a plurality of data groups in persistent storage, the data groups associated with data to be written to the data storage unit, wherein at least one of the data groups is too large to be recorded in a single atomic operation; and
adding the data groups to a journal stored in persistent storage, wherein adding the data groups to the journal comprises storing information about the location of each of the data groups in the journal, wherein the information about the location of each of the data groups is of a size that is capable of being written in a single atomic operation;
wherein each data group is added to the journal in a single atomic operation such that the journal is not in an inconsistent state following a write failure to update the journal.
21. A system for journaling data to be written to a data storage unit, comprising:
a computer-readable storage;
a processor configured to access the computer-readable storage; and
a journal module configured to:
add a first journal entry to a journal, wherein the journal comprises a plurality of journal entries, wherein each journal entry comprises a portion that is of a size at which the portion may be written in a single atomic operation;
add a second journal entry to the journal, wherein the second journal entry of the journal comprises a first portion that is of a size at which the portion may be written in a single atomic operation and a second portion that is too large to be written in a single atomic operation;
add a third journal entry to the journal; and
remove the second journal entry from the journal in a single atomic operation, wherein removing the second journal entry comprises performing an atomic write to a journal entry portion that is of a size at which the portion may be written in a single atomic operation, such that immediately after the atomic write the first portion of the second journal entry and the second portion of the second journal entry have been removed from the journal as a consequence of the single atomic operation.
22. The system of claim 21 , wherein journaled data is designated to be written to a location on a storage device associated with the data storage unit, wherein the second journal entry is removed in response to a communication from the storage device that the data has been stored at the location.