Shadow address space for sharing storage
A computing device includes a persistent storage and a processor. The processor includes a local storage. The local storage includes blocks and an address space. The address space includes a first portion of entries that specify blocks of the local storage and a second portion of entries that specify blocks of the remote data storage. The processor obtains data for storage and makes a determination that the data cannot be stored in the local storage. In response to the determination, the processor stores the data in the remote storage using the second portion of entries.
1. A computing device, comprising:
a persistent storage comprising:
a local storage comprising blocks, and
an address space comprising:
a first plurality of entries that specify the blocks of the local storage, and
a second plurality of entries that indicate blocks of a remote storage; and
a processor programmed to:
obtain data for storage;
make a first determination that the data cannot be stored in the local storage, wherein at a time the first determination is made there is not sufficient space to store the data in the local storage; and
in response to the first determination, store the data in the remote storage using the second plurality of entries, wherein storing the data in the remote storage using the second plurality of entries comprises:
allocating a block of the remote storage indicated by an entry of the second plurality of entries, wherein the entry of the second plurality of entries comprises: a computing device identifier that identifies a second computing device that hosts the remote storage and a logical address identifier that specifies an entry of an address space of the second computing device;
locking the allocated block by updating a use state of the entry of the second plurality of entries to locked;
writing a portion of the data to the locked allocated block;
unlocking the locked allocated block after writing the portion of the data by updating the use state of the entry of the second plurality of entries to unlocked;
after the unlocking,
make a second determination that there is available space in the local storage;
in response to the second determination, identify a block of the remote storage that includes the portion of the data; and
migrate the identified block from the remote storage to the local storage.
2. The computing device of claim 1 , wherein each entry of the second plurality of entries does not specify a block of the remote storage.
3. The computing device of claim 1 , wherein the entry of the address space of the second computing device specifies a block of the blocks of the remote storage.
4. The computing device of claim 1 , wherein the entry of the second plurality of entries further comprises:
a block status.
5. The computing device of claim 4 , wherein the block status specifies a state of the remote storage hosting a block specified by the entry of the second plurality of entries is available.
6. The computing device of claim 1 , wherein the use state specifies whether a block specified by the entry of the second plurality of entries is available.
7. The computing device of claim 1 , wherein the processor is further programmed to:
obtain second data for storage;
make a third determination that the second data can be stored in the local storage; and
in response to the third determination, store the second data in the local storage using the first plurality of entries.
8. The computing device of claim 7 , wherein each entry of the first plurality of entries specifies a block identifier associated with a respective block of the blocks of the local storage.
9. The computing device of claim 1 , wherein migrating the identified block from the remote storage to the local storage comprises:
obtaining a copy of identified block from the remote storage;
storing the copy of the identified block in a block of the local storage;
deleting the identified block from the remote storage; and
updating the address space based on the copy and the deletion.
10. The computing device of claim 9 , wherein updating the address space based on the copy and the deletion comprises:
deleting a computing device identifier and a logical address identifier associated with a second entry of the second plurality of entries that specifies the identified block; and
adding a block identifier that specifies the block of the local storage.
11. The computing device of claim 1 , wherein each block of the plurality of blocks of the local storage comprises a predetermined number of bytes.
12. A method of operating a computing device, comprising:
obtaining, by the computing device, data for storage;
making, by the computing device, a first determination that the data cannot be stored in a local storage comprising blocks, wherein the blocks of the local storage are specified by a first plurality of entries of an address space of the computing device and wherein at a time the first determination is made there is not sufficient space to store the data in the local storage; and
in response to the first determination, storing, by the computing device, the data in a remote storage using a second plurality of entries that indicate blocks of the remote storage, wherein storing the data in the remote storage using the second plurality of entries comprises:
allocating a block of the remote storage indicated by an entry of the second plurality of entries, wherein the entry of the second plurality of entries comprises: a computing device identifier that identifies a second computing device that hosts the remote storage and a logical address identifier that specifies an entry of an address space of the second computing device;
locking the allocated block by updating a use state of the entry of the second plurality of entries to locked;
writing a portion of the data to the locked allocated block;
unlocking the locked allocated block after writing the portion of the data by updating the use state of the entry of the second plurality of entries to unlocked;
after the unlocking,
making a second determination that there is available space in the local storage;
in response to the second determination, identifying a block of the remote storage that includes the portion of the data; and
migrating the identified block from the remote storage to the local storage.
13. The method of claim 12 , further comprising:
obtaining, by the computing device, second data for storage;
making a third determination, by the computing device, that the second data can be stored in the local storage; and
in response to the third determination, storing, by the second computing device, the second data in the local storage using the first plurality of entries.
14. 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 operating a computing device, the method comprising:
obtaining, by the computing device, data for storage;
making, by the computing device, a first determination that the data cannot be stored in a local storage comprising blocks, wherein the blocks of the local storage are specified by a first plurality of entries of an address space of the computing device and, wherein at a time the first determination is made there is not sufficient space to store the data in the local storage; and
in response to the first determination, storing, by the computing device, the data in a remote storage using a second plurality of entries that indicate blocks of the remote storage, wherein storing the data in the remote storage using the second plurality of entries comprises:
allocating a block of the remote storage indicated by an entry of the second plurality of entries, wherein the entry of the second plurality of entries comprises a computing device identifier that identifies a second computing device that hosts the remote storage and a logical address identifier that specifies an entry of an address space of the second computing device;
locking the allocated block by updating a use state of the entry of the second plurality of entries to locked;
writing a portion of the data to the locked allocated block;
unlocking the locked allocated block after writing the portion of the data by updating the use state of the entry of the second plurality of entries to unlocked;
after the unlocking,
make a second determination that there is available space in the local storage;
in response to the second determination, identify a block of the remote storage that includes the portion of the data; and
migrate the identified block from the remote storage to the local storage.
15. The non-transitory computer readable medium comprising computer readable program code of claim 14 , wherein the method further comprises:
obtaining, by the computing device, second data for storage;
making a second determination, by the computing device, that the second data can be stored in the local storage; and
in response to the second determination, storing, by the second computing device, the second data in the local storage using the first plurality of entries.