Systems and methods for relocating data in a persistent storage device
In some embodiments, there is provided a system which allows data to be received into a placement intelligence. After the data is analyzed, the data is written to a persistent storage device. Subsequently, the data may be written. Periodically, self-optimization may occur to improve read speeds or other metrics.
1. A method for relocating data in a persistent storage device comprising an array of drives, wherein the array of drives comprises multiple solid-state drives, the method comprising:
determining a condition for data relocation is met;
determining metadata, wherein determining the metadata comprises determining an expected read pattern; and
sequentially relocating the data, based on the metadata, from a first physical location in the persistent storage device to a second physical location different from the first physical location in the persistent storage device, the sequential relocation comprising:
reading data matching the expected read pattern into memory; and
distributing the data matching the expected read pattern from the memory across multiple dies on the multiple solid-state drives.
2. The method of claim 1 , wherein relocating the data improves read performance.
3. The method of claim 1 , wherein determining condition for data relocation is met is based on at least one of a number of gaps on the persistent storage device exceeding a first threshold; an activity level of the persistent storage device falling below a second threshold; a free capacity of the persistent storage device falling below a third threshold; and a time period having passed.
4. The method of claim 1 , wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
sequentially writing the data matching the expected read pattern from the memory to a contiguous region in the persistent storage device.
5. The method of claim 1 , wherein determining the metadata further comprises determining a priority value.
6. The method of claim 5 , wherein the array of drives further comprises an array of hard disk drives, and wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
reading data for which the priority value was determined into the memory; and
sequentially writing the data for which the priority value was determined from the memory to a fast zone on the array of hard disk drives.
7. The method of claim 5 , wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
reading data for which the priority value was determined into the memory; and
sequentially writing the data for which the priority value was determined from the memory to the multiple solid-state drives.
8. The method of claim 1 , further comprising updating a mapping from a virtual location to the first physical location to map between the virtual location and the second physical location.
9. A storage system comprising:
a memory;
a persistent storage device comprising an array of drives, wherein the array of drives comprises multiple solid-state drives; and
a storage controller configured to perform steps comprising:
determining a condition for data relocation is met;
determining metadata, wherein determining the metadata comprises determining an expected read pattern; and
sequentially relocating the data, based on the metadata, from a first physical location in the persistent storage device to a second physical location different from the first physical location in the persistent storage device, the sequential relocation comprising:
reading data matching the expected read pattern into the memory; and
distributing the data matching the expected read pattern from the memory across multiple dies on multiple solid-state drives.
10. The storage system of claim 9 , wherein relocating the data improves read performance.
11. The storage system of claim 9 , wherein determining the condition for data relocation is met is based on at least one of a number of gaps on the persistent storage device exceeding a first threshold; an activity level of the persistent storage device falling below a second threshold; a free capacity of the persistent storage device falling below a third threshold; and a time period having passed.
12. The storage system of claim 9 , wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
sequentially writing the data matching the expected read pattern from the memory to a contiguous region in the persistent storage device.
13. The storage system of claim 9 , wherein determining the metadata further comprises determining a priority value.
14. The storage system of claim 13 , wherein the array of drives further comprises an array of hard disk drives, and wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
reading data for which the priority value was determined into the memory; and
sequentially writing the data for which the priority value was determined from the memory to a fast zone on the array of hard disk drives.
15. The storage system of claim 13 , wherein sequentially relocating the data, based on the metadata, from the first physical location in the persistent storage device to the second physical location in the persistent storage device further comprises:
reading data for which the priority value was determined into the memory; and
sequentially writing the data for which the priority value was determined from the memory to the multiple solid-state drives.
16. The storage system of claim 9 , wherein the storage controller is further configured to update a mapping from a virtual location to the first physical location to map between the virtual location and the second physical location.