Efficient pre-fetching on a storage system
Managing a cache memory in a storage system includes maintaining a first queue that stores data indictive of the read requests for a particular logical storage unit of the storage system in an order that the read requests are received by the storage system and maintaining a second queue that stores data indictive of the read requests for the particular logical storage unit in a sort order corresponding to page numbers of the read requests, the second queue persisting for a plurality of iterations of read requests. A read request is received and data indicative of the read request is placed in the first queue and in the second queue while maintaining the sort order of the second queue. The second queue is used to determine a prefetch metric that varies according to a number of adjacent elements in the second queue.
1. A method of managing a cache memory in a storage system, comprising:
maintaining a first queue that stores data indictive of the read requests for pages of a particular logical storage unit of the storage system in an order that the read requests are received by the storage system;
maintaining a second queue that stores data indictive of the read requests for the pages of the particular logical storage unit in a sort order corresponding to page numbers of the read requests, the second queue persisting for a plurality of iterations of read requests, wherein the page numbers correspond to a position of the pages within an ordered sequence of pages of the logical storage unit;
receiving a read request for a particular one of the pages of the particular logical storage unit;
placing data indicative of the read request in the first queue;
placing the data indicative of the read request in the second queue while maintaining the sort order of the second queue;
determining a plurality of difference values according to a difference of page numbers between each pair of adjacent elements in the second queue;
determining a prefetch metric value that varies according to a count of a number of times each of the difference values is less than a predetermined value and greater than zero; and
prefetching a plurality of pages having page numbers that come after the particular one of the pages sequentially in response to the prefetch metric value being greater than a predefined value.
2. A method, according to claim 1 , wherein the first queue and the second queue have a fixed size.
3. A method, according to claim 1 , wherein the pre-defined value is based on a look-ahead window.
4. A method, according to claim 1 , wherein the pre-fetch metric value is determined by dividing the count of the number of times each of the difference values is less than a predetermined value and greater than zero by a number of the difference values that are greater than zero.
5. A method, according to claim 1 , wherein a number of elements are discarded from the first queue and the second queue in response to the read request occurring when the queues are full.
6. A method, according to claim 5 , wherein the number of elements that are discarded is 50% of the elements.
7. A method, according to claim 1 , wherein there is a relationship between elements of the first queue and elements of the second queue that facilitates accessing corresponding elements of the second queue by initially accessing elements of the first queue.
8. A non-transitory computer readable medium containing software that manages a cache memory in a storage system that also contains a first queue that stores data indictive of the read requests for pages of a particular logical storage unit of the storage system in an order that the read requests are received by the storage system and a second queue that stores data indictive of the read requests for the pages of the particular logical storage unit in a sort order corresponding to page numbers of the read requests, the second queue persisting for a plurality of iterations of read requests and the page numbers correspond to a position of the pages within an ordered sequence of pages of the logical storage unit, the software comprising:
executable code that receives a read request for a particular one of the pages of the particular logical storage unit;
executable code that places data indicative of the read request in the first queue;
executable code that places the data indicative of the read request in the second queue while maintaining the sort order of the second queue;
executable code that determines a plurality of difference values according to a difference of page numbers between each pair of adjacent elements in the second queue;
executable code that determines a prefetch metric value that varies according to a count of a number of times each of the difference values is less than a predetermined value and greater than zero; and
executable code that prefetches a plurality of pages having page numbers that come after the particular one of the pages sequentially in response to the prefetch metric value being greater than a predefined value.
9. A non-transitory computer readable medium, according to claim 8 , wherein the first queue and the second queue have a fixed size.
10. A non-transitory computer readable medium, according to claim 8 , wherein the pre-defined value is based on a look-ahead window.
11. A non-transitory computer readable medium, according to claim 8 , wherein the pre-fetch metric value is determined by dividing the count of the number of times each of the difference values is less than a predetermined value and greater than zero by a number of the difference values that are greater than zero.
12. A non-transitory computer readable medium, according to claim 8 , wherein a number of elements are discarded from the first queue and the second queue in response to the read request occurring when the queues are full.
13. A non-transitory computer readable medium, according to claim 12 , wherein the number of elements that are discarded is 50% of the elements.
14. A non-transitory computer readable medium, according to claim 8 , wherein there is a relationship between elements of the first queue and elements of the second queue that facilitates accessing corresponding elements of the second queue by initially accessing elements of the first queue.
15. A storage system, comprising:
a plurality of storage units that store data written to the storage system;
a first subset of directors coupled to the storage units;
a second subset of directors that provide data access to the storage system, including data access to write data to the storage system and data access to read data that was previously written to the storage system;
a memory coupled to at least some of the directors, the memory providing a cache memory, and providing a first queue that stores data indictive of read requests for pages of a particular logical storage unit of the storage system in an order that the read requests are received by the storage system and a second queue that stores data indictive of the read requests for the pages of the particular logical storage unit in a sort order corresponding to page numbers of the read requests, the second queue persisting for a plurality of iterations of read requests and the page numbers correspond to a position of the pages within an ordered sequence of pages of the logical storage unit; and
a non-transitory computer readable medium containing software that manages the cache memory, the software having executable code that receives a read request for a particular one of the pages of the particular logical storage unit, executable code that places data indicative of the read request in the first queue, executable code that places the data indicative of the read request in the second queue while maintaining the sort order of the second queue, executable code that determines a plurality of difference values according to a difference of page numbers between each pair of adjacent elements in the second queue, executable code that determines a prefetch metric value that varies according to a count of a number of times each of the difference values is less than a predetermined value and greater than zero, and executable code that prefetches a plurality of pages having page numbers that come after the particular one of the pages sequentially in response to the prefetch metric value being greater than a predefined value.
16. A storage system, according to claim 15 , wherein the first queue and the second queue have a fixed size.
17. A storage system, according to claim 15 , wherein the pre-defined value is based on a look-ahead window.
18. A storage system, according to claim 15 , wherein the pre-fetch metric value is determined by dividing the count of the number of times each of the difference values is less than a predetermined value and greater than zero by a number of the difference values that are greater than zero.
19. A storage system, according to claim 15 , wherein a number of elements are discarded from the first queue and the second queue in response to the read request occurring when the queues are full.
20. A storage system, according to claim 15 , wherein there is a relationship between elements of the first queue and elements of the second queue that facilitates accessing corresponding elements of the second queue by initially accessing elements of the first queue.