Method and system for memory-based data caching
According to some embodiment, a backup storage system receives a request from a client for reading a data segment associated with a file object stored in a storage system. In response to the request, the system determines whether a cache hit counter associated with the data segment exceeds a cache hit threshold. The system further determines whether the data segment is associated with a file region of the file object that is frequently accessed. The system writes the data segment into a memory responsive to determining that the cache hit counter exceeds the cache hit threshold and the data segment is associated with the frequently accessed file region. Otherwise, the system writes the data segment into a solid state device (SSD) operating as a cache device.
1. A computer-implemented method for memory-based data caching, the method comprising:
receiving a request from a client for reading a data segment associated with a file object stored in a storage system;
in response to the request,
determining whether a cache hit counter associated with the data segment exceeds a cache hit threshold;
determining whether the data segment is associated with a file region of the file object that is frequently accessed;
writing the data segment into a memory responsive to determining that the cache hit counter exceeds the cache hit threshold and the data segment is associated with the frequently accessed file region, wherein the memory has stored therein a hash, table or map including one or more buckets, each of the one or more buckets including a respective bucket chain and a respective head entry, each bucket chain storing one or more data segments, and wherein a bucket hit counter and an access bit indicative if any data segment in the bucket is frequently accessed are encapsulated within the head entry; and
otherwise, writing the data segment into a solid state device (SSD) operating as a cache device.
2. The method of claim 1 , further comprising:
hashing the data segment to obtain one or more buckets;
for each bucket:
determining whether a bucket hit counter associated with a bucket chain of the bucket is low;
determining whether any data segment in the bucket chain is frequently accessed; and
evicting the bucket from the memory responsive to determining that the bucket hit counter is low or that no data segment in the bucket is frequently accessed.
3. The method of claim 2 , wherein a data segment in the bucket chain is frequently accessed if the data segment is accessed within a predetermined time period within the bucket chain.
4. The method of claim 1 , further comprising tracking the cache hit counter associated the data segment, wherein the cache hit counter is indicative of cache hits from the memory, the SSD operating as a cache device, and a disk storage.
5. The method of claim 4 , wherein the cache hit counter is stored in a cache index within the memory.
6. The method of claim 1 , wherein the cache hit threshold is 10 hits.
7. The method of claim 1 , wherein the frequently accessed file region is a region that is accessed within a previous hour.
8. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for memory-based data caching, the operations comprising:
receiving a request from a client for reading a data segment associated with a file object stored in a storage system;
in response to the request,
determining whether a cache hit counter associated with the data segment exceeds a cache hit threshold;
determining whether the data segment is associated with a file region of the file object that is frequently accessed;
writing the data segment into a memory responsive to determining that the cache hit counter exceeds the cache hit threshold and the data segment is associated with the frequently accessed file region, wherein the memory has stored therein a hash table or map including one or more buckets, each of the one or more buckets including a respective bucket chain and a respective head entry, each bucket chain storing one or more data segments, and wherein a bucket hit counter and an access bit indicative if any data segment in the bucket is frequently accessed are encapsulated within the head entry; and
otherwise, writing the data segment into a solid state device (SSD) operating as a cache device.
9. The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise:
hashing the data segment to obtain one or more buckets;
for each bucket:
determining whether a bucket hit counter associated with a bucket chain of the bucket is low;
determining whether any data segment in the bucket chain is frequently accessed; and
evicting the bucket from the memory responsive to determining that the bucket hit counter is low or that no data segment in the bucket is frequently accessed.
10. The non-transitory machine-readable medium of claim 9 , wherein a data segment in the bucket chain is frequently accessed if the data segment is accessed within a predetermined time period within the bucket chain.
11. The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise:
tracking the cache hit counter associated the data segment, wherein the cache hit counter is indicative of cache hits from the memory, the SSD operating as a cache device, and a disk storage.
12. The non-transitory machine-readable medium of claim 11 , wherein the cache hit counter is stored in a cache index within the memory.
13. The non-transitory machine-readable medium of claim 8 , wherein the cache hit threshold is 10 hits.
14. The non-transitory machine-readable medium of claim 8 , wherein the frequently accessed file region is a region that is accessed within a previous hour.
15. A data processing system, comprising:
a processor; and
a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations, the operations comprising:
receiving a request from a client for reading a data segment associated with a file object stored in a storage system;
in response to the request,
determining whether a cache hit counter associated with the data segment exceeds a cache hit threshold;
determining whether the data segment is associated with a file region of the file object that is frequently accessed;
writing the data segment into a memory responsive to determining that the cache hit counter exceeds the cache hit threshold and the data segment is associated with the frequently accessed file region, wherein the memory has stored therein a hash table or map including one or more buckets, each of the one or more buckets including a respective bucket chain and a respective head entry, each bucket chain storing one or more data segments, and wherein a bucket hit counter and an access bit indicative if any data segment in the bucket is frequently accessed are encapsulated within the head entry; and
otherwise, writing the data segment into a solid state device (SSD) operating as a cache device.
16. The data processing system of claim 15 , wherein the operations further comprise:
hashing the data segment to obtain one or more buckets;
for each bucket:
determining whether a bucket hit counter associated with a bucket chain of the bucket is low;
determining whether any data segment in the bucket chain is frequently accessed; and
evicting the bucket from the memory responsive to determining that the bucket hit counter is low or that no data segment in the bucket is frequently accessed.
17. The data processing system of claim 15 , wherein the operations further comprise:
tracking the cache hit counter associated the data segment, wherein the cache hit counter is indicative of cache hits from the memory, the SSD operating as a cache device, and a disk storage.