Memory device performance based on storage traffic pattern detection
A processing device of a memory sub-system can monitor a plurality of received commands to identify a forced unit access command. The processing device can identify a metadata area of the memory device based on the forced unit access command. The processing device can also perform an action responsive to identifying a subsequent forced unit access command to the metadata area.
1. A system comprising:
a memory device; and
a controller coupled to the memory device comprising memory cells operable as single level cells (SLCs) and multi-level cells (MLCs), wherein the memory device is configured to:
monitor incoming commands to determine a journal area of the memory device;
determine whether the incoming commands to a start logical block address (LBA) and an end LBA are write commands;
responsive to determining that the incoming commands are write commands, determine whether the incoming commands are forced unit access commands;
responsive to determining that the write commands are the forced unit access commands and responsive to the start LBA to the end LBA being accessed a threshold number of times, mark the start LBA to the end LBA as the journal area of the memory device; and
responsive to determining the journal area, direct subsequent writes to the start LBA to the end LBA of the journal area to SLC virtual blocks reserved for writes to the journal area.
2. The system of claim 1 , wherein the journal area is a fourth extended (EXT4) filesystem journal area.
3. The system of claim 1 , wherein the controller is further configured to configure the memory cells of the journal area to function as a virtual block including the start LBA and the end LBA of the journal area.
4. The system of claim 3 , wherein the controller is further configured to configure the memory cells of the virtual block wherein the virtual block comprises blocks from a plurality of memory devices including the memory device.
5. The system of claim 4 , wherein the controller is further configured to configure the virtual block from the plurality of memory devices as comprising a same block from each of the plurality of memory devices.
6. The system of claim 5 , wherein the plurality of memory devices are a managed NAND device and wherein each of the plurality of memory device are NAND devices.
7. A method, comprising:
monitoring incoming commands to determine a journal area of a memory device;
determining whether the incoming commands to a start logical block address (LBA) and an end LBA are write commands;
responsive to determining that the incoming commands are write commands, determining whether the incoming commands are forced unit access commands;
responsive to determining that the incoming commands are the forced unit access commands and responsive to the start LBA to the end LBA being accessed a threshold number of times, marking the start LBA to the end LBA as the journal area of the memory device; and
responsive to determining the journal area, direct subsequent writes to the start LBA to the end LBA of the journal area to single level cells (SLCs) virtual blocks reserved for writes to the journal area.
8. The method of claim 7 , further comprising:
responsive to determining the journal area, implementing a garbage collection policy specific to the start LBA to the end LBA of the journal area; and
wherein implementing the garbage collection policy specific to the journal area further comprises delaying a garbage collection of the journal area as compared to different garbage collection policies specific to non-journal areas.
9. The method of claim 8 , wherein the garbage collection policy of the journal area includes a delay time of the garbage collection based on a circular writing of the journal area.
10. The method of claim 8 , wherein delaying the garbage collection of the journal area further comprises indefinitely delaying the garbage collection of the journal area.
11. The method of claim 8 , further comprising separating journal data of the journal area from non-journal data in a separate block of the memory device.
12. The method of claim 11 , further comprising implementing the garbage collection policy specific to the start LBA and the end LBA of the journal area on the separate block of the memory device.
13. The method of claim 12 , further comprising implementing the garbage collection policy specific to the journal area on the separate block of the memory device, wherein the separate block is an SLC block.
14. The method of claim 13 , wherein the garbage collection policy specific to the journal area limits a write amplification factor of the journal area as compared to other blocks of the memory device.