Minimizing effects of cache thrashing by altering a persistence policy for non-temporal workloads

Patent metadata

Field	Value
Publication number	US-2026072836-A1
Application number	US-202418883172-A
Country	US
Kind code	A1
Filing date	Sep 12, 2024
Priority date	Sep 12, 2024
Publication date	Mar 12, 2026
Grant date	—

Abstract

implemented method, system, and computer program product for minimizing the effects of cache thrashing involving non-temporal workloads. The cache activities of a workload, including the cache activities (e.g., number of cache hits) involving local and peer caches, are monitored. Based on analyzing the metrics of such monitored cache activities, a determination is made as to whether a non-temporal workload is identified. For example, such a determination may be based on comparing the metrics of the monitored cache activities of the workload to a threshold value. Upon identifying a non-temporal workload, the cache line(s) associated with the non-temporal workload are identified. The persistence policy for the identified cache line(s) is then altered. For example, the persistence policy for the identified cache line(s) may be altered by reducing the tenure of such a cache line(s) thereby reducing the number of cache misses or evictions and minimizing the effects of cache thrashing.

First claim

1 . A computer-implemented method for minimizing effects of cache thrashing, the method comprising:
monitoring cache activity of a workload; identifying a non-temporal workload based on said monitored cache activity of said workload; identifying one or more cache lines associated with said non-temporal workload; and altering a persistence policy for said one or more cache lines associated with said non-temporal workload.

2 . The method as recited in claim 1 , wherein said persistence policy is altered to reduce a tenure of said one or more cache lines.

3 . The method as recited in claim 2 , wherein said persistence policy is altered to reduce said tenure of said one or more cache lines via setting a particular value in a hint bit.

4 . The method as recited in claim 1 , wherein said persistence policy is altered to cast-out said one or more cache lines directly to memory.

5 . The method as recited in claim 1 further comprising:
analyzing metrics of said monitored cache activity of said workload.

6 . The method as recited in claim 5 , wherein said metrics of said monitored cache activity of said workload comprise one or more of the following selected from the group consisting of: cache hit/miss ratio, cache miss data sources, fetch/cast-out ratio, and percentage of cast-outs with changed data.

7 . The method as recited in claim 5 , wherein said cache activity of said workload is monitored in connection with a peer cache, wherein said metrics of said monitored cache activity of said workload comprise a cast-in/fetch-hit ratio provided by said peer cache.

8 . The method as recited in claim 1 , wherein said cache activity of said workload is monitored in connection with local caches and peer caches.

9 . The method as recited in claim 1 , wherein said workload streams data from a memory.

10 . A computer program product for minimizing effects of cache thrashing, the computer program product comprising one or more computer readable storage mediums having program code embodied therewith, the program code comprising programming instructions for:
monitoring cache activity of a workload; identifying a non-temporal workload based on said monitored cache activity of said workload; identifying one or more cache lines associated with said non-temporal workload; and altering a persistence policy for said one or more cache lines associated with said non-temporal workload.

11 . The computer program product as recited in claim 10 , wherein said persistence policy is altered to reduce a tenure of said one or more cache lines.

12 . The computer program product as recited in claim 11 , wherein said persistence policy is altered to reduce said tenure of said one or more cache lines via setting a particular value in a hint bit.

13 . The computer program product as recited in claim 10 , wherein said persistence policy is altered to cast-out said one or more cache lines directly to memory.

14 . The computer program product as recited in claim 10 , wherein the program code further comprises the programming instructions for:
analyzing metrics of said monitored cache activity of said workload.

15 . The computer program product as recited in claim 14 , wherein said metrics of said monitored cache activity of said workload comprise one or more of the following selected from the group consisting of: cache hit/miss ratio, cache miss data sources, fetch/cast-out ratio, and percentage of cast-outs with changed data.

16 . The computer program product as recited in claim 14 , wherein said cache activity of said workload is monitored in connection with a peer cache, wherein said metrics of said monitored cache activity of said workload comprise a cast-in/fetch-hit ratio provided by said peer cache.

17 . The computer program product as recited in claim 10 , wherein said cache activity of said workload is monitored in connection with local caches and peer caches.

18 . A system, comprising:
a memory for storing a computer program for minimizing effects of cache thrashing; and a processor connected to said memory, wherein said processor is configured to execute program instructions of the computer program comprising:
monitoring cache activity of a workload;
identifying a non-temporal workload based on said monitored cache activity of said workload;
identifying one or more cache lines associated with said non-temporal workload; and
altering a persistence policy for said one or more cache lines associated with said non-temporal workload.

19 . The system as recited in claim 18 , wherein said persistence policy is altered to reduce a tenure of said one or more cache lines.

20 . The system as recited in claim 19 , wherein said persistence policy is altered to reduce said tenure of said one or more cache lines via setting a particular value in a hint bit.

21 . The system as recited in claim 18 , wherein said persistence policy is altered to cast-out said one or more cache lines directly to memory.

22 . The system as recited in claim 18 , wherein the program instructions of the computer program further comprise:
analyzing metrics of said monitored cache activity of said workload.

23 . The system as recited in claim 22 , wherein said metrics of said monitored cache activity of said workload comprise one or more of the following selected from the group consisting of: cache hit/miss ratio, cache miss data sources, fetch/cast-out ratio, and percentage of cast-outs with changed data.

24 . The system as recited in claim 22 , wherein said cache activity of said workload is monitored in connection with a peer cache, wherein said metrics of said monitored cache activity of said workload comprise a cast-in/fetch-hit ratio provided by said peer cache.

25 . The system as recited in claim 18 , wherein said cache activity of said workload is monitored in connection with local caches and peer caches.

Assignees

• IBM

Inventors

• Sonnelitter Iii Robert J
• Walters Craig R
• Francois Avery

Classifications

• G06F2212/502

  using adaptive policy · CPC title

• G06F12/0888

  using selective caching, e.g. bypass · CPC title

• G06F2212/60

  Details of cache memory · CPC title

• G06F12/0802 Primary

  Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches · CPC title