SYSTEMS AND METHODS FOR SELF-TUNING MEMORY

• US 20070233989A1
• Filed: 03/30/2006
• Published: 10/04/2007
• Est. Priority Date: 03/30/2006
• Status: Active Grant

• Current Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Original Assignee: International Business Machines Corporation
• Inventors: Xue, Xun, Kuczynski, Wojciech, Huras, Matthew, Lightstone, Sam, Carroll, Matthew, Storm, Adam, Garcia-Arellano, Christian
• Granted Patent: US 7,694,102 B2
• Time in Patent Office: Days
• US Class Current: 711/170
• CPC Class Codes: G06F 12/0223 User address space allocation, e.g. contiguous or non contiguous base addressing
• CPC Class Codes: G06F 12/0646 Configuration or reconfiguration

Abstract

There are disclosed systems, computer program products, and methods for self-tuning memory. In an embodiment, a method for self-tuning memory comprises setting a tuning interval and a target range for free memory for a memory pool. If memory consumption falls outside of the set target range for free memory during a tuning interval, a new target size for the memory pool is set based on the target range for free memory. Memory allocation for the memory pool is increased or decreased for the next tuning interval, such that the new target size for the memory pool is reached. A decrement rate may be used to provide a controlled decrease of the memory pool over a plurality of tuning intervals if necessary, until the new target size for the memory pool is reached.

Claims

1. A data processing system implemented method of self-tuning memory, comprising:

setting a tuning interval;

for a memory pool, setting a target range for free memory;

if memory consumption increases above the set target range for free memory during a tuning interval, then setting a new target size for the memory pool based on the target range for free memory, and increasing memory allocation for the memory pool for the next tuning interval based on the new target size;

if memory consumption decreases below the set target range for free memory during a tuning interval, then setting a new target size for the memory pool based on the target range for free memory, and decreasing memory allocation for the memory pool for the next tuning interval using a decrement rate.

2. The data processing implemented method of claim 1, further comprising repeating over a plurality of tuning intervals setting a new target size for the memory pool based on the target range for free memory and decreasing memory allocation for the memory pool for the next tuning interval using a decrement rate if memory consumption decreases below the set target range for free memory during a tuning interval, until the new target size for the memory pool is reached.

3. The data processing implemented method of claim 2, further comprising selecting the decrement rate to be a percentage of the size of the memory pool.

4. The data processing implemented method of claim 3, further comprising rounding the decrement rate percentage into a specified minimum block of memory, and decreasing the memory allocation for the memory pool in units of the minimum block of memory.

5. The data processing implemented method of claim 2, further comprising setting the target range for free memory to be between about 50% and 60%.

6. The data processing implemented method of claim 1, further comprising setting a maximum size for the memory pool.

7. The data processing implemented method of claim 6, further comprising setting the maximum size of the memory pool to be a percentage of available database memory.

8. The data processing implemented method of claim 7, further comprising providing an overflow memory that may accommodate a temporary overflow beyond the set maximum size of the memory pool.

9. The data processing implemented method of claim 1, further comprising setting the maximum amount of memory consumable by a single transaction each time memory allocation is changed.

10. The data processing implemented method of claim 8, wherein memory blocks are immediately allocated from the overflow memory during a single interval.

11. A data processing system for self-tuning memory, comprising:

means for setting a tuning interval;

means for setting a target range for free memory for a memory pool;

means for determining if memory consumption increases above the set target range for free memory during a tuning interval, setting a new target size for the memory pool based on the target range for free memory, and increasing memory allocation for the memory pool for the next tuning interval based on the new target size;

means for determining if memory consumption decreases below the set target range for free memory during a tuning interval, setting a new target size for the memory pool based on the target range for free memory, and decreasing memory allocation for the memory pool for the next tuning interval using a decrement rate.

12. The data processing system of claim 11, further comprising means for repeating the decrease in memory allocation for the memory pool over a plurality of tuning intervals if necessary, until the new target size for the memory pool is reached.

13. The data processing system of claim 12, further comprising means for selecting the decrement rate to be a percentage of the size of the memory pool.

14. The data processing system of claim 13, further comprising means for rounding the decrement rate percentage into a specified minimum block of memory, and decreasing the memory allocation for the memory pool in units of the minimum block of memory.

15. The data processing system of claim 12, further comprising means for setting the target range for free memory to be between about 50% and 60%.

16. The data processing system of claim 11, further comprising means for setting a maximum size for the memory pool.

17. The data processing system of claim 16, further comprising means for setting the maximum size of the memory pool to be a percentage of available database memory.

18. The data processing system of claim 17, further comprising means for providing an overflow memory that may accommodate a temporary overflow beyond the set maximum size of the memory pool.

19. A computer program product comprising a data processor readable medium storing code that, when loaded into a data processing device, adapts the device to self tune memory, the data processor readable medium including:

code for setting a tuning interval;

code for setting a target range for free memory for a memory pool;

code for determining if memory consumption increases above the set target range for free memory during a tuning interval, setting a new target size for the memory pool based on the target range for free memory, and increasing memory allocation for the memory pool for the next tuning interval based on the new target size;

code for determining if memory consumption decreases below the set target range for free memory during a tuning interval, setting a new target size for the memory pool based on the target range for free memory, and decreasing memory allocation for the memory pool for the next tuning interval using a decrement rate.

20. The computer program product of claim 19, wherein the data processor readable medium further includes code for repeating the decrease in memory allocation for the memory pool using the decrement rate over a plurality of tuning intervals if necessary, until the new target size for the memory pool is reached.

21. The computer program product of claim 20, wherein the data processor readable medium further includes code for selecting the decrement rate to be a percentage of the size of the memory pool.

22. The computer program product of claim 21, wherein the data processor readable medium further includes code for rounding the decrement rate percentage into a specified minimum block of memory, and decreasing the memory allocation for the memory pool in units of the minimum block of memory.