Section Number and Name:
5.5. Tuning Virtual Memory
Describe the issue:
The current description of vm.swappiness have two shortcomings:
- it doesn't explain what swappiness really do: tune the trade-off between page cache and anonymous page.
- since rhel 6.4, swappiness==0 is more aggressive in avoiding swapping out. When the I/O activity is high, it increase risk of OOM killing.
Suggestions for improvement:
A value from 0 to 100 which controls the degree to which the system swaps. A high value prioritizes system performance, aggressively swapping processes out of physical memory when they are not active. A low value prioritizes interactivity and avoids swapping processes out of physical memory for as long as possible, which decreases response latency. The default value is 60.
A value from 0 to 100 which controls the degree to which the system favour anonymous memory or the page cache. A high value improves file-system performance, while aggressively swapping less active processes out of RAM.
A low value avoids swapping processes out of memory, which usually decrease latency, at the cost of I/O performance. The default value is 60.
Warning: Since RHEL 6.4, setting swappiness==0 will even more aggressively avoids swapping out, which increase the risk of OOM killing under strong memory and I/O pressure.
Adding docs-async to Devel Whiteboard, since this can be corrected asynchronous to the release schedule.
Any idea on what the turn around for this would be?
Latest version of the Red Hat Enterprise Linux 6 Performance Tuning Guide on the Red Hat Customer Portal:
Closing this BZ.