Bug 683513 - NUMA aware scheduling & memory allocation
Summary: NUMA aware scheduling & memory allocation
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: kernel
Version: 7.0
Hardware: All
OS: Linux
high
high
Target Milestone: alpha
: 7.0
Assignee: Rik van Riel
QA Contact: Virtualization Bugs
URL:
Whiteboard:
: 817919 909252 (view as bug list)
Depends On:
Blocks: 817917 824630 824634 824636 824980 872549 919431 985488 1032535 1032538 1071730
TreeView+ depends on / blocked
 
Reported: 2011-03-09 15:54 UTC by Rik van Riel
Modified: 2014-06-18 05:28 UTC (History)
18 users (show)

Fixed In Version: kernel-3.10.0-35.el7
Doc Type: Release Note
Doc Text:
NUMA-Aware Scheduling and Memory Allocation In Red Hat Enterprise Linux 7.0, the kernel automatically relocates processes and memory between NUMA nodes in the same system, in order to improve performance on systems with non-uniform memory access (NUMA).
Clone Of:
Environment:
Last Closed: 2014-06-13 11:27:55 UTC
Target Upstream Version:

Attachments (Terms of Use)
Add an attachment (proposed patch, testcase, etc.)

Description Rik van Riel 2011-03-09 15:54:22 UTC 
Description of problem:

Give processes a NUMA home node, try to schedule them near their memory and implement a low impact way of load balancing the processes between NUMA nodes.

Needs to be accepted upstream and solve the problem for many people and many workloads.
Comment 4 Jarod Wilson 2013-10-18 12:48:36 UTC 
Patch(es) available on kernel-3.10.0-35.el7
Comment 6 Linda Wang 2013-10-21 21:12:46 UTC 
*** Bug 909252 has been marked as a duplicate of this bug. ***
Comment 9 Hai Huang 2013-12-18 19:30:32 UTC 
*** Bug 817919 has been marked as a duplicate of this bug. ***
Comment 10 Ludek Smid 2014-06-13 11:27:55 UTC 
This request was resolved in Red Hat Enterprise Linux 7.0.

Contact your manager or support representative in case you have further questions about the request.
Note You need to log in before you can comment on or make changes to this bug.