Whose Cache Line Is It Anyway? Mihir Nanavati, Mark Spear, Nathan Taylor, Shriram Rajagopalan, Dutch...
-
Upload
edwin-sanders -
Category
Documents
-
view
217 -
download
0
Transcript of Whose Cache Line Is It Anyway? Mihir Nanavati, Mark Spear, Nathan Taylor, Shriram Rajagopalan, Dutch...
Whose Cache Line Is It Anyway?
Mihir Nanavati, Mark Spear, Nathan Taylor, Shriram Rajagopalan, Dutch T. Meyer, William Aiello, and Andrew
WarfieldUniversity of British Columbia
Operating System Support for LiveDetection and Repair of False Sharing
13
1 2 3 4 5 6 7 80
5
10
15
20
25
30
35
40
Serial ParallelRegular (FS) Source Fixed
No. of Cores
Tim
e (
s)
14
1 2 3 4 5 6 7 80
5
10
15
20
25
30
35
40
Serial ParallelRegular (FS) Source Fixed
No. of Cores
Tim
e (
s)
15
1 2 3 4 5 6 7 80
5
10
15
20
25
30
35
40
Serial ParallelRegular (FS) Source Fixed
No. of Cores
Tim
e (
s)
16
1 2 3 4 5 6 7 80
5
10
15
20
25
30
35
40
Serial ParallelRegular (FS) Source Fixed
No. of Cores
Tim
e (
s)
17
1 2 3 4 5 6 7 80
5
10
15
20
25
30
35
40
Serial ParallelRegular (FS) Source Fixed
No. of Cores
Tim
e (
s)
7.5x
Linux Kernel [OSDI ’10], JVM [Dice, 2012],
Software Transactional Memory [HPCA ’06]
25
Does this signifyfalse sharing?
Performance Counters
Log Page Reads
Instruction Emulation
Log-Analysis
Does contention exist?What pages are involvedin the contention?
What are the byte rangesbeing accessed?
Rules for remapper
36
0 1000 2000 3000 4000 5000 60000
100
200
300
400
500
600
Time (ms)
Pro
gre
ss (
mill
ion
reco
rds)
Remappings Established
Version with false sharing under Plastic
Coherence Invalidations
Source-fixed Version
110 M/sec
160 M/sec
37
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1 Regular w/PlasticCCBenc
hPhoenix Parsec
1.4x
3.6x
5.4x
Norm
aliz
ed
Perf
orm
an
ce