I think I understand the stats now. >From left to
right: L1 state, L2 state, Directory state, event, count, percentage. Is
this correct?
Could anyone answer the second question please? Is
there a way in Ruby to figure out where most of the latencies come from (like
gprof for c code)? I want to understand where the protocol spend the most time
stalling.
Thank you!
Lei
----- Original Message -----
Sent: Thursday, March 29, 2007 2:02
PM
Subject: [Gems-users] Understanding ruby
stats
Dear list,
How do I understand the following from Ruby
stats?
Request vs. System State
Profile --------------------------------
L1_I
L2_M
NA
GETS
13
0.050333 L1_I
L2_M
NA
GETX
12
0.0464612 L1_I
L2_M
NA GET_INSTR
32
0.123897 L1_I L2_MT
NA
PUTX 7547
29.2202 L1_I L2_NP
I
GETS
694
2.68701 L1_I L2_NP
I
GETX 5524
21.3876 L1_I L2_NP
I GET_INSTR
752
2.91157 L1_I L2_NP
M
GETS
95
0.367818 L1_I L2_NP
M
GETX
718
2.77993 L1_I L2_NP
M GET_INSTR
86
0.332972 L1_I L2_NP
NP
GETX
102
0.39492 L1_I L2_NP
O
GETS
6
0.0232306 L1_I L2_NP
O
GETX
57
0.220691 L1_I L2_NP
O GET_INSTR
4
0.0154871 L1_I L2_NP
S
GETS
94
0.363946 L1_I L2_NP
S
GETX 1553
6.01285 L1_I L2_NP
S GET_INSTR
96
0.37169 L1_I
L2_O
NA GET_INSTR
4
0.0154871 L1_I
L2_S
NA
GETS
4
0.0154871 L1_I
L2_S
NA GET_INSTR
2
0.00774353 L1_I L2_SO
NA
PUTS
47
0.181973 L1_I L2_SO
NA
PUTX
5
0.0193588 L1_I L2_SS
NA
PUTS 1181
4.57256 NA
L2_NP
M
PUTX 7033
27.2301 NA
L2_NP MM
PUTX
34
0.13164 NA
L2_NP
O
PUTX
124
0.480099 NA
L2_OIV M
PUTX
9
0.0348459
Also, is there a way in Ruby to figure out where
most of the latencies come from (like gprof for c code)? I want to understand
where the protocol spend the most time stalling.
Thanks in advance!
Lei
_______________________________________________ Gems-users mailing
list Gems-users@xxxxxxxxxxx https://lists.cs.wisc.edu/mailman/listinfo/gems-users Use
Google to search the GEMS Users mailing list by adding
"site:https://lists.cs.wisc.edu/archive/gems-users/" to your
search.
|