[
https://issues.apache.org/jira/browse/SYSTEMML-843?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15403525#comment-15403525
]
Matthias Boehm commented on SYSTEMML-843:
-----------------------------------------
with a slight script modification I get the following
{code}
Total elapsed time: 5.795 sec.
Total compilation time: 1.476 sec.
Total execution time: 4.319 sec.
Number of compiled MR Jobs: 1.
Number of executed MR Jobs: 0.
Cache hits (Mem, WB, FS, HDFS): 136520/0/0/1.
Cache writes (WB, FS, HDFS): 49016/0/2.
Cache times (ACQr/m, RLS, EXP): 0.884/0.019/0.319/0.314 sec.
HOP DAGs recompiled (PRED, SB): 0/0.
HOP DAGs recompile time: 0.011 sec.
Functions recompiled: 1.
Functions recompile time: 0.049 sec.
Total JIT compile time: 9.145 sec.
Total JVM GC count: 11.
Total JVM GC time: 0.226 sec.
Heavy hitter instructions (name, time, count):
-- 1) tsne 3.089 sec 1
-- 2) x2p 3.032 sec 1
-- 3) MR-Job_CSV_REBLOCK 0.893 sec 1
-- 4) exp 0.544 sec 8034
-- 5) / 0.330 sec 24103
-- 6) write 0.315 sec 2
-- 7) tsmm 0.281 sec 1
-- 8) * 0.252 sec 16069
-- 9) leftIndex 0.252 sec 16468
-- 10) + 0.204 sec 22840
{code}
please do the following change for now, which allows to to update-in-place on P
{code}
Di = D[i,];
while (abs(Hdiff) > tol & itr < 50) {
Pi = exp(-Di * beta[i,1])
Pi[1,i] = 0.
sum_Pi = sum(Pi)
H = log(sum_Pi) + beta[i,1] * sum(Di * Pi) / sum_Pi
Pi = Pi/sum_Pi
Hdiff = as.scalar(H - logU)
#print("iter : " + itr)
if (Hdiff > 0.) {
#print ("in Hdiff > 0.0")
betamin = as.scalar(beta[i,1])
if (betamax == INF) {
beta[i,1] = beta[i,1] * 2.
} else {
beta[i,1] = (beta[i,1] + betamax) / 2.
}
} else {
#print("in Hdiff <= 0.0")
betamax = as.scalar(beta[i,1])
if (betamin == 0.) {
beta[i,1] = beta[i,1] / 2.
} else {
beta[i,1] = (beta[i,1] + betamin) / 2.
}
}
itr = itr + 1
}
P[i,] = Pi;
{code}
> leftIndex and cache release extremely slow
> ------------------------------------------
>
> Key: SYSTEMML-843
> URL: https://issues.apache.org/jira/browse/SYSTEMML-843
> Project: SystemML
> Issue Type: Bug
> Reporter: Imran Younus
> Attachments: tSNT.tar.gz
>
>
> I'm running the tSNE script in standalone mode with a subset of MNIST data
> (2500 points). I ran this with and without `-exec singlenode`. Here are the
> stats:
> (BTW, the same function implemented in python takes less than 10 sec!)
> -> with singlenode flag
> {code}
> ./bin/systemml scripts/staging/tSNE.dml -stats -nvargs
> X=/home/iyounus/workspace/tsne_python/mnist2500_X.txt Y=Y_out.txt C=C_out.txt
> 16/08/01 16:46:54 INFO api.DMLScript: SystemML Statistics:
> Total elapsed time: 109.667 sec.
> Total compilation time: 0.407 sec.
> Total execution time: 109.260 sec.
> Number of compiled MR Jobs: 0.
> Number of executed MR Jobs: 0.
> Cache hits (Mem, WB, FS, HDFS): 223692/0/0/1.
> Cache writes (WB, FS, HDFS): 80351/0/2.
> Cache times (ACQr/m, RLS, EXP): 0.289/0.015/85.192/0.043 sec.
> HOP DAGs recompiled (PRED, SB): 0/0.
> HOP DAGs recompile time: 0.007 sec.
> Functions recompiled: 1.
> Functions recompile time: 0.039 sec.
> Total JIT compile time: 4.924 sec.
> Total JVM GC count: 312.
> Total JVM GC time: 1.12 sec.
> Heavy hitter instructions (name, time, count):
> -- 1) tsne 109.202 sec 1
> -- 2) x2p 109.189 sec 1
> -- 3) leftIndex 106.728 sec 32136
> -- 4) tsmm 0.564 sec 1
> -- 5) exp 0.376 sec 8034
> -- 6) rangeReIndex 0.201 sec 40170
> -- 7) / 0.183 sec 24103
> -- 8) * 0.161 sec 16069
> -- 9) + 0.144 sec 22840
> -- 10) uak+ 0.106 sec 8036
> 16/08/01 16:46:54 INFO api.DMLScript: END DML run 08/01/2016 16:46:54
> {code}
> -> without singlenode flag
> {code}
> > ./bin/systemml scripts/staging/tSNE.dml -stats -nvargs
> > X=/home/iyounus/workspace/tsne_python/mnist2500_X.txt Y=Y_out.txt
> > C=C_out.txt
> 16/08/01 16:52:59 INFO api.DMLScript: SystemML Statistics:
> Total elapsed time: 127.290 sec.
> Total compilation time: 0.396 sec.
> Total execution time: 126.894 sec.
> Number of compiled MR Jobs: 1.
> Number of executed MR Jobs: 0.
> Cache hits (Mem, WB, FS, HDFS): 223693/0/0/1.
> Cache writes (WB, FS, HDFS): 80352/0/2.
> Cache times (ACQr/m, RLS, EXP): 0.421/0.016/100.974/0.041 sec.
> HOP DAGs recompiled (PRED, SB): 0/0.
> HOP DAGs recompile time: 0.009 sec.
> Functions recompiled: 1.
> Functions recompile time: 0.038 sec.
> Total JIT compile time: 4.835 sec.
> Total JVM GC count: 312.
> Total JVM GC time: 1.226 sec.
> Heavy hitter instructions (name, time, count):
> -- 1) tsne 126.426 sec 1
> -- 2) x2p 126.412 sec 1
> -- 3) leftIndex 123.982 sec 32136
> -- 4) exp 0.427 sec 8034
> -- 5) MR-Job_CSV_REBLOCK 0.412 sec 1
> -- 6) tsmm 0.308 sec 1
> -- 7) rangeReIndex 0.242 sec 40170
> -- 8) / 0.208 sec 24103
> -- 9) + 0.172 sec 22840
> -- 10) * 0.151 sec 16069
> 16/08/01 16:52:59 INFO api.DMLScript: END DML run 08/01/2016 16:52:59
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
