[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2012-02-13 Thread Peter Schuller (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13207466#comment-13207466
 ] 

Peter Schuller commented on CASSANDRA-3589:
---

I just realized something. I wasn't looking into this but had a separate 
realization investigating streaming performance - ever since Cassandra moved to 
single-pass streaming (CASSANDRA-2677) streaming easily becomes CPU bound. If 
the mpbs numbers earlier in this ticket (35, 19) are mega*bytes*, the numbers 
are well within what you might reasonably expect from being CPU bound. I have 
yet to look into how the bulk loader stuff works, but is it possible it's going 
through the same path, such that it's spending CPU time re-creating the sstable 
on reception?

(This may be obvious to folks, I have actually never look at the bulk loader 
support before.)

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-14 Thread Samarth Gahire (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13169343#comment-13169343
 ] 

Samarth Gahire commented on CASSANDRA-3589:
---

I have used apache-cassandra-2011-12-14_03-12-43-bin.tar.gz binary downloaded 
from https://builds.apache.org/job/Cassandra/1255/artifact/cassandra/build/; 
for testing sstable generation.
I hope ,I have used a proper binary for testing .Please correct me if I am 
wrong.

There is a massive improvement.Thank you so much for the fix.
Eagerly waiting for sstable-loader fix.


 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Assignee: Sylvain Lebresne
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-13 Thread Samarth Gahire (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13168321#comment-13168321
 ] 

Samarth Gahire commented on CASSANDRA-3589:
---

No, I do not have any secondary indexes on any of the column family and I have 
done the fair comparison and seen some performance hit in sstable-loader 
utility. 

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Assignee: Sylvain Lebresne
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-13 Thread Jonathan Ellis (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13168467#comment-13168467
 ] 

Jonathan Ellis commented on CASSANDRA-3589:
---

Have you been able to benchmark Sylvain's patch?

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Assignee: Sylvain Lebresne
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-12 Thread Sylvain Lebresne (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13168010#comment-13168010
 ] 

Sylvain Lebresne commented on CASSANDRA-3589:
-

For the generation first, the problem is CASSANDRA-3618, but I've opened a 
separate ticket with a more accurate description and because this doesn't only 
affect the sstable generator. Note that in simple tests, I saw up to a ~2x 
difference between 0.8 and 1.0, but never a 3-4x. That could be because what 
you do for the sstable generation exacerbate the problem more than my tests, 
but it would still be great if you could try the patch on CASSANDRA-3618 and if 
it fully solve the sstable generation problem for you.

I have yet to check for the sstable loading performance problem, but a quick 
question first: do you have secondary indexes on the column family you're 
loading data in?

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Assignee: Sylvain Lebresne
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-09 Thread Samarth Gahire (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13166011#comment-13166011
 ] 

Samarth Gahire commented on CASSANDRA-3589:
---

I have checked the CPU and IO utilization

1) While generating sstables CPU utilization with cassandra-0.8.7 is around 80% 
while it is around 90-95% in cassandra-1.0.5 
2) While generating sstables We can see I/O that is disk write after each 20 
-25 seconds and cassandra0.8.7 write to disk with around 35mbps while 
cassandra-1.0.5 write to disk with 19mbps 

Apart from this i cant see any deference let me know for additional information.

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Assignee: Sylvain Lebresne
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira




[jira] [Commented] (CASSANDRA-3589) Degraded performance of sstable-generator api and sstable-loader utility in cassandra 1.0.x

2011-12-08 Thread Samarth Gahire (Commented) (JIRA)

[ 
https://issues.apache.org/jira/browse/CASSANDRA-3589?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13165139#comment-13165139
 ] 

Samarth Gahire commented on CASSANDRA-3589:
---

Actually i am just calculating the total time taken by the our program to 
generate the sstables.I can see this deference when i change the cassandra jar 
included in classpath of program.
About CPU and IO utilisation i will check and let you know as soon as possible.

 Degraded performance of sstable-generator api and sstable-loader utility in 
 cassandra 1.0.x
 ---

 Key: CASSANDRA-3589
 URL: https://issues.apache.org/jira/browse/CASSANDRA-3589
 Project: Cassandra
  Issue Type: Bug
  Components: Tools
Affects Versions: 1.0.0
Reporter: Samarth Gahire
Priority: Minor

 we are using Sstable-Generation API and Sstable-Loader utility.As soon as 
 newer version of cassandra releases I test them for sstable generation and 
 loading for time taken by both the processes.Till cassandra 0.8.7 there is no 
 significant change in time taken.But in all cassandra-1.0.x i have seen 3-4 
 times degraded performance in generation and 2 times degraded performance in 
 loading.Because of this we are not upgrading the cassandra to latest version 
 as we are processing some TeraBytes of data everyday time taken is very 
 important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira