We have mimix replication replicating same data to 2 servers , one is remote(570 mma power6)and other is local(It is a partition on 570 mma power 6). The replication to local server takes more time than remote server. Remote server has 2 processors, local server has 3 processors, disk space and usage is same, however remote server has 42 disk more than local server. Is there any way to reduce replication time on local server other than increasing of number of disks.
Since the local system is slower than the remote, network differences shouldn't be involved unless that local system has the previously mentioned line speed issue. Comparing the line description against the other two systems is the best first step. It's easy to get out of the way. That would be worth feeling lucky over.
If configuration looks right, then an actual test should be next. FTP a good sized savefile to both the remote and local systems to track transfer rates, and possibly repeat the test three or more times. Significant differences should start to indicate issues with cabling, switches or what have you between systems.
Show us some transfer rates to get us started.
If networking is out of the way, then you're in to normal "performance" analysis. Comparing basic CPYFs from one library to another on the same systems would give some very basic guidelines for how the two systems reacted.
Once you have a feeling for how the two major components, network and disk, contribute separately, you should have something to guide your next steps.
It's been a little while since "more disks" is assured to result in best performance results. It can still be a good solution, but "better disks" can give even better results. Disk performance can be greatly improved nowadays by removing lots of older drives and replacing with newer/bigger/faster yet many fewer drives.
And it's possible that none of the above will help much if the systems simply aren't properly tuned for the work they're doing. That local might just be running slower because it doesn't know how to do things better.
Tom
last time we too had a problem with mimix. after modifying the speed of the router, it really slowed down the replication speed.
after restarting the IPS on both sides the replication was really fast and the line was max out till 95-98%.
Free Guide: Managing storage for virtual environments
Complete a brief survey to get a complimentary 70-page whitepaper featuring the best methods and solutions for your virtual environment, as well as hypervisor-specific management advice from TechTarget experts. Don’t miss out on this exclusive content!
Discuss This Question: 2  Replies