Unsolved
This post is more than 5 years old
5 Posts
0
445
June 13th, 2016 02:00
RecoverPoint doubles service time
In a production environment when RecoverPoint local async replication is activated the service time figures double.
The number of RPAs at each side of the RecoverPoint cluster was recently increased from 2 to 5. The corresponding VNX5300s however are close to the max supported drive number. In our test environment RecoverPoint behaved the same way ie latency doubles.
Since RecoverPoint operates out of band, shouldn’t service time remain unaffected? Is this due to the impact of write splitting and the entries on the journal?
No Events found!
forshr
2 Intern
•
1.1K Posts
0
June 13th, 2016 12:00
RP is out of band in that the splitter will only wait a short time for the RPA to ACK the write but ACKing the wrote carries a small latency overhead. Any performance related issues in the system should be identified by a highload on the basis that this this async replication.
Regards,
Rich Forshaw
Consultant Corporate Systems Engineer - RecoverPoint & VPLEX (EMEA)
Data Protection and Availability Solutions
EMC Europe Limited
Mobile: 44 (0) 7730 781169 44%20(0)%207730%20781169>
E-mail: richard.forshaw@emc.com
Twitter: @rw4shaw