This post is more than 5 years old
1 Rookie
•
35 Posts
0
2472
January 6th, 2016 14:00
Have Recoverpoint, want RP4VM's
Hello everyone,
I currently have recoverpoint protecting my environment (100% vmware). It is a VNX7500 to VNX7500. I like it, but what I don't like is recoverpoint has to do the entire Lun/datastore. Doing it this way I have to keep my VM's organized by lun, so my groups can come over correctly.
It would seem that with RP4VM's, it doesn't matter what lun/datastore the VM is on, it will just replicate the VM and it's datastores, so I could have VM's scattered across multiple datastores, and still create consistency groups with those VM's.
Am I understanding this correctly? I have highly considered moving from RP to RP4VM's because of this.
What are people's thoughts?
Thanks,
Brandon
forshr
2 Intern
•
1.1K Posts
0
January 6th, 2016 15:00
This is the primary difference between the two products in respect of VM protection so your summary is correct.
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
brettesinclair
2 Intern
•
715 Posts
1
January 6th, 2016 22:00
If you're 100% VMware I reckon it's an excellent move.
If you're currently using SRM you can also ditch it and use the built in orchestration of RP4VM's. If you're not using SRM, you'll enjoy the new ability to orchestrate dr operations.
BAronson1
1 Rookie
•
35 Posts
0
January 7th, 2016 07:00
Thank you both. Does anyone know if there is a performance impact with moving the splitter from the SAN to the esxi hosts?
forshr
2 Intern
•
1.1K Posts
1
January 7th, 2016 08:00
There are always mitigating factors when quoting performance numbers but due to the sequential nature of the ESXi splitting process the latency is slightly higher than array based splitter. Added response is around 0.6 - 0.8ms for 64kb writes during low to medium levels of IOPS activity. Please note that this is still negligible to host or app latency.
arosen_tca
6 Posts
0
January 8th, 2016 09:00
So I just happened to have done some (basic) testing with our new RP4VM install. I had one VM sitting on a VNXe3200. When running an IOMeter test with 4K blocks, 50% reads, 0% random, I was able to get 6.2MBps and 1500 IOPS. When the server was protected by RP4VM, my IOPS dropped to 1000 and MBps dropped to 4.3. That's a 30% drop in performance, which I'm not happy with. I understand that this test is with higher load, but once I load my host up I don't want my brand new VNXe to all of a sudden be the bottleneck due to RP. Unfortunately, since all my servers are currently physical I don't have a real-world workload I can easily simulate for better testing results.
Thoughts/experiences from the community?
Adam
forshr
2 Intern
•
1.1K Posts
0
January 8th, 2016 16:00
Can you check whether DelayACK is disabled for the iSCSI Storage Adapters. If it is enabled, can you disable it and retry the test. Refer to p25 of the following doc:
https://support.emc.com/docu55593_VNXe3200-Best-Practices-For-Performance---Applied-Best-Practices-Guide.pdf?language=en_US
Rich Forshaw
Consultant Corporate Systems Engineer - RecoverPoint & VPLEX (EMEA)
Data Protection and Availability Solutions
EMC Europe Limited
Mobile: +44 (0) 7730 781169
E-mail: richard.forshaw@emc.com
Twitter: @rw4shaw
EMC Europe Limited
Registered in England with Company No. 00990752 Registered office address: Level 1, Exchange House, Primrose Street, London C2A 2EG
The information contained in this e-mail message and any files transmitted with it are confidential. It is intended only for the addressee and others authorised to receive it. If you are not the intended recipient or the person responsible for delivering the message to the intended recipient, you are advised that you have received the e-mail in error; please delete it and notify the sender immediately. You should not retain the message or disclose its contents to anyone. Any disclosure, copying, distribution or action taken in reliance on the contents of the e-mail and its attachments is strictly prohibited.
BAronson1
1 Rookie
•
35 Posts
0
January 11th, 2016 09:00
Hi Arosen,
I am currently setting up a POC for RP4VM's.
I would like to do the same test you did, and I can post my results. Do you have the profile handy that you used in IOMeter, and the VM sepcs you used? I am 100% virtual, so my test might be able to offer some insight for others.