This post is more than 5 years old
32 Posts
0
5471
February 17th, 2016 03:00
How to get the RPO and RTO values from current RPA setup
Hello All,
I would need your expert advice and bit of understanding on RPO and RTO part of our Recoverpoint Environment. We are using RPA EX 4.1 CRR for replicating between US and Germany. I wanted to know my RPO values which we are achieving with the current bandwidth. I was told that Journal Capacity also plays an important role in achieving better RPO.
Also what are other ways besides increasing our bandwidth which can be implemented in our setup to get better results.
Sincerely,
Ivan
No Events found!
echolaughmk
2 Intern
•
522 Posts
0
February 17th, 2016 14:00
You current RPO values is what you would set in the CG section for the RPO you want to maintain:
In terms of seeing what it is, I think there is a CLI command that will spit it out for you (or used to be), but you can look at your journal volume and the snapshots it has listed and the time difference between them is your relative RPO. It won't be an exact period of time between each because RP operates dynamically in async mode to basically be optimized as fast as it can be unless otherwise manually regulated or set to sync mode. The values above look like you have an RPO of seconds and under 10 seconds for the CG's listed. These will fluctuate to be under the RPO you designate with the copy but won't be the same periods of time in between each as designed. I don't know of any actual report on the average RPO, but I think the tool you are referring to is the long_term_stats tool which I believe can be utilized to see some of the RPO averages from the consolidated statistics output.
-Keith
echolaughmk
2 Intern
•
522 Posts
0
February 17th, 2016 04:00
Are you looking for the RPO values in terms of looking at your journal and seeing the relative time difference between the snapshots? That will give you the first indication of the RPO you can achieve. Or are you looking about trying to figure out the sizing of what your RPO values will become if you change the bandwidth, etc?
ivan_skinner
32 Posts
0
February 17th, 2016 06:00
I am trying to find out my current RPO values, below is the output I captured from two different CG. These are just Journal usage that I have shared. I was once told that the current Image is the RPO value you are getting. This may be correct but these values always change, and we never get the similar values some time the current image goes back to many hours or even days in some worst case.
When my customer seeks for the report on RPO we dont have any assured values on data protection as it tends to fluctuate.
CG1
Journal:
Usage: 66.93 GB
Total: 68.95 GB
Latest image: Wed Feb 17 2016 08:55:57
Current image: Wed Feb 17 2016 08:48:30
Journal lag: 30.05 MB
Protection window:
Current:
Value: 5 days 6 hr
Status:N/A
Predicted:
Value: 4 days 19 hr
CG2
Journal:
Usage: 454.83 GB
Total: 488.92 GB
Latest image: Wed Feb 17 2016 08:58:05
Current image: Wed Feb 17 2016 08:55:27
Journal lag: 28.01 MB
Protection window:
Current:
Value: 1 weeks 2 days
Status:N/A
Predicted:
Value: 1 weeks 1 days
I do understand that EMC has there own sizing tool which can be used to analyze the data and can share the predicted bandwidth usage as per the current IO trend.
ivan_skinner
32 Posts
0
February 18th, 2016 01:00
Thank you so much for clarifying my doubt. So RPO values will be decided dynmically but as per the default setting it will be 25 second. I guess that 25 second interval can further be customized?
You have any recommendation which can be implemented for better results, most of our CG's are VM SRM protected.
echolaughmk
2 Intern
•
522 Posts
1
February 18th, 2016 05:00
correct - you can change that RPO value and RP will try to meet it as fast or faster than that you set out of the box. If it can't meet it, you will see an RPO elongation and then have to take any of the necessary actions to correct it - WAN issues, jvol issues, RPA utilization, etc.
The default of 25 seconds works well for most deployments so if you don't have a need to tweak them, the PiT images it will create for you should be more than enough and you probably don't need to worry about them unless you start having issues.
With SRM, by default it takes the last snapshot upon failover/test so you will likely want that to be seconds behind and the default RPO of 25 seconds will capture that. Since you are using SRM, you can also check out the VSI plugin out there to chose the PiT image on failover if that is of any use to you or your customer.