Unsolved
This post is more than 5 years old
5 Practitioner
•
274.2K Posts
0
723
How do I find Reliability of backup products?
What is specific method to be followed testing and tell that this reliability of the product at this point
We performed backup of 2000 Clients for a month performed restart of NetWorker Services only once. Can tell that reliability as 97% ?
How do I estimate reliability
What data needs to collected? (Throughtput,Backup time , no of errors in daemon.log)
What are factors needs to be consider to predict reliability?
ble1
14.3K Posts
1
October 23rd, 2014 01:00
Reliability of backup goes into two directions:
- platform stability and expected operational response
- data retrieval availability
Focussing on 1st group is easier, but it is 2nd parameter which is most important. In total, both will give you feeling of how it is going, but how you plan to quantify this is not easy task. Normally, you define test/parameters/observations which you count and you assign importance/marking to them - that way you may quantify it, but which parameters you will use and how will you rate it - depends on environment/customer.
ble1
14.3K Posts
1
October 23rd, 2014 01:00
You need to define matrix and rate it. For example, you said there was one restart. That alone means nothing unless you rate with given weight like
a) restart due to application configuration requirement (eg. renaming client)
b) restart due to application error
b1) has this restart been planned and executed where operator wanted without giving much headache
b2) has this restart been done due to application stand still
c) restart due to 3rd party influence (hosting, network)
d) restart due to planned change
Obviously, just a single restart can be differently viewed. If you assign different points to it, and you create model for your environment with other points, you can compare what you get at the end of the month compared to what would be ideal situation.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
October 23rd, 2014 01:00
Want to take platform stability and expected operational response
For that started measuring daemon.log errors, throughput of backup, memory/cpu, utilization I/O pattern
no defects obtained etc.
I have data which we have done testing for 30 days. Yet unable conclude Reliability of this product
Unable find any formula suitable for this . How do go about this ?
muser2
15 Posts
0
October 23rd, 2014 06:00
Reliability or success is measured by each customer...
Success could be measured by the various reports in Networker (good) or DPA (best)
Starting backup at designated time, might be important to certain clients. Successful backup is always a factor. But the most important measure of a successful backup, is the ability to restore the data backed up. If you back it up, but are unable to restore, you have done nothing.
Are you asking which variables need to be looked at measure backup success?