Overview

This playbook describes how to analyze reduction ratio on a DXi

Analysis

For this playbook we will use the Harris IT system that has been used in the other Dxi7500 Playbooks.  The purpose of looking at the reduction ratio is to determine if a customer is adding data that is not susceptible to our dedup engine which has an impact on both performance and capacity.  We are using the link for the following graphs on the Harris IT system:

http://dart.quantum.com/cgi-bin/stats?DB=CX0929BVA00446-20110223-123021&pr=0&start=1293080747&end=1298482221&base=1000

This timeline is from 12/23/2010 to 2/23/2011 and we are looking at the Reduction Ratio graph found under Overview or Reduction using the Main Menu.  From this graph we can see:

• Reduction ratio is over 10 for the past 9 weeks

Conclusion

A large decrease in reduction ratio would indicate that the customer started backing up a dataset that was not susceptible to deduplication or that something is wrong with our deduplication engine or backup application filter as was the case with Simpana 8.  A large increase would mean that the data ingested is highly susceptible to deduplication which, when excessive, can negatively impact performance simply because the system has to work harder.  A Total Reduction Factor of less than 5 should be investigated.  According to the graph for this system, we don’t not have a problem related to the Reduction Ratio.0