Troubleshooting

Improving Oracle OLTP database performance with Dell Fluid Cache for DAS

6.4 Re-setup Fluid Cache as noted in Appendix C.1.1 Enabling Fluid Cache.

6.5 Repeat step 5 to run the test again, if necessary.

7. Disabling Fluid Cache:

7.1 Fluid Cache was disabled. For details on the steps followed, refer to Appendix C.1.2

Removing Fluid Cache.

Performance results and analysis

This section presents the performance results and analysis of the two storage solutions described in the

Solution design and reference architecture section:

 Traditional HDD-based storage solution (Baseline Configuration)

 Fluid Cache for DAS-based storage solution (Fluid Cache)

For the remainder of this section each of the solutions will be referred to with the short description

mentioned in bold in the brackets above. These short descriptions will also be used as legends in the

graphs to represent each of the solution.

Note: The results presented in this section are intended only to showcase the relative performance of

the two solutions configured as described in this document. The results are not indicative of the

maximum capabilities of any system, database software, storage, or the solution.

For each of the solutions under test, multiple tests were conducted to check for consistency of the

results. Two seconds was used as the acceptable max average response time (ART) for each of the

tests. The tests were either manually terminated or the results were discarded past that point. Also, to

account for the standard deviation between each test run, the average of the results was used to

evaluate the performance of each of the storage solution.

Figure 4 shows the transactions per second (TPS) performance behavior spanning the entire duration of

the test for the two storage solutions evaluated in this paper. The graph plots the TPS behavior until

the average response time (ART) of two seconds was exceeded.