Troubleshooting

Improving Oracle OLTP database performance with Dell Fluid Cache for DAS
19
Figure 9 compares the average response time (ART) performance of both the storage solutions at 3100
user load. As can be seen from the graph, the Fluid Cache based solution delivers 95% lower ART at
3100 user load, which is near the peak performance of the baseline configuration.
Figure 10 compares the maximum user load that the two storage solutions were able to deliver while
keeping the response time at two seconds or less. As seen in Figure 10, the Fluid Cache based solution
delivered 34% more user load compared to the baseline configuration.
Performance graph: Max user load with 2secs ART SLA Figure 10.
The results in this section show that by adding Fluid Cache based solution to a saturated baseline
configuration can vastly improve the OLTP database performance without having to change any existing
hardware or software.
Conclusion
This paper demonstrates how a Dell PowerEdge server configured with Dell Fluid Cache for DAS can be
easily added to a traditional backend spinning HDD-based storage solution to significantly improve
OLTP performance in a DAS environment.
Fluid Cache can be a great solution to boost performance for the following DAS environments and use
cases
 Existing traditional spinning HDD-based storage environments looking to boost performance for
OLTP databases. This can be achieved by replacing their existing single node database server
with a Dell’s PowerEdge server that supports Dell PowerEdge Express Flash PCIe SSDs and
configured with Dell Fluid Cache for DAS.
3200
4300
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Baseline Fluid Cache
Userload
OLTP (R720) - Max Userload with 2secs ART SLA
34%