Varying Trade clients

To study the impact of the number of clients on performance, we ran several tests where we modified the number of clients for the Command cache and DistributedMap cache modes.

Figure 1 shows the impact the number clients had on performance. We varied the number of clients for the cache modes Command caching and DistributedMap caching.

Observations

Figure 1 shows the impact of varying clients when using Command caching. Between 15 and 40 clients, the throughput amount fluctuates around the average about plus or minus 1%. Of particular interest is the strong linear relationship of the response times.

Using DistributedMap caching shows the same behavior. Here, throughput amount fluctuates also around the average about plus or minus 1%. The response time shows the same strong linear relationship, but at lower values (80 - 90%) than for Command caching (see Figure 1).

Conclusions

If the number of clients exceeds a certain maximum, the throughput decreases very slightly, but the response time scales linearly with the number of clients. When the maximum is reached, throughput is no longer improved by adding clients. Therefore, when observing response time, it is important to look at values from runs with the same number of clients.