THIS IS A TEST INSTANCE. ALL YOUR CHANGES WILL BE LOST!!!!
...
3.) 1-hour test execution with high load (tokens were generated between 1-2 seconds on 15 threads)- without token eviction:
| Code Block |
|---|
perf.test.usecase.knoxtoken.numOfThreads=15 perf.test.usecase.knoxtoken.testDurationInSecs=3600 perf.test.usecase.knoxtoken.requestDelayLowerBoundInSecs=1 perf.test.usecase.knoxtoken.requestDelayUpperBoundInSecs=2 |
| Number of tokens created | Number of tokens renewed | Average response time (token creation)* | Average response time (token renewal) |
|---|---|---|---|
| ms | ms |
4.) 1-hour test execution with high load (tokens were generated between 1-2 seconds on 15 threads) - with token eviction:
In this test I kept the previous config and, additionallyAdditionally, I lowered the Token TTL to 10 minutes from 10 hours (by default, the token eviction grace period is set to 0). That triggered automated token eviction every 10 minutes.
| Number of tokens created | Number of tokens renewed | Average response time (token creation)* | Average response time (token renewal) |
|---|---|---|---|
| 49596 | 1430 | 85.5 ms | 19.09 ms |
Memory graph (using JConsole)
*: There is a short warmup period at the beginning of each test round (see test results where token creation time takes ~2 seconds a couple of times). We could even see better results if those values were not recorded at the end (the longer the test period the better the average result is).