Changes between Version 58 and Version 59 of Check
- Timestamp:
- Oct 29, 2010, 1:30:10 PM (14 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Check
v58 v59 413 413 === Performance === 414 414 415 The lag between the detection of a node failing and restoring the service, by turning it on for another node, is some tens of miliseconds. We have tried a few measurements, for both local islands and remote islands. In these examples, the Self-healing service is running in the same network as the Bucharest island and we introduce node failures in both Bucharest and Dresden islands.415 The lag between the detection of a node failing and restoring the service, by turning it on for another node, is some tens of miliseconds. We have tried a few measurements, for both local islands and remote islands. In these examples, the Self-healing service is running in the same network as the Bucharest island and we introduce node failures causing node starts in both Bucharest and Dresden islands. 416 416 417 417 [[Image(sh-performance.png, center, 600px)]] … … 419 419 The data from the above screenshot is better shown in this table: 420 420 421 ||= Test no. =||= Node failure=||= Detection time [s] =||= Service restored time [s] =||= Lag [ms] =||421 ||= Test no. =||= Node started =||= Detection time [s] =||= Service restored time [s] =||= Lag [ms] =|| 422 422 || 1 || remote || 24.182 || 24.297 || 115 || 423 423 || 2 || local || 30.741 || 30.775 || 34 || … … 425 425 || 4 || local || 37.247 || 37.279 || 32 || 426 426 427 You may notice that the mean restore time for local failures is ~30-40 ms and the mean restore time for remote failures is ~100-150 ms. The difference between remote and local times is most probably caused by network lag times and remote times can vary even more depending on the connection between the computer running the Self-healing service and the node island.427 You may notice that the mean restore time for local node starts is ~30-40 ms and the mean restore time for remote node starts is ~100-150 ms. The difference between remote and local times is most probably caused by network lag times. Remote times can vary even more depending on the connection between the computer running the Self-healing service and the node island. 428 428 429 429 = Proposed Check Demonstration Sketch =