RIA: Efficient Monitoring of Operating Systems

9410116 Elnozahy本项目的目标是开发一个软件监测工具，以便能够在实际工作量和网络活动下收集性能数据。与其他硬件和软件监控工具不同，新工具可以收集长时间的执行跟踪，对实际系统几乎没有干扰。据估计，监测执行的持续时间比使用现有技术可能的时间长几个数量级。实现完全在软件中进行，并依赖于指令计数器，而指令计数器在现代处理器中正变得越来越常见。由于系统的执行不受监控过程的影响，因此该处理可以应用于在系统使用时收集非常长的痕迹。这将允许进行比今天更雄心勃勃的绩效研究。在程序的最后测试阶段，监控工具还可以用于调试目的，以发现间歇性错误。这些错误被称为Heizenbugs，是最难发现的错误，也是系统软件被认为不可靠的原因。***

9410116 Elnozahy The objective of this project is to develop a software monitoring tool that would allow the collection of performance data under realistic workloads and network activities. Unlike other hardware and software monitoring tools, the new tool can collect long execution traces with very little perturbation to the actual system. The duration of monitored execution is estimated to be orders of magnitude longer than has been possible using existing techniques. The implementation takes place entirely in software and relies on the instruction counters that are increasingly becoming common with modern processors. Because the execution of the system is not affected by the monitoring process, the processing can be applied for collecting very long traces while the system is in use. This will permit more ambitious performance studies than have been possible today. The monitoring tool can also be used for debugging purposes during the final testing phases of programs to uncover intermittent errors. These errors, known as Heizenbugs, are the most difficult to discover and are the reason for the perceived unreliability of systems software. ***