Attributes | Values |
---|
rdf:type
| |
Description
| - Computers are often used as a tool to measure time in various experiments. Benchmarking and hardware performance evaluation are very common examples. There one typically needs to measure the duration of certain computer operation. The precision of hardware clocks is typically more than satisfying for this purpose but the measured time is also influenced by other factors such as the used operating system. In this paper, we evaluate how does the variance of benchmark results depend on the operating system. We developed a set of benchmarks to measure the characteristics of memory subsystem, the cost of task preemption and the cost of task migration between CPU cores/sockets. We run the benchmarks on NOVA microkernel as well as on two different configurations of Linux, in all cases on x86 architecture. We captured the data from several tenths of hours of running experiments and compare the obtained results. The mean measured values are in all cases similar. Besides the mean values, we compare the variance of subsequent runs of the benchmark. The results vary depending on the experiment, but there is a clearly visible trend in all experiments - the noise generated by the microkernel is in most cases the lowest, Debian/Linux sits in the middle and the highest noise is produced by a minimal configuration of the Linux kernel.
- Computers are often used as a tool to measure time in various experiments. Benchmarking and hardware performance evaluation are very common examples. There one typically needs to measure the duration of certain computer operation. The precision of hardware clocks is typically more than satisfying for this purpose but the measured time is also influenced by other factors such as the used operating system. In this paper, we evaluate how does the variance of benchmark results depend on the operating system. We developed a set of benchmarks to measure the characteristics of memory subsystem, the cost of task preemption and the cost of task migration between CPU cores/sockets. We run the benchmarks on NOVA microkernel as well as on two different configurations of Linux, in all cases on x86 architecture. We captured the data from several tenths of hours of running experiments and compare the obtained results. The mean measured values are in all cases similar. Besides the mean values, we compare the variance of subsequent runs of the benchmark. The results vary depending on the experiment, but there is a clearly visible trend in all experiments - the noise generated by the microkernel is in most cases the lowest, Debian/Linux sits in the middle and the highest noise is produced by a minimal configuration of the Linux kernel. (en)
|
Title
| - Operating System Noise: Linux vs. Microkernel
- Operating System Noise: Linux vs. Microkernel (en)
|
skos:prefLabel
| - Operating System Noise: Linux vs. Microkernel
- Operating System Noise: Linux vs. Microkernel (en)
|
skos:notation
| - RIV/68407700:21230/12:00198380!RIV13-GA0-21230___
|
http://linked.open...avai/predkladatel
| |
http://linked.open...avai/riv/aktivita
| |
http://linked.open...avai/riv/aktivity
| |
http://linked.open...vai/riv/dodaniDat
| |
http://linked.open...aciTvurceVysledku
| |
http://linked.open.../riv/druhVysledku
| |
http://linked.open...iv/duvernostUdaju
| |
http://linked.open...titaPredkladatele
| |
http://linked.open...dnocenehoVysledku
| |
http://linked.open...ai/riv/idVysledku
| - RIV/68407700:21230/12:00198380
|
http://linked.open...riv/jazykVysledku
| |
http://linked.open.../riv/klicovaSlova
| - Linux; Microkernel; Noise (en)
|
http://linked.open.../riv/klicoveSlovo
| |
http://linked.open...ontrolniKodProRIV
| |
http://linked.open...in/vavai/riv/obor
| |
http://linked.open...ichTvurcuVysledku
| |
http://linked.open...cetTvurcuVysledku
| |
http://linked.open...vavai/riv/projekt
| |
http://linked.open...UplatneniVysledku
| |
http://linked.open...iv/tvurceVysledku
| - Sojka, Michal
- Waechtler, S.
|
http://localhost/t...ganizacniJednotka
| |