|
|
Acceleration method of measuring electrical machines
Ředina, Ondřej ; Mach, Martin (referee) ; Červinka, Dalibor (advisor)
This thesis deals with the acceleration method of measuring electrical machines. The beginning of the work is dedicated to the description of standard loading methods, the description of the acceleration method and the analysis of losses in the electrical machine. Two speed modulations are selected: linear and sinusoidal. Each modulation requires a different procedure for determining the speed profile parameters. To apply the method, a diagram of the connection of the workplace is drawn and its optimization is carried out. Subsequently, the usability of the oscilloscope card is verified, which ensures the determination of the instantaneous frequency. Before the test itself, the moment of inertia is measured and the parameters of the speed profile are calculated. The acceleration method is supplemented by a full load test for a more detailed comparison of the results. The measured data almost coincide with the predicted values and trends.
|
|
|
Reproducible Partial-Load Experiments in Workload Colocation Analysis
Podzimek, Andrej ; Bulej, Lubomír (advisor) ; Pena, Tomás Fernández (referee) ; van Hoorn, André (referee)
Hardware concurrency is common in all contemporary computer systems. Efficient use of hardware resources requires parallel processing and sharing of hardware by multiple workloads. Striking a balance between the conflicting goals of keeping servers highly utilized and maintaining a predictable performance level requires an informed choice of performance isolation techniques. Despite a broad choice of resource isolation mechanisms in operating systems, such as pinning of workloads to disjoint sets of processors, little is known about their effects on overall system performance and power consumption, especially under partial load conditions common in practice. Performance and performance interference under partial processor load is analyzed only after the fact, based on historical data, rather than proactively tested. This dissertation contributes a systematic approach to experimental analysis of application performance under partial processor load and in workload colocation scenarios. We first present a software tool set called Showstopper, capable of achieving and sustaining a variety of partial processor load conditions. Based on arbitrary pre-existing computationally intensive workloads, Showstopper replays processor load traces using feedback control mechanisms to maintain the desired load. As opposed to...
|
|
|
Reproducible Partial-Load Experiments in Workload Colocation Analysis
Podzimek, Andrej ; Bulej, Lubomír (advisor) ; Pena, Tomás Fernández (referee) ; van Hoorn, André (referee)
Hardware concurrency is common in all contemporary computer systems. Efficient use of hardware resources requires parallel processing and sharing of hardware by multiple workloads. Striking a balance between the conflicting goals of keeping servers highly utilized and maintaining a predictable performance level requires an informed choice of performance isolation techniques. Despite a broad choice of resource isolation mechanisms in operating systems, such as pinning of workloads to disjoint sets of processors, little is known about their effects on overall system performance and power consumption, especially under partial load conditions common in practice. Performance and performance interference under partial processor load is analyzed only after the fact, based on historical data, rather than proactively tested. This dissertation contributes a systematic approach to experimental analysis of application performance under partial processor load and in workload colocation scenarios. We first present a software tool set called Showstopper, capable of achieving and sustaining a variety of partial processor load conditions. Based on arbitrary pre-existing computationally intensive workloads, Showstopper replays processor load traces using feedback control mechanisms to maintain the desired load. As opposed to...
|
guest ::
