Hardware support for performance measurements and energy estimation of OpenRISC processor

Bara, Lucian; Boncalo, Oana; Marcu, Marius

doi:10.1109/saci.2015.7208237

Cited by 1 publication

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bara et al [1] develop hardware support for performance monitoring in an processor core and show that performance monitoring also increases the energy consumption of a processor. In contrast to this work, they focus on hardware level performance monitors.…”

Section: Measurementmentioning

confidence: 99%

“…The examination of performance changes, performance anomalies or the general performance behavior of a software usually requires the measurement of the performance. 1 The measurement in production environments is called monitoring [21]. 2 Performance monitoring is possible using two options: (1) using instrumentation, i.e., by inserting measurement probes into the source code, or (2) using sampling, i.e., by frequently obtaining the current state of the application from the outside.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Solving the Challenge of Minimal Overhead Monitoring

Reichelt¹,

Kühne

Hasselbring

2023

Companion of the 2023 ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

The examination of performance changes or the performance behavior of a software requires the measurement of the performance. This is done via probes, i.e., pieces of code which obtain and process measurement data, and which are inserted into the examined application. The execution of those probes in a singular method creates overhead, which deteriorates performance measurements of calling methods and slows down the measurement process. Therefore, an important challenge for performance measurement is the reduction of the measurement overhead.To address this challenge, the overhead should be minimized. Based on an analysis of the sources of performance overhead, we derive the following four optimization options: (1) Source instrumentation instead of AspectJ instrumentation, (2) reduction of measurement data, (3) change of the queue and (4) aggregation of measurement data. We evaluate the effect of these optimization options using the MooBench benchmark. Thereby, we show that these optimizations options reduce the monitoring overhead of the monitoring framework Kieker. For MooBench, the execution duration could be reduced from 4.77 𝜇𝑠 to 0.39 𝜇𝑠 per method invocation on average. CCS CONCEPTS• General and reference → Performance; • Software and its engineering → Software performance; Software maintenance tools.

show abstract

Section: Measurementmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%