A5101719359- Software System Performance and Reliability
- Advanced Software Engineering Methodologies
- Software Testing and Debugging Techniques
- Parallel Computing and Optimization Techniques
- Scientific Computing and Data Management
- Software Engineering Research
- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Distributed systems and fault tolerance
- Machine Learning and Data Classification
- Radiomics and Machine Learning in Medical Imaging
- Anomaly Detection Techniques and Applications
- Service-Oriented Architecture and Web Services
- Software Reliability and Analysis Research
- Real-Time Systems Scheduling
- Advanced Data Storage Technologies
- Advanced Malware Detection Techniques
- Security and Verification in Computing
- Formal Methods in Verification
- Logic, programming, and type systems
- Education, Psychology, and Social Research
- Health Systems, Economic Evaluations, Quality of Life
- Algorithms and Data Compression
- Caching and Content Delivery
- Model-Driven Software Engineering Techniques
Charles University
2015-2024
Czech Academy of Sciences, Institute of Physics
2024
Technical University of Liberec
2024
Czech Academy of Sciences
2005-2024
Brno University of Technology
2024
Technical University of Munich
2016
University of West Bohemia
2012
Czech Academy of Sciences, Institute of Computer Science
2005
Institut national de recherche en informatique et en automatique
1999
Established benchmark suites for the Java Virtual Machine (JVM), such as DaCapo, ScalaBench, and SPECjvm2008, lack workloads that take advantage of parallel programming abstractions concurrency primitives offered by JVM Class Library. However, are fundamental understanding way in which modern applications data-processing frameworks use JVM's features, validating new just-in-time (JIT) compiler optimizations enable more efficient execution workloads. We present Renaissance, a suite composed...
The rapid adoption and the diversification of cloud computing technology exacerbate importance a sound experimental methodology for this domain. This work investigates how to measure report performance in cloud, well research community is already doing it. We propose set eight important methodological principles that combine best-practices from nearby fields with concepts applicable only clouds, new ideas about time-accuracy trade-off. show these are using practical use-case experiment. To...
Abstract The objective of the study is to evaluate evolution performance new ATLAS Inner-Tracker (ITk) strip sensors as a function radiation exposure, ensure proper operation upgraded detector during lifetime High-Luminosity Large Hadron Collider (HL-LHC). Full-size ITk Barrel Short-Strip (SS) with final layout design, ATLAS18SS, have been irradiated neutrons and gammas, confirm results obtained prototypes miniature development phase. irradiations cover wide range fluences doses that will...
While workload collocation is a necessity to increase energy efficiency of contemporary multi-core hardware, it also increases the risk performance anomalies due interference. Pinning certain workloads subset CPUs simple approach increasing isolation, but its effect depends on type and system architecture. Apart from common sense guidelines, pinning has not been extensively studied so far. In this paper we study impact CPU interference for pairs collocated workloads. Besides various...
Component ensembles provide expressive abstractions that address challenges in engineering software architectures for Internet of Things and cyber-physical systems, particularly when combined with mobile ad hoc networks. A case study collaborative parking demonstrates the open-endedness component their ability to self-adapt.
Although methods and tools for unit testing of performance exist over a decade, anecdotal evidence suggests is not nearly as common functionality. We examine this situation in study GitHub projects written Java, looking occurrences evaluation code frameworks. quantify the use such frameworks, identifying most relevant approaches, describe how we adjust design our SPL framework to follow these conclusions.
Compared to functional unit testing, automated performance testing is difficult, partially because correctness criteria are more difficult express for than functionality. Where existing approaches rely on absolute bounds the execution time, we aim assertions code in relative, hardware-independent terms. To this end, introduce Stochastic Performance Logic (SPL), which allows making statements about relative method performance. Since SPL interpretation based statistical tests applied...
Many decisions taken during software development impact the resulting application performance. The key whose potential is large are usually carefully weighed. In contrast, same care not used for many individual likely to be small -- simply because costs would outweigh benefits. Developer opinion common deciding factor these cases, and our goal provide developer with information that help form such opinion, thus preventing performance loss due accumulated effect of poor decisions.
The ATLAS experiment will replace its existing Inner Detector with the new all-silicon Tracker (ITk) to cope operating conditions of forthcoming high-luminosity phase LHC (HL-LHC). outer regions ITk be instrumented ∼18000 ATLAS18 strip sensors fabricated by Hamamatsu Photonics K.K. (HPK). With launch full-scale sensor production in 2021, community has undertaken quality control (QC) testing these ensure compliance mechanical and electrical specifications agreed HPK. is conducted at seven QC...
The tutorial aims at practitioners - researchers or developers who need to execute small scale performance experiments in Java. goal is provide the attendees with a compact overview of some issues that can hinder experiment mislead evaluation, and discuss methods tools help avoid such issues. will examine multiple elements software execution stack impact performance, including common virtual machine mechanisms (just-in-time compilation garbage collection together associated runtime...
We investigate the duet measurement procedure, which helps improve accuracy of performance comparison experiments conducted on shared machines by executing measured artifacts in parallel and evaluating their relative together, rather than individually. Specifically, we analyze behavior procedure multiple cloud environments use experimental evidence to answer research questions concerning assumption underlying procedure. demonstrate improvements ranging from 2.3x 12.5x (5.03x average) for...
Dynamic analysis tools are often implemented using instrumentation, particularly on managed runtimes including the Java Virtual Machine (JVM). Performing instrumentation robustly is especially complex such runtimes: existing frameworks offer limited coverage and poor isolation, while previous work has shown that apparently innocuous can cause deadlocks or crashes in observed application. This paper describes ShadowVM, a system for instrumentation-based dynamic analyses JVM which combines...
Numerous projects have evaluated the performance of CORBA middleware over past decade. Interestingly, many published results are either gathered or analyzed imprecisely. We point out common causes such imprecision related to gathering timing information and effects warm-up, randomization, cross talk delayed hidden functionality, demonstrate their impact on evaluation. also present suggestions reporting in a manner that is more relevant
Engineering a large software project involves tracking the impact of development and maintenance changes on performance. An approach for is regression benchmarking, which automated benchmarking evaluation performance at regular intervals. Regression must tackle nondeterminism inherent to contemporary computer systems execution environments results. On example fully environment mono open-source project, we show how problems associated with can be tackled using statistical methods.