Performance Modeling and Benchmarking of Event-Based ... - DVS

More documents

Recommendations

Info

24 CHAPTER 2. BACKGROUND R W R 0HPRU\ 3DUWLWLRQV P W 'LVN W R R R S W P &38 W R R R R R S 7HUPLQDOV S S S W R 'LVN W R R R S Figure 2.8: A QPN Model of a Central Server with Memory Constraints (reprinted from [26]). Solving of QPNs & Tools for QPNs For QPNs, the analytic solving approach is well-defined [26] and implemented by several tools, e.g. [25, 23]. However, the analytic approach has limitations regarding the number of possible tokens and places which lead to a state explosion for models of real world applications [123]. Therefore, we decided to use a simulation-based QPN solver for our models. Such a simulationbased approach was presented in [123] which is implemented by the QPME tool (Queueing Petri net Modeling Environment) [127, 129, 128]. We employed this tool to build and analyze our QPN models. QPME provides a QPN editor including a graphical user interface, which helps to construct QPN models and the optimized simulation engine SimQPN [127, 123] for model analysis. As a result of our work, several new features were added to QPME and to the SimQPN engine. Further, the performance of the solver was increased significantly. 2.4 Concluding Remarks In this chapter we provided the background needed for the understanding of this thesis. We started with the concept of events and event-based systems in general. As part of this discussion, we reviewed related concepts such as publish/subscribe communication in detail and described underlying technologies such as Active Databases and their properties. Focusing on MOMs and DEBS, established standards were specified in detail. Additionally, we composed a comprehensive list of existing middleware products and supported standards. At the end of this chapter, we introduced the queueing Petri net paradigm and gave an overview of existing tools and solvers. QPNs provide several benefits over conventional modeling paradigmas by combining the ideas of QNs and SPNs. As a result, system aspects such as blocking, software and hardware contention and synchronization can be easily reflected in a QPN model.
Chapter 3 Related Work In this chapter, we provide an overview of the current state in performance modeling and benchmarking of event-based systems. We will focus on distributed event-based systems as well as MOMs. A comprehensive overview can also be found in [131]. 3.1 Performance Modeling of Event-Based Systems In this section we present an overview of existing performance modeling techniques for eventdriven systems considering both centralized and distributed environments. Modeling of MOM A method for modeling MOM systems using performance completions is presented in [87]. Performance completions provide a general mechanism for including low-level details of execution environments into abstract performance models. A pattern-based language for configuring the type of message-based communication is proposed and model-to-model transformations are used to integrate low-level details of the MOM system into high-level software architecture models. A case study based on part of the SPECjms2007 workload (more specifically Interaction 4) is presented as a validation of the approach. However, no interactions involving multiple message exchanges or interaction mixes are considered. In [141], an approach for predicting the performance of messaging applications based on the Java Enterprise Edition (JavaEE) is proposed. The forecast is carried out during application design, without access to the application implementation. This is achieved by modeling the interactions among messaging components using queueing network models, calibrating the performance models with architecture attributes associated with these components, and populating the model parameters using a lightweight application-independent benchmark. However, again the workloads considered are very simple and do not include any complex messaging interactions. In [88], the dependency between the MOMs usage and its performance was analyzed using statistical inference. For the validation of the approach, parts of SPECjms2007 and jms2009-PS were used. Modeling of Distributed Publish/Subscribe Systems Several performance modeling techniques specifically targeted at distributed publish/subscribe systems [40] exist in the literature. However, such techniques are normally focused on modeling the routing of events through distributed broker topologies from publishers to subscribers as opposed to modeling interactions and message flows between communicating components in event-driven applications. In [115], an analytical model of publish/subscribe systems that use hierarchical identity-based routing is presented. The model is based on continuous time 25
Page 1: Performance Modeling and Benchmarki
Page 4 and 5: priorities. Finally, we validated o
Page 6 and 7: 4.3.4 Tool Extension . . . . . . .
Page 8 and 9: 5.7 Proportions of the Interactions
Page 11 and 12: List of Acronyms Acronym λ t,k j A
Page 13 and 14: Acronym SPEC-OSG SPE SPN SQL ST SUT
Page 15 and 16: Chapter 1 Introduction 1.1 Motivati
Page 17 and 18: PC Server 310 PC Server 310 PC Serv
Page 19 and 20: 1.3. APPROACH AND CONTRIBUTIONS OF
Page 21 and 22: 1.4. THESIS ORGANIZATION 7 • Stan
Page 23 and 24: Chapter 2 Background In this chapte
Page 25 and 26: 2.1. EVENT-BASED SYSTEMS 11 is gene
Page 27 and 28: 2.2. TECHNOLOGY PLATFORMS OF EVENT-
Page 35 and 36: 2.3. INTRODUCTION TO QUEUEING PETRI
Page 37: 2.3. INTRODUCTION TO QUEUEING PETRI
Page 41 and 42: 3.2. BENCHMARKING OF EVENT-BASED SY
Page 43 and 44: 3.2. BENCHMARKING OF EVENT-BASED SY
Page 45 and 46: 3.4. CONCLUDING REMARKS 31 hierarch
Page 47 and 48: Chapter 4 Performance Engineering o
Page 49 and 50: 4.1. MODELING METHODOLOGY FOR EBS 3
Page 57 and 58: 4.2. PERFORMANCE MODELING PATTERN 4
Page 87 and 88: 4.3. EXTENSIONS OF QPNS 73 ,-).$%&'
Page 89 and 90:
4.4. CONCLUDING REMARKS 75 Without
Page 91 and 92:
Chapter 5 Benchmarking of Event-Bas
Page 93 and 94:
5.1. SPECJMS2007 - A STANDARD BENCH
Page 95 and 96:
Page 97 and 98:
Page 99 and 100:
Page 101 and 102:
Page 103 and 104:
Page 105 and 106:
Page 107 and 108:
Page 109 and 110:
Page 111 and 112:
Page 113 and 114:
Page 115 and 116:
5.2. CASE STUDY I: SPECJMS2007 101
Page 117 and 118:
Page 119 and 120:
Page 121 and 122:
Page 123 and 124:
Page 125 and 126:
5.3. JMS2009-PS - A PUBLISH /SUBSCR
Page 127 and 128:
5.3. JMS2009-PS - A PUBLISH /SUBSCR
Page 129 and 130:
5.4. CASE STUDY II: JMS2009-PS 115
Page 131 and 132:
5.4. CASE STUDY II: JMS2009-PS 117
Page 133 and 134:
Chapter 6 Performance Modeling of E
Page 135 and 136:
6.2. MODELING SPECJMS2007 121 Scena
Page 137 and 138:
6.2. MODELING SPECJMS2007 123 BASE*
Page 139 and 140:
6.2. MODELING SPECJMS2007 125 1‘O
Page 141 and 142:
6.2. MODELING SPECJMS2007 127 Sc. 1
Page 143 and 144:
6.2. MODELING SPECJMS2007 129 Perce
Page 145 and 146:
6.2. MODELING SPECJMS2007 131 Serve
Page 147 and 148:
6.3. CONCLUDING REMARKS 133 the mod
Page 149 and 150:
Chapter 7 Conclusions and Outlook W
Page 151 and 152:
7.1. ONGOING AND FUTURE WORK 137 JA
Page 153 and 154:
7.1. ONGOING AND FUTURE WORK 139 th
Page 155 and 156:
Bibliography [1] J. Abbott, K. B. M
Page 157 and 158:
BIBLIOGRAPHY 143 [29] BiCEP Researc
Page 159 and 160:
BIBLIOGRAPHY 145 [61] G. Cugola and
Page 161 and 162:
BIBLIOGRAPHY 147 [92] R. Henjes, M.
Page 163 and 164:
BIBLIOGRAPHY 149 [128] S. Kounev an
Page 165 and 166:
BIBLIOGRAPHY 151 [162] Object Compu
Page 167 and 168:
BIBLIOGRAPHY 153 [197] K. Sachs, S.
Page 169 and 170:
BIBLIOGRAPHY 155 [228] P. Tran, P.
Page 171:
Erklärung Hiermit erkläre ich, di
show all

Performance Modeling and Benchmarking of Event-Based ... - DVS

Create successful ePaper yourself

Delete template?

Save as template?