Details

Solving Enterprise Applications Performance Puzzles


Solving Enterprise Applications Performance Puzzles

Queuing Models to the Rescue
1. Aufl.

von: Leonid Grinshpan

69,99 €

Verlag: Wiley
Format: PDF
Veröffentl.: 06.02.2012
ISBN/EAN: 9781118161890
Sprache: englisch
Anzahl Seiten: 256

DRM-geschütztes eBook, Sie benötigen z.B. Adobe Digital Editions und eine Adobe ID zum Lesen.

Beschreibungen

Poorly performing enterprise applications are the weakest links in a corporation's management chain, causing delays and disruptions of critical business functions. This groundbreaking book frames enterprise application performance engineering not as an art but as applied science built on model-based methodological foundation. The book introduces queuing models of enterprise application that visualize, demystify, explain, and solve system performance issues. Analysis of these models will help to discover and clarify unapparent connections and correlations among workloads, hardware architecture, and software parameters.
<p>Acknowledgments ix</p> <p>Preface xi</p> <p><b>1. Queuing Networks as Applications Models 1</b></p> <p>1.1. Enterprise Applications—What Do They Have in Common? 1</p> <p>1.2. Key Performance Indicator—Transaction Time 6</p> <p>1.3. What Is Application Tuning and Sizing? 8</p> <p>1.4. Queuing Models of Enterprise Application 9</p> <p>1.5. Transaction Response Time and Transaction Profile 19</p> <p>1.6. Network of Highways as an Analogy of the Queuing Model 22</p> <p>Take Away from the Chapter 24</p> <p><b>2. Building and Solving Application Models 25</b></p> <p>2.1. Building Models 25</p> <p>Hardware Specification 26</p> <p>Model Topology 28</p> <p>A Model’s Input Data 29</p> <p>Model Calibration 31</p> <p>2.2. Essentials of Queuing Networks Theory 34</p> <p>2.3. Solving Models 39</p> <p>2.4. Interpretation of Modeling Results 47</p> <p>Hardware Utilization 47</p> <p>Server Queue Length, Transaction Time, System Throughput 51</p> <p>Take Away from the Chapter 54</p> <p><b>3. Workload Characterization and Transaction Profiling 57</b></p> <p>3.1. What Is Application Workload? 57</p> <p>3.2. Workload Characterization 60</p> <p>Transaction Rate and User Think Time 61</p> <p>Think Time Model 65</p> <p>Take Away from the Think Time Model 68</p> <p>Workload Deviations 68</p> <p>“Garbage in Garbage out” Models 68</p> <p>Realistic Workload 69</p> <p>Users’ Redistribution 72</p> <p>Changing Number of Users 72</p> <p>Transaction Rate Variation 75</p> <p>Take Away from “Garbage in Garbage out” Models 78</p> <p>Number of Application Users 78</p> <p>User Concurrency Model 80</p> <p>Take Away from User Concurrency Model 81</p> <p>3.3. Business Process Analysis 81</p> <p>3.4. Mining Transactional Data from Production Applications 88</p> <p>Profiling Transactions Using Operating System Monitors and Utilities 88</p> <p>Application Log Files 90</p> <p>Transaction Monitors 91</p> <p>Take Away from the Chapter 93</p> <p><b>4. Servers CPUs and Other Building Blocks of Application Scalability 94</b></p> <p>4.1. Application Scalability 94</p> <p>4.2. Bottleneck Identification 95</p> <p>CPU Bottleneck 97</p> <p>CPU Bottleneck Models 97</p> <p>CPU Bottleneck Identification 97</p> <p>Additional CPUs 100</p> <p>Additional Servers 100</p> <p>Faster CPUs 100</p> <p>Take Away from the CPU Bottleneck Model 104</p> <p>I/O Bottleneck 105</p> <p>I/O Bottleneck Models 106</p> <p>I/O Bottleneck Identification 106</p> <p>Additional Disks 107</p> <p>Faster Disks 108</p> <p>Take Away from the I/O Bottleneck Model 111</p> <p>Take Away from the Chapter 113</p> <p><b>5. Operating System Overhead 114</b></p> <p>5.1. Components of an Operating System 114</p> <p>5.2. Operating System Overhead 118</p> <p>System Time Models 122</p> <p>Impact of System Overhead on Transaction Time 123</p> <p>Impact of System Overhead on Hardware Utilization 124</p> <p>Take Away from the Chapter 125</p> <p><b>6. Software Bottlenecks 127</b></p> <p>6.1. What Is a Software Bottleneck? 127</p> <p>6.2. Memory Bottleneck 131</p> <p>Memory Bottleneck Models 133</p> <p>Preset Upper Memory Limit 133</p> <p>Paging Effect 138</p> <p>Take Away from the Memory Bottleneck Model 143</p> <p>6.3. Thread Optimization 144</p> <p>Thread Optimization Models 145</p> <p>Thread Bottleneck Identification 145</p> <p>Correlation Among Transaction Time, CPU Utilization, and the Number of Threads 148</p> <p>Optimal Number of Threads 150</p> <p>Take Away from Thread Optimization Model 151</p> <p>6.4. Other Causes of Software Bottlenecks 152</p> <p>Transaction Affinity 152</p> <p>Connections to Database; User Sessions 152</p> <p>Limited Wait Time and Limited Wait Space 154</p> <p>Software Locks 155</p> <p>Take Away from the Chapter 155</p> <p><b>7. Performance and Capacity of Virtual Systems 157</b></p> <p>7.1. What Is Virtualization? 157</p> <p>7.2. Hardware Virtualization 160</p> <p>Non-Virtualized Hosts 161</p> <p>Virtualized Hosts 165</p> <p>Queuing Theory Explains It All 167</p> <p>Virtualized Hosts Sizing After Lesson Learned 169</p> <p>7.3. Methodology of Virtual Machines Sizing 171</p> <p>Take Away from the Chapter 172</p> <p><b>8. Model-Based Application Sizing: Say Good-Bye to Guessing 173</b></p> <p>8.1. Why Model-Based Sizing? 173</p> <p>8.2. A Model’s Input Data 177</p> <p>Workload and Expected Transaction Time 177</p> <p>How to Obtain a Transaction Profile 179</p> <p>Hardware Platform 182</p> <p>8.3. Mapping a System into a Model 186</p> <p>8.4. Model Deliverables and What-If Scenarios 188</p> <p>Take Away from the Chapter 193</p> <p><b>9. Modeling Different Application Configurations 194</b></p> <p>9.1. Geographical Distribution of Users 194</p> <p>Remote Office Models 196</p> <p>Users’ Locations 196</p> <p>Network Latency 197</p> <p>Take Away from Remote Office Models 198</p> <p>9.2. Accounting for the Time on End-User Computers 198</p> <p>9.3. Remote Terminal Services 200</p> <p>9.4. Cross-Platform Modeling 201</p> <p>9.5. Load Balancing and Server Farms 203</p> <p>9.6. Transaction Parallel Processing Models 205</p> <p>Concurrent Transaction Processing by a Few Servers 205</p> <p>Concurrent Transaction Processing by the Same Server 209</p> <p>Take Away from Transaction Parallel Processing Models 213</p> <p>Take Away from the Chapter 214</p> <p>Glossary 215</p> <p>References 220</p> <p>Index 223</p>
<p><strong>Leonid Grinshpan</strong> Ph.D. (Stamford, CT) has been working for Oracle Corporation for more than 14 years as Consulting Technical Director. His customers include Dell, Citibank, Verizon, Clorox, Bank of America, AT&T, Best Buy, Aetna and many other major enterprises. He obtained Ph.D. in queuing models of computer systems from Russian Academy of Science.
<b>A groundbreaking scientific approach to solving enterprise applications performance problems</b> <p>Enterprise applications are the information backbone of today's corporations, supporting vital business functions such as operational management, supply chain maintenance, customer relationship administration, business intelligence, accounting, procurement logistics, and more. Acceptable performance of enterprise applications is critical for a company's day-to-day operations as well as for its profitability.</p> <p>Unfortunately, troubleshooting poorly performing enterprise applications has traditionally relied upon a trial-and-error series of actions rather than a systematic, scientific methodology.</p> <p>Written by the Consulting Technical Director of Oracle Corporation, one of the largest enterprise software providers in the world, this groundbreaking book frames enterprise applications performance engineering as an applied science built on a model-based foundation. Readers will discover how queuing models of enterprise applications can be used to visualize, demystify, explain, and help to solve the performance issues of many applications. The author demonstrates how these queuing models help discover the critical connections among users' workloads, hardware architecture, and software parameters.</p> <p>Without resorting to complex mathematics, <i>Solving Enterprise Applications Performance Puzzles</i> covers such important topics as:</p> <ul> <li> <p>Queuing networks as application models</p> </li> <li> <p>Building and solving application models</p> </li> <li> <p>User's workload characterization</p> </li> <li> <p>Servers, CPUs, and other building blocks of application scalability</p> </li> <li> <p>Software bottlenecks</p> </li> <li> <p>Performance and capacity of virtual systems</p> </li> <li> <p>Model-based application sizing</p> </li> </ul> <p>Throughout the book, readers will find tables and charts summarizing and visualizing important data. In addition, readers will find plenty of illustrations to help them better understand the core drivers of application performance.</p> <p><i>Solving Enterprise Applications Performance Puzzles</i> offers performance engineers, programmers, and systems analysts a new perspective on why enterprise applications can underperform and fail to support businesses. It puts forward a scientific approach for discovering the underlying problems and correcting them.</p>
<i>Solving Enterprise Applications Performance Puzzles : Queuing Models to the Rescue</i> by Leonid Grinshpan is a pretty interesting book about application of queuing models to solving enterprise performance and I believe the book fills a few gaps in practical application of queuing theory. Another good name for this book could be “Building queuing models by example”. <p>I spent a lot of time trying to use queuing models to solve practical performance issues and would testify that it is pretty challenging. There are a few areas where it was developed a little further (for example, around capacity planning of existing systems), but if you trying to do something else – you won’t find much help. You have a lot of books about systems performance, you have a lot of books about queuing theory with simple examples, but not much in between to solve practical tasks. And here Leonid’s book may help, especially if you are new in this area.</p> <p>Chapter 1, Queuing Networks as Applications Models, is an introduction into the topic. It discusses how queuing theory may be used to model enterprise applications. A lot of analogues are used to introduce the subject.</p> <p>Chapter 2, Building and Solving Application Models, is an overview of the whole process, including short discussions about essentials of queuing theory and using of tools to solve models.</p> <p>Chapter 3, Workload Characterization and Transaction Profiling, discusses what input data for models are and how to gather them.</p> <p>Chapter 4, Servers, CPUs, and Other Building Blocks of Application Scalability, discusses scalability, bottlenecks, how to identify bottlenecks and ways to fix them (mostly on CPU and I/O examples).</p> <p>Chapter 5, Operating System Overheads, discusses main components of operating systems, where overheads come from, how to measure them, and their impact on transaction time.</p> <p>Chapter 6, Software Bottlenecks, is devoted to software bottlenecks, which are rarely discussed in application to queuing models – while in practice software bottlenecks happen all the time. Memory bottlenecks and thread optimizations and their modeling are discussed in details. Multiple other software bottlenecks are also reviewed.</p> <p>Chapter 7, Performance and Capacity of Virtual Systems, is an overview of performance issues related to virtualization , their explanation with queuing theory, and a methodology of virtual machine sizing.</p> <p>Chapter 8, Model-Based Application Sizing: Say Good-Bye to guessing, explains why to use model-based sizing and discusses it step-by-step from gathering input data to model deliverables and what-if scenarios.</p> <p>Chapter 9, Modeling Different Application Configurations, discusses several specials cases including geographical distribution of users, cross-platform modeling, remote terminal services, load balancing, and parallelization of transactions.</p> <p>The book covers a lot of topics. However, to avoid disappointments, I’d like to point out what this book is not:</p> <p>- It is not a textbook about queuing theory. The section 2.2 Essentials of Queuing Networks Theory has 5 pages in it.</p> <p>- It is not a book about tools to solve queuing models. Available tools are listed and there are references, but they are just mentioned as a way to solve models (with one tool used as an illustration of the process). You don’t need to know any tool to read the book (but you will need one when you try to solve your own models).</p> <p>- It is not a comprehensive book about enterprise application performance. There is plenty of important information and practical recommendations about enterprise application performance in the book, but it is shared as needed to build models and analyze their results.</p> <p>So the book is exactly what the title says: a practical book about building queuing models to investigate enterprise applications performance issues.</p> <p>- Alexander Podelko, Oracle</p>

Diese Produkte könnten Sie auch interessieren:

Cyber Security and Digital Forensics
Cyber Security and Digital Forensics
von: Mangesh M. Ghonge, Sabyasachi Pramanik, Ramchandra Mangrulkar, Dac-Nhuong Le
PDF ebook
190,99 €
Smart Systems for Industrial Applications
Smart Systems for Industrial Applications
von: C. Venkatesh, N. Rengarajan, P. Ponmurugan, S. Balamurugan
EPUB ebook
190,99 €
Green Internet of Things and Machine Learning
Green Internet of Things and Machine Learning
von: Roshani Raut, Sandeep Kautish, Zdzislaw Polkowski, Anil Kumar, Chuan-Ming Liu
EPUB ebook
190,99 €