Product Description
Software, Environments, and Tools 20

Scientific computing has often been called the third approach to scientific discovery, emerging as a peer to experimentation and theory. Historically, the synergy between experimentation and theory has been well understood: experiments give insight into possible theories, theories inspire experiments, experiments reinforce or invalidate theories, and so on. As scientific computing has evolved to produce results that meet or exceed the quality of experimental and theoretical results, it has become indispensable.

Parallel processing has been an enabling technology in scientific computing for more than 20 years. This book is the first in-depth discussion of parallel computing in 10 years; it reflects the mix of topics that mathematicians, computer scientists, and computational scientists focus on to make parallel processing effective for scientific problems. Presently, the impact of parallel processing on scientific computing varies greatly across disciplines, but it plays a vital role in most problem domains and is absolutely essential in many of them.

Parallel Processing for Scientific Computing is divided into four parts: The first concerns performance modeling, analysis, and optimization; the second focuses on parallel algorithms and software for an array of problems common to many modeling and simulation applications; the third emphasizes tools and environments that can ease and enhance the process of application development; and the fourth provides a sampling of applications that require parallel computing for scaling to solve larger and realistic models that can advance science and engineering.

This edited volume serves as an up-to-date reference for researchers and application developers on the state of the art in scientific computing. It also serves as an excellent overview and introduction, especially for graduate and senior-level undergraduate students interested in computational modeling and simulation and related computer science and applied mathematics aspects.

ContentsList of Figures; List of Tables; Preface; Chapter 1: Frontiers of Scientific Computing: An Overview; Part I: Performance Modeling, Analysis and Optimization. Chapter 2: Performance Analysis: From Art to Science; Chapter 3: Approaches to Architecture-Aware Parallel Scientific Computation; Chapter 4: Achieving High Performance on the BlueGene/L Supercomputer; Chapter 5: Performance Evaluation and Modeling of Ultra-Scale Systems; Part II: Parallel Algorithms and Enabling Technologies. Chapter 6: Partitioning and Load Balancing; Chapter 7: Combinatorial Parallel and Scientific Computing; Chapter 8: Parallel Adaptive Mesh Refinement; Chapter 9: Parallel Sparse Solvers, Preconditioners, and Their Applications; Chapter 10: A Survey of Parallelization Techniques for Multigrid Solvers; Chapter 11: Fault Tolerance in Large-Scale Scientific Computing; Part III: Tools and Frameworks for Parallel Applications.Chapter 12: Parallel Tools and Environments: A Survey; Chapter 13: Parallel Linear Algebra Software; Chapter 14: High-Performance Component Software Systems; Chapter 15: Integrating Component-Based Scientific Computing Software; Part IV: Applications of Parallel Computing. Chapter 16: Parallel Algorithms for PDE-Constrained Optimization; Chapter 17: Massively Parallel Mixed-Integer Programming; Chapter 18: Parallel Methods and Software for Multicomponent Simulations; Chapter 19: Parallel Computational Biology; Chapter 20: Opportunities and Challenges for Parallel Computing in Science and Engineering; Index. ... Read more

Product Description
In this book, the authors review important developments and discuss new strategies for the performance evaluation of parallel, distributed and emergent systems. The purpose is to make analytical, simulation and testbed-based modelling and evaluation tools and methodology possible and understandable to a wider audience and to report the state-of-the-art innovative tools and techniques for system performance evaluation. Consequently, this book can serve as a valuable source of reference and indispensable reading for researchers, educators, engineers, graduate students, and practitioners in the field of design and analysis of such systems. ... Read more

Product Description
The primary purpose of this book is to capture the state-of-the-art in Cloud Computing technologies and applications. The book will also aim to identify potential research directions and technologies that will facilitate creation a global market-place of cloud computing services supporting scientific, industrial, business, and consumer applications. We expect the book to serve as a reference for larger audience such as systems architects, practitioners, developers, new researchers and graduate level students. This area of research is relatively recent, and as such has no existing reference book that addresses it.

This book will be a timely contribution to a field that is gaining considerable research interest, momentum, and is expected to be of increasing interest to commercial developers. The book is targeted for professional computer science developers and graduate students especially at Masters level. As Cloud Computing is recognized as one of the top five emerging technologies that will have a major impact on the quality of science and society over the next 20 years, its knowledge will help position our readers at the forefront of the field. ... Read more

Product Description
Divided into three information packed parts, Distributed Computing provides a broad-ranging introduction to the theory of distributed computing. The first part covers the fundamental models, issues, and techniques. Part II is organized around key simulations in distributed computing, and their applications. Part III consists of a sampling of more advanced topics. ... Read more

Customer Reviews (5)

Theoritical text with lots of lemmas.
In my opinion, this book is difficult to read with very few exhibits. The author's comprehension is profound with tons of references. This book should be used between intermediate to an advanced course of Distributed Computing area.

formal maths approach
[A review of the 2nd edition.]
The authors give a mathematically sophisticated analysis of various modes of distributed computing. Where the distribution might refer to separate CPUs in a multiprocessor architecture, or perhaps to separate computers inside a LAN, or to computers scattered across the Internet.

We see that issues of latency and reliability can [and will] arise. Coordinating a task across the processors gives rise to amazing complexity. What if some processors crash? A consensus problem occurs. How to solve it is explained.

There are also impossibilities in task solving that might occur, and these need to be treated carefully. The narrative has suggestions on how to diagnose if such events happen. The reader will see that fault tolerance can be awkward to handle.

The treatment may be too mathematical for some readers. You need a strong background in maths; preferably including discrete maths.

A good book
This is a good book in the field of distributed computing, but it is very mathematically oriented. It concentrates on impossibility proofs and lower bound proofs. I advise people new to this material to read a more descriptive book before reading this one

well-written, in-depth overview of distributed computing
I used this book for teaching an under-graduate primer course in distributed computing. The book is readable, coherent, well-structured and very efficient as a textbook. It strikes a good balance between the sea ofdetails and the basic principles. I am familiar with the core of this booksince it was a collection of lecture notes (alas, no longer available).It's a pity that some important topics have been omitted from the bookversion (e.g., Gallager, Humblett, Spira alg). There are some minor errorsand imperfections in pseudocodes and exercise definitions which are alittle bit annoying. That's why I'm giving this book four points and notfive. Bottom line: I would recommend this book as a course textbook.

Terse prose fraught with errors and omissions
I've struggled to read this text, since it contains many very recent results distilled into intelligently organized chapters. Unfortunately, even though this book is intended for a savvy audience, the text is oftentoo detailed and technical, while important "big picture"intuition is never relayed. Frequent errors in the algorithms and proofs,ranging from simple subscript swaps to more subtle errors in logic to the(rarer) complete lack of logic make this a difficult book to recommend. Inaddition, the exercises are frequently too vague (sometimes meaningless) --this book is definitely not recommended for class work. ... Read more

Product Description

A clear illustration of how parallel computers can be successfully applied
to large-scale scientific computations. This book demonstrates how a
variety of applications in physics, biology, mathematics and other sciences
were implemented on real parallel computers to produce new scientific
results. It investigates issues of fine-grained parallelism relevant for
future supercomputers with particular emphasis on hypercube architecture.

Product Description
* New approaches to parallel computing are being developed that make better use of the heterogeneous cluster architecture
* Provides a detailed introduction to parallel computing on heterogenous clusters
* All concepts and algorithms are illustrated with working programs that can be compiled and executed on any cluster
* The algorithms discussed have practical applications in a range of real-life parallel computing problems, such as the N-body problem, portfolio management, and the modeling of oil extraction ... Read more

Product Description
* Focuses on learning patterns and knowledge from data generated by mobile users and mobile technology.
* Covers research and application issues in applying computational intelligence applications to mobile computing
* Delivers benefits to a wide range of applications
* Introduces the state of the art of computational intelligence to the mobile paradigm ... Read more

Product Description
Live smarter! Discover how intelligent environments are revolutionizing industry, home, and office

Research on smart environments represents some of the most innovative work being done in computer science, electrical engineering, and information systems today. The interdisciplinary scope of the field integrates aspects of machine learning, human-machine interfacing, wireless networking, mobile communications, sensor networks, and pervasive computing.As research efforts devise intelligent environments for use in homes, offices, classrooms, hospitals, and automobiles, far-reaching applications--from design and architecture to health care issues and software engineering--become increasingly evident.

With contributions from leading researchers in a wide array of disciplines, Smart Environments: Technology, Protocols, and Applications presents the state of the art in this emerging field.Important topics covered include:

• Wireless sensor networks
• Pervasive technology
• Middleware
• Home and office networking and appliances
• Prediction algorithms
• Location and estimation techniques
• Automated decision-making
• Privacy and security issues
• Assistive environments for individuals with special needs
• Future trends

Supplementary features include an accompanying Web site and a comprehensive listing of Web sites for smart environment projects.Encompassing both theory and practical implementation, with in-depth discussion of existing applications, Smart Environments is an authoritative resource for practicing engineers and students in the field. ... Read more

Product Description
As more and more hardware platforms support parallelism, parallel programming is gaining momentum. Applications can only leverage the performance of multi-core processors or graphics processing units if they are able to split a problem into smaller ones that can be solved in parallel. The challenges emerging from the development of parallel applications have led to the development of a great number of tools for debugging, performanceanalysis and other tasks. The proceedings of the 3rd International Workshop on Parallel Tools for High Performance Computing provide a technical overview in order to help engineers, developersand computer scientists decide which tools are best suited to enhancing their current development processes. ... Read more

Product Description
This book introduces the basic concepts of parallel and vector computing in the context of an introduction to numerical methods.It contains chapters on parallel and vector matrix multiplication and solution of linear systems by direct and iterative methods.It is suitable for advanced undergraduate and beginning graduate courses in computer science, applied mathematics, and engineering.Ideally, students will have access to a parallel or Vector computer, but the material can be studied profitably in any case.

* Gives a modern overview of scientific computing including parallel an vector computation
* Introduces numerical methods for both ordinary and partial differential equations
* Has considerable discussion of both direct and iterative methods for linear systems of equations, including parallel and vector algorithms
* Covers most of the main topics for a first course in numerical methods and can serve as a text for this course
... Read more

Customer Reviews (1)

As the title says: an introduction
It has all the basic bits that a beginner needs to get started in numerical computation: polynomial approximation, numerical integration, and linear systems. The latter is a real strength, since it offer not just exact techniques (like Gaussian elimination), but iterative techniques including sparse-system special cases and more general conjugate gradient techniques. There isn't much cut&paste code here, but theorems usually appear only as conclusions, not as arguments. The last sections, on iterative solutions of linear systems, give more in the way of concepts than guidance, but a determined reader will find value.

Despite strengths, this book has significant weaknesses. Error analysis is thin, and that's what really sets good analysts ahead of the pack. The book predates wide acceptance of symbolic algebra packages, so it favors Taylor series approximations over the superior but tedious orthogonal polynomials. It notes cache:memory penalties under 1:10, where they're typically over 1:100 today. And its discussion of performance processor architecture barely approaches adequate, even by 1993 standards. That dates back 10 generations of Moore's Law, ten doublings of transistor count or 1000x. That's a lot, and a world with a Blue Gene in it is a very different place.

Still, the basics haven't changed. Despite some obscurity in the later chapters, it's still good for the first few things a numerical programmer needs to know, including a little parallelism awareness. If I were teaching a basic course in scientific computation, it would still be in the running when I went to choose a text.

//wiredweird ... Read more

Product Description
A state-of-the-art guide for the implementation of distributed simulation technology.

The rapid expansion of the Internet and commodity parallel computers has made parallel and distributed simulation (PADS) a hot technology indeed. Applications abound not only in the analysis of complex systems such as transportation or the next-generation Internet, but also in computer-generated virtual worlds for military and professional training, interactive computer games, and the entertainment industry.

In this book, PADS expert Richard M. Fujimoto provides software developers with cutting-edge techniques for speeding up the execution of simulations across multiple processors and dealing with data distribution over wide area networks ,including the Internet. With an emphasis on parallel and distributed discrete event simulation technologies, Dr. Fujimoto compiles and consolidates research results in the field spanning the last twenty years, discussing the use of parallel and distributed computers in both the modeling and analysis of system behavior and the creation of distributed virtual environments.

While other books on PADS concentrate on applications, Parallel and Distributed Simulation Systems clearly shows how to implement the technology. It explains in detail the synchronization algorithms needed to properly realize the simulations, including an in-depth discussion of time warp and advanced optimistic techniques. Finally, the book is richly supplemented with references, tables and illustrations, and examples of contemporary systems such as the Department of Defense's High Level Architecture (HLA), which has become the standard architecture for defense programs in the United States. ... Read more

Customer Reviews (2)

Excellent book as a primer to distributed computing
This book is a very readable introduction to parallel and distributed computing.It presents concepts well, and offers enough examples to allow one to make sure they have a firm grasp on what is being presented.It is NOT a textbook.It is not full of code, or detailed descriptions of exactly how to implement a parallel/distributed system in a given situation.Those already expert in the field or those seeking low-level detail and code should look elsewhere.I recommend this book to everyone else.

#1 Source for Parallel and Distributed Simulation Techniques
Parallel and Distributed Simulation Systems, by RichardFujimoto, brings together all of the leading techniques fordesigning and operating parallel and distributedsimulations.This is the first book to bringthis materialin a single source.Previously, simulation developers had to research a library to journal and conference articles toidentify,master, and select techniques appropriate totheir problem.

Fujimotodivides the material into three sections.Thefirst introduces the readerto the real world problems thathave given rise to the need for PADStechniques. He alsoincludes an introduction to the field of discreteeventsimulation, laying the necessary groundwork for the readertounderstand the rest of the book.

In the second section the book beginsto describe many ofthe techniques that are used to accomplish parallel or distributed operations for discrete event simulations. These arepresented very clearly and most readers willgrasp them immediately.Inthis section, the reader gainsan appreciation for the difficulty involvedin creatingparallel simulations that maintain the causality of events within the entire simulation.Distributing the applicationacrossmultiple computers has the potential to greatlyincrease its executionspeed, but must do so withoutjeopardizing the causal accuracy of thesimulation.Topicscovered include conservative synchronization, timewarp,deadlock avoidance, lookahead, repeatability, globalvirtual time,scheduling, and performance. Fujimoto alsoincludes a description of theGeorgia Tech Time Warp systemin which he and others have implemented manyof thesetechniques.

Finally, the third section describes theapplication ofPADS techniques to distributed virtual environments. Fujimoto focuses distributed military training systems andpoints towardthe potential for applying these techniquesin entertainment applications. This section is motivatedby a description of the evolution ofdistributed militarytraining using protocols like simulator networking (SIMNET), distributed interactive simulation (DIS),aggregate levelsimulation protocol (ALSP), and the morerecent high level architecture(HLA). Fujimoto takes greatcare to describe how PADS techniques are usedwithinsoftware implementations of the HLA. Topics includedistributedinteractive simulation, dead reckoning, thehigh level architecture,network technologies,communication protocols, data distributionmanagement, andmessage ordering.

The book is an essential tool forthose creating parallelor distributed simulation applications orconductingresearch in the field.It provides and excellent level of detail and sufficient explanation to allow others toimplement thetechniques.It is available directly fromWiley-Interscience and frommost of the Web-basedbookstores. ... Read more

Product Description
Foundations of Multithreaded, Parallel, and Distributed Programming covers-and then applies-the core concepts and techniques needed for an introductory course in this topic. The book emphasizes the practice and application of parallel systems, using real-world examples throughout.

Greg Andrews teaches the fundamental concepts of multithreaded, parallel and distributed computing and relates them to the implementation and performance processes. He presents the appropriate breadth of topics and supports these discussions with an emphasis on performance. ... Read more

Customer Reviews (6)

Difficult to follow with bad examples.
I took this class with Greg Andrews and he is as bad of a teacher as he is a textbook author.He is jumpy and scattered while teaching and his book is organized similarly, very hard to follow, doesn't explain things really well.I'm a book learner, and this text is not the right one for the self-taught type.So if you want to learn about parallel computing, try another book, and hopefully this isn't a required text!

Wide and deep on fundamentals, and timely
First a little about me and of what use I found the book.Maybe you are looking to determine whether this book is suitable for your purposes too.I studied this book cover to cover.This took a little over one year.However I have to add, to this day I did not yet solve the example problems, which are plentiful and probably carefully designed and necessary forunderstanding all the ins and outs and cementing the lessons into your head.But I work all day writing software for a living and felt I needed a change of pace, meaning I wanted to spend time studying and absorbing new information after work hours this year rather than writing still more code atnight for now.I've done the night coding thing many a time already.I undoubtedly will shortly be writing code using the book's concepts.That's really the point of me reading it: to write a lot more concurrent code, and to do it better.To my detriment I did not have the benefit of enrolling in a real upper level undergraduate course or a graduate course as guided by an instructor as is probably more typically the case with this textbook's readers.To my benefit, during my day work I had already designed and developed some relatively successful (meaning people are actually using them today, and they paid some money for it) multithreaded, parallel, and distributed applications, such as a web statistics and analytics clustered parallel application, and communication middleware of an access control system.At mid-career it occurred to me that having already made concurrent systems by relying on my cultivated developer's intuition and OJT, now was maybe a good time to try to fill the gaps in my knowledge of concurrent programming and climb to higher levels of mastery in this specialty.

Parallelism is making a strong comeback again in all sorts of systems small and large, such as GPUs and multicore CPUs as well as the giant Beowulf clusters and the classical manyprocessor supercomputers.The heat generation that comes with ever faster clock cycles has put up a pretty tough barrier in the way for the famous CPU chip makers.You can see for yourself that the rise in the gigahertz numbers has flattened out lately, and now it's the number of cores which is beginning to rise in commodity computing equipment, instead of the frequency of (serial) operations.The new parallel generation is happening now after a decade or two of diversion into emphasizing faster sequential processors, Intel and AMD being the notables in that effort.Well now the Intels and AMDs have pushed sequential processing and hardware instruction pipelining really fast but maybe they also have found the thermal limits, and the memory speed limits, and exploited all the pipelining and predictive branching that they could, and now they have to find something else to make similar progress going forward to make Mr. Moore and his Law work right again.Increasing the total aggregate throughput of operations executed across multiple processors or cores appears to be a way forward now in computing performance.

I needed some good, thorough material to perform as the center pillar of a concurrent programming learning initiative.

To select this book over the other textbooks that are available, whereas some seemed specialized or narrow in scope, and some were too formal or dense in their presentation for do-it-yourselfers, I noticed on the web that a pretty large number of universities are using the textbook by Andrews for their introductory course in parallel programming or concurrent programming.

Now notice that the book was copyright in 2000 which is almost a decade ago now which is a bit of a limitation, yet the reality is this is not quite a problem. You can fill in the rare gap using other sources of information, like also using other books, as well as online course lecture videos graciously provided for free public consumption at web sites like the Cloudera company, and the MIT and UC Berkeley university web sites.Keep in mind there is a reason why a book becomes a classic and keeps being used at universities.Consequently however, the famous MapReduce is not really represented in the book.Globus, an earlier distributed framework, is mentioned though.Google and Yahoo and Facebook, and other such sites who are now programming innovatively in the very large for mostly nonscientific applications, would not yet hit the big time and share their concurrent computing innovations with the public until a couple more years into the future, when this book was written.

In my opinion today's massively parallel applications underpinning a few of the famous web sites might well now have some of the world's biggest concurrent application clusters even rivaling supercomputers, since the "supers" don't seem to concurrently use "clusters of clusters" linked across the whole planet, like MapReduce already does every day for millions of users.The traditional supercomputers, on the other hand, even the biggest, baddest ones who are hitting the top 500 fastest lists, seem to be located at just one site at a time if I'm not mistaken.And I'm not talking about content caching or simple load balancing at Google, because GFS and Mapreduce as a parallel coordination language is much more than simple web site front ending.Mapreduce is an application development framework.I suspect Google and its globe-spanning cluster application might be even faster than any of the world's fastest single-site highly parallel supercomputers doing the atom bomb development simulations and cryptanalysis and communications traffic analytics for the DOE or military.The actual numbers are unknown it seems but I suspect the world's largest Beowulf cluster is already in use at Google and they might already be achieving application and system concurrency across perhaps a half-million compute nodes.

Also, Single Instruction Multiple Data (SIMD) programming is not covered enough to my taste in Andrews book.Yet I want to program some massively parallel SIMD GPUs as seen a lot lately in daughterboards or "video cards."There is a30-processor GPU with thousands of parallel hardware threads, organized in a multi-level thread/warp/block hierarchy, with its own separate NUMA memory subsystem, running in my workstation right now as I write this review.Also, I understand Cray just announced their intent to include SIMD GPUs in an upcoming supercomputer.So SIMD is making a comeback.But the book provides nearly no instruction for learning SIMD design and coding.I am left to prepare for SIMD using other sources like the vendor-specific NVidia CUDA documentation or perhaps the nascent OpenCL language.SIMD computing has existed long before year 2000, but at the time of the book, SIMD had already fallen out of favor apparently, because Multi Instruction Multiple Data (MIMD) architecture had pretty much taken over the computer scientists' attention, and this book reflects that.

Despite the minor quibbles, it is accurate to say the book has a broad coverage of topics within the field.And, at the same time there is the depth and detail such that the reader will develop a feeling of being equipped for a good amount of languages or communication models you will eventually select whatever the job at hand.Like the author says, you can only put so many pages into a single textbook. There are probably whole books out there about any one of the chapters in this book.The bread and butter skill of how to effectively think about concurrent and parallel and distributed systems of nearly all types is presented with clarity and simplicity.This benefit is really the strength of this book.There are many examples built from different models and approaches so you get a sense of what works in what situation.There are also plenty of pseudocode and near-code examples in many languages.Make no mistake, there is also a significant amount of detail and depth of instruction on the essentials such as building correct and high performance and fair critical sections of shared memory.The reader develops a sense for fine grained and coarse grained concurrency; effective control of nondeterministic instruction histories; shared memory versus distributed memory programming; parallel and sequential and distributed and concurrent programming (each is different); concurrent systems-level programming versus concurrent applications-level programming; surveys of important features in different languages including their strengths and weaknesses with regard to the suitability for your system including hardware, network, and software; and parallelizing compilers and language abstractions.You will develop readiness to tackle situations with Ps and Vs (semaphores), monitors, message passing, pthreads, and critical sections.

Now, please put aside the tone of the minor criticisms I told you earlier.Andrews book is far more timely than you might have been thinking.Let me demonstrate.Yesterday a brand new language named "Go" was made available to the public by Google.Google appears to be presenting Go as an open-source concurrent-friendly systems programming language.Reading the tutorial for Go, one can see Go provides intrinsics which largely mimic the CSP style of synchronous message passing for interprocess communication and synchronization.Anyone who reads Andrews book will spot the CSP similarity quickly in the new Go language.Moreover the reader will also bring a good preparation on how to use Go's communication model.That's because Andrews' book has provided its readers with good instruction on the synchronous message passing model.Having read Andrews' book, I expect you are better prepared to start programming with what is perhaps one of the more sophisticated and essential parts of the Go language, its intrinsic message passing model.I would suggest that someone who approaches Go without any preparatory knowledge of CSP's guarded synchronous communications model might be at risk of getting mired down in confusion for a while.Go supports concurrency, yet you will not find multithreading and forks and joins in it if I'm understanding it correctly.Those won't be seen explicitly because with message passing those are absent.Perhaps multithreading is the only concurrent technology that some programmers have used, especially if C and pthreads or java or C#.net are among your other area of expertise.My suggestion is to pick up a copy of Andrews book if you want to program some concurrent systems using message passing like Go's.I don't expect a vendor's language tutorial is going to have the time and space to provide all the educational tools, such as a comparitive analysis in different application situations. The book explains the comm and synch model in the context of several different examples;and further, you would want to know where this communication and synchronization model is efficient and simple to code, versus where it is not so simple at times, and this is illustrated as well.Asynchronous message passing code using MPI can be simpler than synchronous code (seen in Go or CSP) depending on the application at hand.Also, maybe you can use CSP to model your Go application.That's because CSP is not just a programming language; it's also a modeling tool which may be useful for Go algorithm design work.

The book was (relatively) easy to read independently compared to some other books I've purchased on similar subjects.The worse ones lay dormant on my shelf.My personal copy of the Andrews book has begun showing some wear and real usage... and that's a good thing.The author obviously knows a lot on this subject, yet he still is able to conjure the sympathy and patience for the beginner. He tends to provide the details that you cannot take for granted in the background of a nonexpert (future expert).Some other books by contrast seem to make too-big logical jumps, or use a terseness that leaves you shaking your head and rereading the same sentences over and over like something was left out.It is apparent that more than just a computer scientist, Andrews is also a teacher of knowledge.Independent students can read his book effectively because getting stuck and frustrated is rare, an important boon in the absence of an expert instructor.

As Andrews said up front, while his book is broad and has also some depth where it's needed, it is still not the only book you will need.This is not much of a problem in the big picture, especially with the web.I am finding good satisfaction in having studied Andrews' broad introductory book closely.Moving ahead from there, it was useful to study complementary materials like the UC Berkely ParLab short course lecture videos (free), Prof. Demmel's CS267 Apps of Parallel Computers course lecture videos (also free), the linear algebra and calculus math material available on free video at MIT's Opencourseware site, and by reading the more narrowly-scoped but interesting textbook Patterns of Parallel Programming by Mattson et al.

Overall, Prof. Andrews' book is a strong basis for learning the principles of design and programming of parallel, concurrent, and distributed systems.

Dr. Andrews knows this topic
The author of this text, Dr. Andrews, has dealt with the theory and implementation of parallel/multithreaded/distributed in computer systems since the 70s. I was fortunate to take his class at the University of Arizona in which this book was used as the primary text.

Unlike many textbooks of its ilk, Dr. Andrews does use coded examples, but they are not complex code excerpts that span several pages. He does an excellent job of covering the topic in both C with Posix, Java, as well as the language he worked on MPD. Since this topic has been his primary focus he really knows the subject matter yet can explain it in a way such that anyone with moderate programming skills can grasp.

Just like his lectures, the fundamentals and theory presented in each chapter is always structured, explained, and numerous examples are given to reinforce the topics that are being taught. I would recommend this book to anyone who requires an introductory to medium exposure to the critical topic of multithreaded, parallel and distributed programming.

An Excellent Book
I used this book for Dr. Andrews' parallel computing class many years ago and LOVED this book.This book covers a wide ranging topics of parallel computing and is a good enough reference for 99% people out there.

The book is extremely useful in that it provides actual working example codes for nearly all the topics covered, in C or Java or both.The codes are also very small, in most cases less than a page.This is very important because a lot cases such as multiple readers single writers are not easy to code from scratch and could easily have synchronization problems unless one has strong overall grasp of the concept.This book does very good analysis of potential pitfalls of them.Even if you already knew the concepts, this book provides a valuable reference and code templates.

Some part of the book may seem to be pseudo code to some people since Dr. Andrews uses the MPD language (which is on top of C) for a far easier time dealing with many of the computing issues.Although I have never used to MPD language, I find the syntax useful in understanding many concepts Dr. Andrews is trying to explain, such as how to partition work into small tasks.

I've never read anything worse than this in any subject ever
I've read a few books on computer science and mathematics, being a university student in both subjects. However, I've never read anything nearly as conjested as this book. Basically its 665 pages of randomly mixed crap. I wonder if Amazon is going to censor this review, still this is what I think about this book.

Many parts of this books shows that the author has a very poor understanding of mathematics, logical deduction and also pays to attention to details. Further he seems to have quite a bit of flawed intuition about programs/processes and threads. Okay that might be harsh but atleast he has no ability to communicate his understanding anyway so...

More specifically I disagree with the following things;
The author repeats definitions (sometimes three or more times!).
The author does not explain his weird pseudo-code notation which I additionally think is counter-intuitive. He presents many copies of the same snippet with the first few versions beeing incorrect and the versions only differ my a few lines. Great way to take of two whole pages with minimal actual information content.

The one thing that made be go online and write this was page 73 (i've not read futher and i'm not sure i'm going to either) where the author delivers a lengthy insult to ones intellect by telling you that a program is BAD if any possible trace of execution leave the program in a BAD state. Further he continues to write that GOOD is equivalent to BAD, and then uses these two concepts as if they were each others logical negations. ... Read more

Product Description
A new model for task scheduling that dramatically improves the efficiency of parallel systems

Task scheduling for parallel systems can become a quagmire of heuristics, models, and methods that have been developed over the past decades. The author of this innovative text cuts through the confusion and complexity by presenting a consistent and comprehensive theoretical framework along with realistic parallel system models. These new models, based on an investigation of the concepts and principles underlying task scheduling, take into account heterogeneity, contention for communication resources, and the involvement of the processor in communications.

For readers who may be new to task scheduling, the first chapters are essential. They serve as an excellent introduction to programming parallel systems, and they place task scheduling within the context of the program parallelization process. The author then reviews the basics of graph theory, discussing the major graph models used to represent parallel programs. Next, the author introduces his task scheduling framework. He carefully explains the theoretical background of this framework and provides several examples to enable readers to fully understand how it greatly simplifies and, at the same time, enhances the ability to schedule.

The second half of the text examines both basic and advanced scheduling techniques, offering readers a thorough understanding of the principles underlying scheduling algorithms. The final two chapters address communication contention in scheduling and processor involvement in communications.

Each chapter features exercises that help readers put their new skills into practice. An extensive bibliography leads to additional information for further research. Finally, the use of figures and examples helps readers better visualize and understand complex concepts and processes.

Researchers and students in distributed and parallel computer systems will find that this text dramatically improves their ability to schedule tasks accurately and efficiently. ... Read more

Customer Reviews (1)

decompose a program into a directed graph
As computer hardware continues to improve, along the lines of Moore's Law, software improvements have lagged. This has been particularly so with a system of parallel machines. The biggest problem is how to translate a given algorithm's implementation, that was written for a single CPU, into one for several CPUs. More to the point, how to do this systematically, instead of in a manual, ad hoc fashion?

That is the promise of Sinnen's book. It takes a program, and finds a task graph. That is, the code is decomposed into a set of smaller modules. Each module can be taken to be a node of a directed graph. The directed nature of the graph is essential to the discussion. The direction of a vertex indicates a dependency on another node. Given this graph, and a set of CPUs, the book explains how to allocate the nodes across the CPUs, in a manner that is optimal.

Important related concepts are developed. Like clustering heuristics. These are used to form groups or clusters of modules. To a programmer, think of the concept of recursion. That's the utility of clustering.

The text is heavily mathematical, as befits any application of graph theory. The crucial part of the text is at its start, as you might surmise. It explains how to algorithmically extract the dependence graph, which is a representation of the dependence structure of the initial program.

Potentially very useful. ... Read more

Product Description
Although the computing demands of real-time signal processing and control applications are increasing rapidly, parallel processors permit several instructions to be dealt with simultaneously so that the "real-time" needed is manageable. This book introduces the advantages of this strategy and details how to use parallel processing to deal with common signal processing and control algorithms. It emphasises the relationship between the computing requirements of algorithms and the appropriate choice of architectures, while demonstrating how to identify processor capabilities and how to exploit them to the fullest. The text includes examples and end-of-chapter exercises to facilitate self- and group study and case studies to put theoretical concepts into a practical context. For advanced students in parallel computing, control and signal processing disciplines, it is an invaluable tool in learning to get the most from their computer systems. ... Read more

Product Description
Rapid changes in the field of parallel processing make this book especially important for professionals who are faced daily with new products—and provides them with the level of understanding they need to evaluate and select the products. It gives readers a fundamental understanding of parallel processing application and system development.Chapter topics include parallel machines and computations, potential for parallel computations, vector algorithms and architectures, MIMD computers and multiprocessors, distributed memory processors, interconnection networks, data dependence and parallelism, implementing synchronization and data sharing, parallel processor performance, temporal behavior of parallel programs, and parallel I/O.For computational scientists, software engineers, computer architects, and computer engineers. ... Read more

Product Description
Volume 2 discusses programming environments and development tools, Java as a language of choice for development in highly parallel systems, and state-of-the-art high performance algorithms and applications. DLC:High performance computing. ... Read more

Customer Reviews (1)

Parallel is the way to go
A pleasant read but I think the Wilkenson and Allen book called Parallel Computing is better written.All in all it wasn't bad and I'd recommend it to anyone looking for a secondary source. ... Read more

Product Description

Author Peter Pacheco uses a tutorial approach to show students how to develop effective parallel programs with MPI, PThreads, and OpenMP. The first undergraduate text to directly address compiling and running parallel programs on the new multi-core and cluster architecture, An Introduction to Parallel Programming explains how to design, debug, and evaluate the performance of distributed and shared-memory programs. User-friendly exercises teach students how to compile, run and modify example programs.

Product Description
ParCo2007 marks a quarter of a century of the international conferences on parallel computing that started in Berlin in 1983. The aim of the conference is to give an overview of the state-of-the-art of the developments, applications and future trends in high performance computing for all platforms. The conference addresses all aspects of parallel computing, including applications, hardware and software technologies as well as languages and development environments. Special emphasis was placed on the role of high performance processing to solve real-life problems in all areas, including scientific, engineering and multidisciplinary applications and strategies, experiences and conclusions made with respect to parallel computing. The book contains papers covering: 1) Applications; The application of parallel computers to solve computationally challenging problems in the physical and life sciences, engineering, industry and commerce. The treatment of complex multidisciplinary problems occurring in all application areas was discussed. 2) Algorithms; Design, analysis and implementation of generic parallel algorithms, including their scalability, in particular to a large number of processors (MPP), portability and adaptability and 3) Software and Architectures; Software engineering for developing and maintaining parallel software, including parallel programming models and paradigms, development environments, compile-time and run-time tools. A number of symposia on specialized topics formed part of the scientific program. The following topics were covered: Parallel Computing with FPGAs, The Future of OpenMP in the Multi-Core Era, Scalability and Usability of HPC Programming Tools, DEISA: Extreme Computing in an Advanced Supercomputing Environment and Scaling Science Applications on Blue Gene. The conference was organized by the renowned research and teaching institutions Forschungszentrum Julich and the RWTH Aachen University in Germany.

IOS Press is an international science, technical and medical publisher of high-quality books for academics, scientists, and professionals in all fields.

Some of the areas we publish in:

-Biomedicine
-Oncology
-Artificial intelligence
-Databases and information systems
-Maritime engineering
-Nanotechnology
-Geoengineering
-All aspects of physics
-E-governance
-E-commerce
-The knowledge economy
-Urban studies
-Arms control
-Understanding and responding to terrorism
-Medical informatics
-Computer Sciences ... Read more

Product Description
The first edition of The Art of Parallel Programming was adopted as a course text by 62 Universities. Now completely revised and updated, this book provides in-depth coverage of the major organizational techniques for parallel programs, including specific parallel algorithms to illustrate these techniques. Author Bruce P. Lester covers topics such as data partitioning, multiprocessor architecture, synchronous parallelism, MPI standard library, communication streams, data sharing, replicated workers, and much more. Key Features of the book include: (1) Parallel algorithms from many areas, including matrix algorithms, graph algorithms, solution to linear equations, sorting, solution to differential equations, image processing, discrete optimization, and communication algorithms. (2) Two comprehensive chapters on parallel computer architecture. (3) Practical performance analysis methods combining mathematical analysis of algorithms with measurement of actual parallel program execution time. (4) Techniques to avoid common sources of performance degradation, such as memory contention, communication delay, load imbalance, synchronization delay, network congestion, process creation time. (5) Practical parallel programming skills developed through short programming exercises and more extensive programming projects using the C language. (6) A free software package containing a Parallel Computer Simulation System that runs on any ordinary computer, allowing you to gain experience with parallel programming without the need for an expensive parallel computer system. ... Read more

Product Description
Essential reading to understand patterns for parallel programming

Software patterns have revolutionized the way we think about how software is designed, built, and documented, and the design of parallel software requires you to consider other particular design aspects and special skills. From clusters to supercomputers, success heavily depends on the design skills of software developers.

Patterns for Parallel Software Design presents a pattern-oriented software architecture approach to parallel software design. This approach is not a design method in the classic sense, but a new way of managing and exploiting existing design knowledge for designing parallel programs. Moreover, such approaches enhance not only build-time properties of parallel systems, but also, and particularly, their run-time properties.

• Features known solutions in concurrent and distributed programming, applied to the development of parallel programs
• Provides architectural patterns that describe how to divide an algorithm and/or data to find a suitable partition and link it with a programming structure that allows for such a division
• Presents an architectural point of view and explains the development of parallel software

Patterns for Parallel Software Design will give you the skills you need to develop parallel software. ... Read more