Customer Reviews

The Grid: Blueprint for a New Computing Infrastructure (The Elsevier Series in Grid Computing)

5 star:		(5)
4 star:		(1)
3 star:		(0)
2 star:		(2)
1 star:		(1)

 

 

The Grid seems like a book from the future. I've bought over 100 books related to the Internet looking for ideas that could lead to a public company, but this one is the best. The Grid starts with simple comparisons (Chicago came from rail and 'caching' of grain elevators and stock yards) then gets technical. The topics are covered by experts and include distributed computing, sensing,and teleimmersion, programming tools, services, schedulers, resource management, visualization, security, protocols, Quality of Service, operating systems and interfaces. The pen-ultimate section was co-authored by recently deceased Internet Society head Jon Postel, and is my favorite. If covers the past, present and future of network infrastructure. The last section is test beds. At the risk of seeming ungrateful for this gold mine of future net business, there are a few omissions that I missed including the Grid in mixed environments. Low-earth orbit satellites and wireless IP broadband could have been covered, as they will be the very important parts of the Grid. The VR section totals only about ten pages and, surprisingly, doesn't even touch on entertainment applications, though entertainment (including porn) has driven many 'seeds of the grid', including video, CD-ROMs, and streaming video. A few books that might be interesting: Peter Glaser's Solar Power Satellites is very complementary: with cheap power everywhere, the grid can cover the earth, seas and even leap up into space. If you haven't read it already, I'd toss Kurzweil's Age of Intelligent Machines into the Amazon shopping cart to fill in the AI and VR gap. If you want to see how grids could grow into gods, I'd also highly recommend David Zindell's The Wild. I'm open to corresponding about where The Grid goes and grows from here, especially from investors with the know how or desire to capitalize on the new entities that will grow out of the grid, or be used to create it.

I don't understand some of the critical reviews of this book. The fact is that Grids do exist today and are being used by Research oriented companies. About two dozen companies are already serving clients in the realm of grid computing. Platform Computing, United Devices, and Avaki are just three companies who are helping to create the future of the grid computing. The web succeded because it connected everybody. Until larger grids are contructed for business enterprises I agree that grid computing will not grow. The book does a great job of showing what the future may be like in terms of grid computing. Someday your computing resources will come from your local Grid Computing Company just as you get electricity from your Power Company. How that comes to be is still the ultimate question.

There is a lot of hype these days about the Next Generation Internet (NGI), the National Science Foundation's High Performance Connections Program to very-high broadband connection services (NSF HPC vBNS), and the University Corporation for Advanced Internet Development's Internet2 Project (UCAID I2). I'd strongly recommend that anyone who is trying to figure out the what? and why? and who? of the emerging advance networks take time to read "The Grid: Blueprint for a New Computing Infrastructure." This weighty volume provides the reader a good baseline about the applications, the programming tools, the services and the infrastructure that is being put into place to serve current and future research and education communities. The book is a collection of articles written by those who are building the various pieces. Some chapters are more easily read than others -- just as some topics are more easily understood. Whether network engineer or manager, there is something here for everyone.

Boy, when was VA declared an imagination-free zone? Don't the critical reviewers of this book have any belief in the possibilities afforded by grid computing? To complain that the book is light on technical detail is somewhat churlish, because the purpose of the book was to examine the possibilities, not show you how to code one up in an afternoon at the comfort of your desk.

Can there be any doubt that computing grids will start to become popular in some form or another? I don't think there is any doubt at all. Perhaps when some of the research and early work is complete and the standards agreed, some enterprising author can write a book on grid computing with more technical meat on the bone, but until that work is done, what would be the point? I suspect there just isn't all that much technical detail to point at yet.

To my way of thinking, all the possible uses and configurations of computing grids are still to be discovered. If you are one of the explorers intent on doing something new and different with the technology, then this book is for you. If you are only content to follow the crowd, once all the technical details have been worked out and served up on a plate to you by somebody else, then by all means skip this book and wait for one with more technical gravitas.

The Grid's most important function is as a manifesto: It says, hey guys, we will plug-in a simple PC into the wall, and we will have at our fingertips any computational power we need. We can simulate a crash of cars with a very fine precision, in program which needs perhaps 4 weeks to run on supercomputers filling a large room.

Some pieces of software, named tools, will find the fastest way to any computer in the world, or to any number of computers in the world, which are available for you. You can hire them, you can lease them, you can pay as you use them. We may even buy a Grid Card, like a plastic calling card, and do all the computing, until the card is depleted.

All it will be possible. One day, every single supercomputer in the world will be part of the GRID making money for its owners.

How people will collaborate? Here there is a void to be filled by some giant future computer utilities corporations, not yet born. AOL can become one of them, and increase its size hundreds of times. The telecom giants of today can take leadership. Or some new start up, a name no one has heard yet, will become the code named GRID Inc.

But the GRID Inc. (my own fiction) is not a fiction in 'The Grid' edited by Ian Foster and Carl Kesselman: the reader sees the blueprint. In the next edition, we would love to add the chapter for the Global Resources Director (GRD) a software that optimizes resources in a distributed heterogeneous environment.

The people minds shift. Computer Power as electricity? Will we have at home a power-meter measured in a computing unit, something equivalent to Kilowatts? Why not?.

Ladies and Gentlemen, the GRID is coming. Ian Foster and Carl Kesselman are the first to announce it.

The rest is history.

The Grid technology provides an infrastructure to couple computers (PCs, workstations, clusters, & so on), scientific instruments, storage, databases, application kernels, and people across the globe/Internet and present them as an unified, integrated resource to the user. The Grid tools hide complexities associated with accessing these resources and offers efficient mechanism to access them transparently. Using resource brokers, you can access compute power and application services as easy as accessing electricity today with ease. Still not convinced, read this book!

Still there is lot left to explore and I think we are still looking for a "brand new" killer application for the Grid. That helps in computational Grids making impact on the world economy similar to that of electrical Grids. Master the principles presented in the book and develop applications that revolutionize the Grid as much as Web did for the Internet and the world!

This book is full of visions of the future and other hyperbole. The "grid" is discussed as if it: a) is already in existence or b) is completely planned and the subsystems are tested, and is just waiting implementation. Neither could be further from the truth. While making hopeful predictions about the future is a good way to guide your thoughts and help in planning, it is sill just a guess! In particular, the notion of a "computational grid", analogous to the power grid, is quite a reach. This concept is not a technical one so much as a social/business one. I just don't buy it.

The technical content of the book is limited. Few new concepts are presented, are scattered and poorly explained. For instance, I don't need this book to learn about 3-tier architecture. The best feature appears to be the surprisingly complete bibliography of reference papers.

Networks will continue to expand, in number and critically, in capacity. Computers, and the services that run on them, will take advantage of this added connectivity in ways that are not yet clear. But, this is far from the ubiquitous utility model where anyone can plug in anywhere receiving any service for a modest fee.

Lastly, the whole pretext of a completely planned grid that is proposed is shot down by the author's own arguments that services and utilities have grown mainly due to unanticipated forces. i.e. the whole example of the city of Chicago. No one planned a vast city, but it grew nonetheless. In this sense, I'm sure that if something like the grid will come to fruition, it will happen without the massive planning and government spending that is implied in the book.

My title says it all. This is one of those "hand wave across the map" kind of books, and to call it a textbook is doing it quite a favor. If that's what you're into - a lot of dreams with no technical info to back it up, this book - and the most recent Starfleet manual - are for you. If you genuinely want to learn about real subjects in the area of high performance computing and networking, tho, skip this thing with prejudice.

This book is full of visions of the future and other hyperbole. The "grid" is discussed as if it: a) is already in existence or b) is completely planned and the subsystems are tested, and is just waiting implementation. Neither could be further from the truth. While making hopeful predictions about the future is a good way to guide your thoughts and help in planning, it is sill just a guess! In particular, the notion of a "computational grid", analogous to the power grid, is quite a reach. This concept is not a technical one so much as a social/business one. I just don't buy it.

The technical content of the book is limited. Few new concepts are presented, are scattered and poorly explained. For instance, I don't need this book to learn about 3-tier architecture. The best feature appears to be the surprisingly complete bibliography of reference papers.

Networks will continue to expand, in number and critically, in capacity. Computers, and the services that run on them, will take advantage of this added connectivity in ways that are not yet clear. But, this is far from the ubiquitous utility model where anyone can plug in anywhere receiving any service for a modest fee.

Lastly, the whole pretext of a completely planned grid that is proposed is shot down by the author's own arguments that services and utilities have grown mainly due to unanticipated forces. i.e. the whole example of the city of Chicago. No one planned a vast city, but it grew nonetheless. In this sense, I'm sure that if something like the grid will come to fruition, it will happen without the massive planning and government spending that is implied in the book.