This review and interview with the authors was originally run in the December 12, 2005 issue of GRID today. To view the original article click here.

We keep hearing about Grid computing evolving from big science to enterprise, but there are precious few new ideas about the specific opportunities and business cases behind the evolution. If you're tired of the insipid "aligning IT with business goals" market-ese on Grid and want some new perspectives, you might want to check out a book that was recently released: Grid Computing The Savvy Manager's Guide.

The book is co-authored by Pawel Plaszczak and Richard Wellner -- both insiders in the open source Globus Toolkit development project that began 10 years ago and is the nucleus of the Grid computing evolution. These aren't marketing guys, they have street cred and hands on experience with grids -- and the book reads that way.

I'll jump right out and tell you what I like most about the book. It is first and foremost a good read. It flows very well, giving the reader a good learning "bang" for the time-investment "buck."

Initial chapters provide the reader with a good foundation for understanding the fundamentals of Grid. Although this information will probably be a repeat for those that have any Grid background at all, it is presented concisely and will serve as a good reminder and handy reference for those in this category. Particularly useful to enterprise end-users are the explanations of the importance of OGSA (Open Grid Services Architecture) and WSRF (Web Services Resource Framework) to the enterprise Grid evolution. The set-up also gives a good amount of discussion to virtualization's tie-in to Grid, and some clarity in distinguishing Grid from other technologies/concepts (clusters, cycle scavenging, distributed computing, Web services, peer-to-peer, etc.).

The book goes on to examine the entry of Grid computing into business. Again, there might be some repetitive information here, but the authors take the extra step and extrapolate on that information. They use a well-defined taxonomy study to predict what, exactly, enterprise Grid will look like when it becomes more mainstream. From departmental grids to enterprise and partner grids, the book explains how different affinity groups and industries will participate.

The second half of the book is where the real value is. Chapters on what it means to actually be on the Grid, technical issues and management issues are the "value adds" to the Grid books that have come before it.

In the case studies and examples, the book is a little heavy on the side of academic and research grids. But given the relative lack of real world enterprise grid deployments, one can hardly blame the authors for this.

To add some first-person insight to my review, I sat down with the authors and asked them a bit more about their motivations for and their thoughts after publishing this book:

NAWROCKI: Many have identified the fact that this is a critical time in Grid's cross-over from academia/science to enterprise. In The Savvy Manager's Guide, you cover this theme throughout -- but do you have three or four specific challenges that you would say crystallize this thought?

PLASZCZAK: The Savvy Manager's Guide is a book for those who want to understand the business value of Grid technologies. It took us two years to complete the book and during this time we engaged in about 60 interviews with Grid visionaries and vendors, as well as the early adopters. As we both came from inside the Grid computing industry, this allowed us to see the subject from a different angle. Needless to say, this also changed our own perspective on the challenges for Grid computing. There is no question that the Grid paradigm brings incredible value, both to the business and to society. As we discuss in Chapter 2, the core concept of grids is the virtualization of resources, which brings consequences and opportunities reaching further than many of us can imagine. As for challenges, some of them we took on ourselves when writing the book.

The first challenge to overcome is the common misunderstanding on what Grid computing is. In the book, we explain why different groups of interest use different definitions to describe the same phenomenon. We also explain why, in fact, these definitions are not contradictory but complementary.

Second is fear and uncertainty on possible Grid business models and benefits. Understanding the concept of Grid computing is a prerequisite to talk about its business value. The next step is to understand business opportunities in the Grid computing markets and calculate the benefits for the participants. In the book, we present various business models, or roles in the Grid market. These models are functioning today. Grid is about tangible and very useful things, often possible with today's technology. A very recent example: Sun Microsystems is using their utility Grid for converting document formats. This is happening today. We should get used to talking about grids in present tense, and instead of inventing futuristic business models, we must admit that down-to-earth, tangible grid installations exist today and are bringing revenue.

Another challenge is related to set of psychological/political/legal/administrative factors, which today give preferences to organizations working in isolation to those cooperating externally at many layers. I believe that globalization, the "net effect," and Grid computing will change this thinking.

WELLNER: Grid technology in the form of Globus Toolkit as an infrastructure and various products from other vendors as point solutions (e.g. SGE/PBS/LSF as schedulers in a larger grid framework) continues to have tremendous emerging uptake in the commercial space. We are currently working with customers in automotive, pharmaceutical, finance and defense vertical markets to name a few of the most interesting projects.

Over the past couple years, it has been great fun to see the market mature from being one of overt skepticism to one that is becoming cautiously optimistic about the real power of the Grid at large and various forms of internal grids, as well. We hope that our book is playing some small part in that migration as it is clear that the Grid has satisfied some serious needs and has the potential to be a really big thing.

This acceptance of Grid technology as a meaningful infrastructure has knocked down one of the most significant barriers that existed in the past. In our book, we spend a brief time talking about this, but most of it assumes that this barrier has indeed been removed and talks about how to go about doing grid systems. The significant challenges that remain can be roughly segmented into cultural and technological concerns.

Culturally, there are still hurdles to leap in terms of creating a comfort level with user communities. They are often concerned about what sorts of brokers will be used to manage compute time between facilities, what kind of schedulers will be used on local clusters, what kind of security will be used and how all this maps to corporate policies that were put into place without consideration for the emergence of the Grid.

These are the kinds of issues that are a natural fit for working with a partner that has done these sorts of things before. We talk about several use cases in the book where these kinds of challenges are met, but it is clear that these kinds of cultural hurdles will be around for a long time.

The second group of challenges surrounds the technology itself. In particular, finding the right combination of tools. For instance, the Globus Toolkit as an infrastructure layer with different pieces on the front and back depending on the needs of various organizations, or perhaps using a proprietary platform from start to finish because vendor lock-in isn't a concern.

As above, often times it is very helpful to work with a partner that has done these kinds of things before. In both the use cases in the book and with customers, we have found that being able to lean on folks who have been through this process before is very helpful. Our book makes no attempt, due to the dynamic nature of the field, to be a comprehensive survey of tool options or how to integrate them into a Grid solution.

In short, the field has matured to the point where the Grid is more about system integration efforts than research and development.

NAWROCKI: What does Grid Computing: The Savvy Manager's Guide accomplish that its predecessor books on Grid have not yet touched on?

WELLNER: Our book is the first purpose-written volume designed to teach managers what they need to know to begin their first grid deployments. It is, in fact, one of the first purpose-written books of any sort on Grid computing. The publishing world, until now, has been dominated by Grid computing books that are essentially a collection of papers. These kinds of books are extremely useful, but they don't touch on the kind of cover-to-cover coherent discussion of a topic the way a book like ours can.

PLASZCZAK: Grid Computing: The Savvy Manager's Guide was designed to be concise and comprehensive, yet not technical. It is a guide, designed to be read in order. As such, it is complementary to several other books, papers and articles on grids that take on the subset of the subject and explain it in low-level details. Our intention was to introduce a business decision-maker to the area of Grid computing. To my knowledge, noone so far has attempted to write a book on Grid computing in business terms.

NAWROCKI: Care to wager any predictions on enterprise Grid, either in terms of what we're likely to see in the next few years, or specific areas in enterprise (e.g., storage, middleware, enterprise developer activity, etc.) where we'll start to see the biggest surges in activity (either from open source Globus community, with vendors or otherwise)?

WELLNER: There is interesting work being done in just about all areas of Grid computing right now. In some areas, things are being refined to the point of being just plain cool. In others, there are broad new areas of research underway that wouldn't have been possible without the kind of stable underpinning that exists today after nearly a decade of research into core areas.

This is a very exciting time to be involved in the Grid and, indeed, computing at large. We are on the verge of a new way of solving problems that has tremendous potential.

PLASZCZAK: Grid computing is part of a worldwide and industry-wide trend, leading to service orientation, on-demand business, virtualization and ubiquity. Already today we are witnessing appearance of Grid resource providers. In the near future, we may see the growth of an on-demand business sector, with a variety of service offerings, available for both organizations and individuals. Many of these users will benefit from it without knowing they are using Grid, as it is already happening today. For businesses, this means an entirely new market sector full of opportunities.

In general, what sets the book apart from other available literature on grids is the practical context, backed up with actual case studies that it gives to would-be enterprise Grid decision-makers. From "building the business case" to understanding the motivations/requirements for "Grid-enabling" a product or application, this is a comprehensive look at some of the key business issues related to enterprise Grid.

close window