Four years ago saw the release of the first book describing in detail how to build a storage network. Now that experienced author, Tom Clark, tackles IP storage networks in a new book.

by Alex Goldman ISP-Planet Associate Editor [September 12, 2002]

Four years ago, Tom Clark wrote Designing Storage Area Networks: A Practical Reference for Implementing Fibre Channel SANs, one of the first books to describe in detail the implementation of a storage area network (SAN). Printed in 1998, the book became so popular that it was translated into Japanese in 2000. Now Clark has written a book on the future of storage technologies, IP-based storage networks.

Clark is an industry insider. As director of technical marketing for Nishan Systems, he talks to the company's largest customers, learns their concerns, and then consults with the company's engineering team. The company he works for is an IP storage innovator. It has released its own storage networking protocols and makes products that are multi-protocol compatible.

Clark's new book, IP SANs: A Guide to iSCSI, iFCP, and FCIP Protocols for Storage Area Networks, is meant to provide a complete overview of this nascent but emerging field. Clark told us, "I gave it enough technical meat to satisfy technical readers and gave them guideposts so they could read more if they wish to, but I also did not make it too intimidating for a storage manager, consultant, or VAR. It's a very complex subject, and I wanted to achieve a level of abstraction above the specifications themselves. I also wanted to talk about the applications of the protocols and not get caught up in describing the performance of the protocols themselves, blow by blow."

Clark starts with an introduction to the concept of storage and follows, in two through seven, by describing storage protocols and technologies that have been in use for several years, including RAID, SCSI, NAS, Fibre Channel, and Gigabit Ethernet. He then describes standard networking protocols, such as IP, UDP, and TCP. Since Nishan Systems is not tied to any of these legacy protocols, Tom Clark is able to critique them with an insider's accuracy.

Chapters 8 and 9 are the "technical meat" of the book, in which Clark describes the new storage networking protocols mentioned in the book's title as well as several additional protocols.

Chapter 10 covers the thorny issue of security in storage networks. Currently, a large number of storage networking projects are being undertaken to increase security by providing backup or mirroring, and it is critical, especially in these applications, to secure corporate data that is transmitted over the Internet.

Chapter 11 covers Quality of Service (QoS) issues which have to be solved in order to deliver many valuable applications. A thorough understanding of protocol and networking issues is an important prerequisite to understanding this chapter.

Chapter 12 describes the InifiniBand initiative, a proprietary initiative that could be already out of date and is run by the Infiniband Trade Association.

Chapter 13 describes IP storage applications, and this is where managers and executives will derive the greatest value from the book. However, they too will have to understand the protocols and technologies—and their various weaknesses—that underlie the applications being delivered in order to understand what IP storage applications are viable and why.

Chapter 14 is the book's conclusion. The last sentence is particularly telling: "As storage and networking technologies become more tightly integrated, storage networking ceases to be a distinct entity and simply enters mainstream data communications. From a user perspective, this is ideal." Left unsaid is the fact that many equipment makers may find that this innovation is not serving their interests.

The book retails at $42.99, for which readers get 288 pages of concise, dense, informative writing as well as handy chapter summaries, a bibliography, glossary, and index.

ISPs in particular should find themselves champions of IP storage networks. In implementing such networks for themselves, ISPs will find they already have most of the necessary networking and storage expertise, and also understand the potential problems that any application may face when it crosses the Internet. If IP storage networks become popular, ISPs can look forward to greater demand for their products, both bandwidth and also higher level consulting services.

—End