-
Lyman, Peter and Hal R. Varian, "How Much Information," 2000. Retrieved from http://www.sims.berkeley.edu/how-much- info on August 18, 2002. -
In addition to its use for promoting SANs, the data explosion myth also resonated with many IT professionals because it provided a convenient explanation both for increasing server downtime and also for growing budgetary expenditures on storage technology. Ultimately, it provided a catch-all justification for virtually any problem confronting storage and server operations, even those that had no relationship whatsoever with storage capacity and its utilization. -
In the mid-1990s, the analyst community had cultivated interest in a phenomenon it called the "thin client revolution." The Gartner Group, followed by other analysts, advanced the notion that the "fat client" Windows PC had become an unsupportable drain on corporate resources. Analysts promoted the concept of a dumb terminal device operating a Web browser and a Java Virtual Machine as an inevitable replacement for expensive-to-maintain, fat-client WINTEL desktops. This was among the first of a consecutive series of events in which the analysts, who were originally thought to watch and report on trends, instead tried to create trends for subsequent monitoring and analysis. After about two years of banging the thin client gong, the thin client revolution had fizzled, and even Gartner had to concede that it had been "something of a red herring." (They did so however in a paragraph written in a very small typeface and buried in the back pages of one of their less important subscriber newsletters: Their retraction was hardly made with the same measure of ballyhoo with which they had started the supposed revolution.) -
Reported in Jon William Toigo, "Storage Area Networks Still on Washington Wish List," Washington Technology , September 11, 2000, http://www.washingtontechnology.com . -
Jon William Toigo, "Avoiding a Data Crunch," Scientific American , March 2000, www.sciam.com . -
Trend data in these ratios was reported in numerous analyst studies in 2000. More recent assessments by industry analysts are skewed only to reflect the current slowdown in IT spending by organizations. -
See quote from Randy Kerns, a senior analyst with The Evaluator Group in Jon William Toigo, "Data Center and Storage Consolidation," A Report for Byte & Switch, July 17, 2002, www.byteandswtich.com . Since Kern's estimate was published there has been little trustworthy data on the size of the SAN market or how many SANs had actually been deployed. Most analysts count numbers of FC ports (on switches) shipped, or numbers of FC host bus adapters shipped, to support their estimates of SAN adoption rates. These numbers are inherently flawed as evidence of widespread SAN adoption. In fact, many companies that deployed early FC SANs have since "upgraded" them once or twice with newer , more capable equipment. Many FC ports are sold "in advance of requirements" because of problems and inefficiencies inherent in deploying SANs as a cascade of smaller switches. Moreover, FC HBA sales do not necessarily reflect SAN attachment. FC provides a high-speed point to point interconnect that can be used very effectively in building server-attached storage platforms as well as "storage area networks." -
As this book goes to press, leading analysts have just declared that SANs have replaced SAS as the leading storage topology being acquired by companies today. Bases for this finding are unclear. Left unanswered are questions such as what is being measured and counted as a SAN. If analysts are counting numbers of host bus adapter (FC) ports, their conclusions may be flawed by the fact that Fibre Channel is increasingly used to connect direct-attached arrays (an application to which it is well-suited). |
