Middleware 2006 - Keynotes
John Wilkes,
Hewlett Packard Laboratories,
Title: Utilification Redux
Abstract: In 2004,
I wrote: "Utility computing has the potential to revolutionize the way we
purchase, organize, and distribute computational power and services. It will do
so by offloading resource provisioning to centralized sites that can benefit
from economies of scale, careful, failure-resilient construction, flexibility
and changeability of hardware choices, and scalable and business-driven
management techniques. But that promise is useless unless we can move
applications from traditional computing environments into utility ones, where
the applications are fronted by service interfaces and resource flexing is the
norm."
The landscape has shifted somewhat since then: now, few doubt that
service-oriented architectures (SOAs) are here to
stay; the early emphasis on performance in utility computing is beginning to
seem misplaced; and the non-performance "ilities"
are going to need much more attention than they've been given so far.
In this talk, I'll look at how these ideas have changed how we should
think about utilification -- and ask not what they
can do for middleware, but what middleware can do for them.
Bio: John
joined HPL in 1982 with a PhD from
Sean Baker, IONA,
Title: Middleware's Teenage Years
Abstract: There are great middleware products, so why is the middleware space so immature? This is not an attack on any individual, company or standard, but a statement on how we behave as a group. We have no consistency in the terminology we use (I'm sure you use terms consistently, but does everyone else?). Users get conflicting advice on key issues, which means that they are a lot less confident in making technology decisions. Surprisingly, given the role that middleware sets itself, interoperability between middleware products is difficult. We start multi-year campaigns across the whole industry - such as loosely coupling systems - without agreeing clear goals. Complementary approaches are declared to be rivals. The list goes on and on.
This talk takes a critical look at our industry, and focuses on a number of areas in which we need to improve.
Bio: Dr. Baker is Chief Corporate Scientist, a member of
In the many years since