|
|
|
Research in Dependable Computing
Behrooz Parhami: 2007/06/19 || E-mail: parhami@ece.ucsb.edu || Problems: webadmin@ece.ucsb.edu Other contact info at: Bottom of this page || Go up to: B. Parhami's research program or his home page
On June 19, 2007, Professor Parhami's UCSB ECE website moved to a new location. For an up-to-date version of this page, visit it at the new address: http://www.ece.ucsb.edu/~parhami/res_dep_comp.htm In
the following descriptions, selected items from B.
Parhami’s list of publications are provided in brackets. Defining
the Field
Dependability
concerns are integral parts of engineering design. Aspects and/or measures of
dependability for computer systems include reliability, availability,
performability (sustained computational capability), safety, security,
testability, and maintainability. Ideally, we would like our computer systems to
be perfect, always yielding timely and correct (or at least safe) results.
However, just as bridges collapse and airplanes crash occasionally, so too
computer hardware and software cannot be made totally immune to undesirable or
unpredictable behavior. Despite great strides in component reliability and
programming methodology, the exponentially increasing complexity of integrated
circuits and software products makes the design of perfect computer systems
nearly impossible. The field of dependable computing [96] is concerned with
studying the causes of imperfection in computer system (impairments to
dependability), tolerance methods for ensuring correct, safe, or timely
operation despite such impairments, and tools for evaluating the quality of
proposed or implemented solutions. Areas
of Work Methods
for ensuring the uninterrupted and reliable functioning of digital computers in
the presence of component faults and subsystem malfunctions are increasingly in
demand as such systems expand in hardware and software complexity, perform a
wider spectrum of critical functions, interact with a growing number of
minimally trained users, and are expected to endure harsh or hostile operating
environments. Thus, redundancy techniques, that were once limited to
high-profile space and military projects, are becoming commonplace in
run-of-the-mill computer systems. Unlike much work in the area of dependable
computing, Professor Parhami’s focus is not solely on hardware reliability or
software certification. Rather, he views result correctness as the central
theme. His work thus combines elements of hardware redundancy, architectural
flexibility, and algorithm robustness to achieve result correctness and graceful
performance degradation. Because system reconfiguration and elimination of
single points of failure invariably involve the use of multiple processors,
Professor Parhami’s projects in dependable computing are intimately related to his
research on parallel processing.
Current Threads
Return to: Top of this page || Go up to: B. Parhami's research program or his home page
|
|
