history

	Who We Are
	What we want	How can we help you?	What we can do for you today	What's up

Elepar began life in Sunnyvale, CA, in 1998 and has been in its current location in Beaverton, OR since early 1999.

President: David DiNucci (see below)
Advisory board:

Pat Cauduro: Based primarily in Boston, Pat's experience includes founding two startups (Cluster Appliances and Windward Networks), Director of Business Development and/or Software at those plus C-Port and Shiva, and other prior techincal roles at Ektron Applied Imaging (Kodak), Advanced NMR, Siemens Medical Electronics Labs, and HP Medical.

Robert Dietrich: In addition to cofounding two startups (Advanced Video Communications and Process Path), Bob has been involved in both technical and management positions in large established companies (e.g. Intel, nCUBE, Tektronix) and smaller/newer ones (Centerspan, Complete Business Solutions).

Robert "Buzz" Hill: Buzz is an experienced entrepreneur in the Portland area, founding EyeDentify, and more recently, Kumulipo and Minds' I.

Miki Tokola: Miki was VP of Marketing at Genedax (which he co-founded) and Jenkon, and has marketing management and/or technical experience from Mentor Graphics, Floating Point, Tektronix, and Boeing Computer Services.

Contact info:
14380 NW Hunters Dr.
Beaverton, OR 97006
503-439-9431
moreinfo@elepar.com

The History of Elepar Technology

Although Elepar's technology has many relatives, the direct lineage for the last few decades effectively began with a graphical notation called Large Grain Data Flow (LGDF), invented by Robert G. Babb II (though obviously based on several predecessors itself). Dr. Babb's idea was to extend traditional "dataflow diagrams" used in software engineering into an automated methodology to simplify the development of programs for uniprocessors and shared-memory multiprocessors. David DiNucci became Dr. Babb's graduate student at Oregon Graduate Institute (then Oregon Graduate Center) in 1985, and in that role spent the next few years extending LGDF to work more portably on distributed memory computers as well, calling the resulting more portable (but somewhat less powerful) method LGDF2. DiNucci then formalized the methodology and the notation into a computational model called F-Nets, which he presented as his PhD dissertation (at OGI) in 1991. Upon leaving OGI, Dr. DiNucci used his spare time to design a full-fledged graphical computer language, based upon the F-Nets computational model but having a precise human-oriented syntax and containing constructs to facilitate object-oriented programming, arrays, data parallelism, and templates. This language is now called Software Cabling, and is the central focus of Elepar's technology.

Dr. David C. DiNucci, President and technology officer

Dr. DiNucci is the inventor and initial developer of the Software Cabling, Cooperative Data Sharing (CDS), and PICA technologies which provide the initial unfair advantage for Elepar. His PhD dissertation at Oregon Graduate Institute (OGI) was entitled "A Formal Model for Architecture-Independent Parallel Software Engineering", and described the F-Nets computational model which serves as a core to SC and CDS. Upon leaving OGI in 1991, he spent a small time in a post-doctoral position in the Languages and Tools Group at the National Energy Research Supercomputing Center (NERSC, then at Lawrence Livermore National Laboratories) and then began working as a contractor at the Numerical Aerodynamic Simulation (NAS) facility at NASA Ames Research Center in late 1991. Here he was a member of the Parallel Tools Group, developing the key principles of the CDS package, and serving temporarily as a member of the international Message-Passing Initiative 2 (MPI-2) Forum (the One-Sided Communication subcommittee). From this same position, he later played a key role, as lead of the IPG Distributed Architectures and Scheduling team, in planning the NASA Information Power Grid, a joint multi-year national research project with the National Science Foundation to investigate the use of computational grid technologies for NASA's future needs. This work allowed him to work with leading researchers in the computational grid field from around the country, and earned him (and others) a NASA Ames Group Achievement Award. He left his position at NASA in August 1998 to form Elepar. In addition to the publications listed on Elepar's technical references page, he also contributed two chapters (concerning the Alliant FX/8 and the Loral Dataflow 100) to the book "Programming Parallel Processors", ed. R. Babb, 1988, Addison-Wesley.

For a PDF version of Dr. DiNucci's formal resume, click here.

Dr. DiNucci can be reached at dave@elepar.com