Hirofumi Tezuka TRC Massively Parallel Software Laboratory

As shown in the above profile, I didn't finish my university course, have changed my employment a good many times, am past the first flush of youth, and have no wife (not shown in the profile). In short, I've had so far quite an irregular life.

I joined the Real World Computing Partnership in April 1995. I am one of the few staff members of the RWCP and am destined to lose my job again when this partnership is dissolved in the future. If so, I will probably look for a new job, and my resume will doubtless spread onto yet another sheet of paper. I have been involved in the development of a number of microcomputer systems and UNIX workstations. As a researcher at the Japan Advanced Institute of Science and Technology, Hokuriku (JAIST), I was engaged in the study of multimedia and real-time processing. To conclude from my own experience in these fields, we must not consider the computer as just a combination of hardware and software. We employ the two concepts just as a convenience.

A decade ago, for example, people tried to reduce loads to software such as compilers by enhancing the performance of processor instructions (e.g. CISC). The advent of RISC, which simplifies and speeds up instructions, however, drastically changed the design of processors, and compilers now have more important roles. So, what is required now is well-organized collaboration between hardware engineers who have an extensive expertise of software and software engineers who possess in-depth knowledge of hardware, to develop a well-designed computer system.

I believe I am this type of software engineer with a store of knowledge about hardware.

As of December 1995, I am in charge of the development of communications software for a workstation cluster which consists of dozens of UNIX workstations for parallel processing. Why I say ﾅgas of December 1995ﾅh is that system engineers have to handle anything, as necessity arises, like an all-rounder, and I have no idea what I will be doing several months from now.$(BEq(B

The essential point of communication is speed. So, I often stay in the office until late at night in an effort to develop a faster program. I am probably one of those researchers who puzzle fellow clerks working in offices on the 13th floor of our building and causes them to sayﾅgwhat on earth are those guys doing on the 16th floor?ﾅhSince I have been chiefly involved in the development of commercial product, I am not a good at writing research papers of doing research itself. However, I will make good use of my experiences on UNIX systems and real-time processing, my strongest fields.