Kazuhiko Kakehi

Formal boundary spanning by industry liaison offices and the changing pattern of university–industry cooperative research: the case of the University of Tokyo

Although university–industry (U–I) collaborations promise numerous benefits, both parties need to overcome high cultural and organisational barriers in order to materialise potentials. A variety of boundary spanning structures have developed in the USA and other countries, and this paper examines the roles and effects of industry liaison offices (ILOs) in Japan based on the case study of the Division of University Corporate Relations (DUCR) at the University of Tokyo. In Japan, because of governmental regulations, traditional U–I cooperative research has been formed within faculty members’ interpersonal networks with corporate researchers. Thus, the cooperative research projects were small in size and limited to bring actual commercial benefits. This paper suggests that formal boundary spanning by ILOs could facilitate the formation of ‘inter-organisational’ alliances between university and industry, and the alliances could generate larger, interdisciplinary, explorative cooperative research, overcoming the limitations of traditional cooperative research projects initiated in informal and interpersonal approaches.

A fusion-embedded skeleton library

This paper addresses a new framework for designing and implementing skeleton libraries, in which each skeleton should not only be efficiently implemented as is usually done, but also be equipped with a structured interface to combine it efficiently with other skeletons. We illustrate our idea with a new skeleton library for parallel programming in C++. It is simple and efficient to use just like other C++ libraries. A distinctive feature of the library is its modularity: Our optimization framework treats newly defined skeletons equally to existing ones if the interface is given. Our current experiments are encouraging, indicating that this approach is promising both theoretically and in practice.

SYSTEMATIC DERIVATION OF TREE CONTRACTION ALGORITHMS

While tree contraction algorithms play an important role in efficient tree computation in parallel, it is difficult to develop such algorithms due to the strict conditions imposed on contracting op...

Efficient Parallel Tree Reductions on Distributed Memory Environments

A new approach for fast parallel reductions on trees over distributed memory environments is proposed. By employing serialized trees as the data representation, our algorithm has a communication-efficient BSP implementation regardless of the shapes of inputs. The prototype implementation supports its real efficacy.

Generators-of-generators library with optimization capabilities in fortress

A large number of studies have been conducted on parallel skeletons and optimization theorems over skeleton programs to resolve difficulties with parallel programming. However, two nontrivial tasks still remain unresolved when we need nested data structures: The first is composing skeletons to generate and consume them; and the second is applying optimization theorems to obtain efficient parallel programs. In this paper, we propose a novel library called Generators of Generators (GoG) library. It provides a set of primitives, GoGs, to produce nested data structures. A program developed with these GoGs is automatically optimized by the optimization mechanism in the library, so that its asymptotic complexity can be improved. We demonstrate its implementation on the Fortress language and report some experimental results.

EMMA: a web-based report system for programming course--automated verification and enhanced feedback

EMMA is a web-based system supporting programming language education. It provides an integrated environment enabling students to upload programming assignments and teachers to evaluate them. EMMA automatically executes all programs submitted by students and provides different sets of inputs for them to test their programs. This has the benefit of letting teachers concentrate only on further hands-on verification. It also enables students to study individually and improve the quality of their programs before submission. The system also offers various communication methods between students and teachers. Students can write to teachers with problems and questions and receive both suggestions and evaluations. Also, notable results can be made available to all users, so that students can see their classmates’ results. Among previous web-based tools, many approaches have been applied to solve various problems: some tools offer learning environments adapting to students’ levels [2], while others facilitate building a local environment suitable for assignments [1]. Above all, we developed a system that automates some part of program verification without losing the flexibility of the submission format as programs.

An efficient staging algorithm for binding-time analysis

Binding-Time Analysis (BTA) is one of the compile-time program analyses which is a general framework for program optimization and program generation [1]. The task of BTA is to divide a source program into two parts according to a given binding-time specification [2]. A binding-time specification gives information about availability of data: static data are available at compile-time while dynamic data are available at run time. BTA determines, from a binding-time specification of the input, on which parts computation can take place by propagating information on static data. Partial evaluators specialize the program for the static data by using the information from BTA to generate a more efficient program than the original.