Search

In 1982, Japan launched its Fifth Generation Computer Systems project (FGCS), designed to develop intelligent software that would run on novel computer hardware. As the first national, large-scale artificial intelligence (AI) research and development (R&D) project to be free from military influence and corporate profit motives, the FGCS was open, international, and oriented around public goods. Although the FGCS did not plan any commercialized technologies, many American computer experts portrayed it as an economic threat to U.S. dominance in computing and the global economy—and policymakers around the developed world believed them and funded AI projects of their own. Later, however, the FGCS was remembered as a failure. Why? This article recasts the FGCS as an interstice in the shift from a state-funded regime of American science organization to the neoliberal privatized regime of R&D now ascendant around the world. By exploring how notions of economic competitiveness and national security shaped R&D, this article reveals AI to be a product of contingent choices by multiple actors—nation-states, government bureaucracies, corporations, and individuals—rather than the outcome of deterministic technological forces.

The technical goal of the Fifth Generation Computer Systems (FGCS) project (1982–1993) was to develop Parallel Inference technologies, namely systematized technologies for realizing knowledge information processing on top of parallel computer architecture. The Logic Programming paradigm was adopted as the central working hypothesis of the project. At the same time, building a large-scale Parallel Inference Machine (PIM) meant to develop a novel form of general-purpose computing technologies that are powerful enough to express various parallel algorithms and to describe a full operating system of PIM. Accordingly, the research goal of the Kernel Language was set to designing a concurrent and parallel programming language under the working hypothesis of Logic Programming. The aim of this article is to describe the design process of the Kernel Language (KL1) in the context of related programming models in the 1980s, the essence of Concurrent Logic Programming and Constraint-Based Concurrency, and how the technologies we developed in those days evolved after their conception.

Data mining is known as data, which promotes the growth of knowledge discovery. It is the process of analyzing descriptive data from divergent perspectives and summarizing it into valuable information, which is high-level music processing out of which a machine intends to decipher the Raaga of a frequency or the pitch of the music. One of the ways to approach the task is by comparing selected music features from the spectrum and a Raaga database. Recognizing emotion from music has become one of the active research themes in image processing and applications based on human-computer interaction. This research conducts an experimental study on recognizing facial emotions. The flow of the emotion recognition system includes the basic process in the singular value decomposition system. These include music acquisition, pre-processing of a spectrum, feature detection, feature extraction, classification, and when the emotions are classified, the system assigns the particular user music according to his emotion. The proposed system focuses on live images taken from the music database. This research aims to develop an automatic music recognition system for innovative manufacturing through the additive manufacturing route. The emotions considered for the experiments include happiness, Sadness, Surprise, Fear, Disgust, and Anger that are universally accepted. This paper overviews the progress of applying Additive manufacturing in Applied Machine learning which sustains the capability of disruptive digital manufacturing.

Artificial Intelligence (AI) is one of the current emerging technologies. In the history of computing AI has been in the similar role earlier - almost every decade since the 1950s, when the programming language Lisp was invented and used to implement self-modifying applications. The second time that AI was described as one of the frontier technologies was in the 1970s, when Expert Systems (ES) were developed. A decade later AI was again at the forefront when the Japanese government initiated its research and development effort to develop an AI-based computer architecture called the Fifth Generation Computer System (FGCS). Currently in the 2010s, AI is again on the frontier in the form of (self-)learning systems manifesting in robot applications, smart hubs, intelligent data analytics, etc. What is the reason for the cyclic reincarnation of AI? This paper gives a brief description of the history of AI and also answers the question above. The current AI “cycle” has the capability to change the world in many ways. In the context of the CE conference, it is important to understand the changes it will cause in education, the skills expected in different professions, and in society at large. acceptedVersion

In the last two decades, researchers have found many advantages related to Artificial Intelligence (AI) in the performance of both service and manufacturing systems. It includes alternate methods to conventional techniques and there are mechanisms of integrated systems. Currently, many researches related to artificial intelligence are under process. At present, artificial intelligence has been used to solve various problems in several fields and has become more popular around the world. This field is extremely difficult to review either chronologically or thematically. AI has a long history of growing constantly and changing tremendously. Connected with the history of computers, artificial intelligence is the main concept behind the fifth generation computers. This review paper deals with the framework of artificial intelligence and focuses on fields related to artificial intelligence. Particularly this paper describes about the recent growth in the field of artificial intelligence and its applications. Several issues of artificial intelligence are analysed. It is concluded that still many researches are yet to be conducted for further growth in the field of Artificial Intelligence (AI).

Since the first formal specifications of modern computing machinery as laid out by Alan Turing and his contemporary fellows, we have been witnessing, during the last three decades, an evolutionary path in computing towards more personalized and contextualized data and knowledge artifacts. Information sharing, co-ordination, co-operation and, to some extent, collaboration among machines has been envisioned for complex problem solving. A prominent example of this problem solving approach has been the Fifth Generation Computer Systems (FGCS) project as launched in Japan in the 1980s and based on the concept of calculation using massive parallelism in logic and hardware. Grid and Distributed Computing, also known as Future Generation Computer Systems, is another similar attempt to exploit massive parallelism in order to solve complex problems. In this article, we explore the notion of Collective Intelligence (CI) in the realm of the Social Web and its potential to become a new computing paradigm for creating solutions or strategies to tackle wicked problems where the synergistic interactions of a group of people with diverse cultural and professional backgrounds are requested.

Fifth Generation computers should not simply be regarded as an enhancement of current computer technology: the intention is that a fresh approach should be taken to computer science and to the use of computers. The argument of this paper is that the fresh approach must encompass education and training, with implications that extend far beyond the simple use of computers.

The technical goal of the Fifth Generation Computer Systems (FGCS) project (1982–1993) was to develop Parallel Inference technologies, namely systematized technologies for realizing knowledge information processing on top of parallel computer architecture [8].

Traditionally, topics such as parallel computing, computer graphics, and artificial intelligence have been taught as stand-alone courses in the computing curriculum. Often these are elective courses, limiting the material to the subset of students choosing to take the course. Recently there has been movement to distribute topics across the curriculum in order to ensure that all graduates have been exposed to concepts such as parallel computing. This work describes one approach to threading topics from AI, graphics, and parallel processing into coursework starting early in a student's program without substantially altering the basic course objectives. This, in turn, allows the student to sample topics early enough to make informed decisions when selecting electives.

The rapid development of computer-based communication technologies, typified by the advent of the Internet, is said to be giving rise to a novel form of community, a global electronic community.The paper considers, primarily in reference to the Japanese context, whether such an electronic community can really claim universality, allowing individuals to communicate with each other across the existing boundaries of nation-states, corporations, and the family, and how it will affect the existing socio-political structure. One focus of discussion is a government-funded Japanese project, the 'Fifth Generation Computer System Project' (FGCSP), which was carried out over a decade beginning in 1982. A critical assessment of the project bears out that its failure should not be ascribed, as is often done, to its having concentrated on developing hardware technologies with limited applicability, but rather to a serious flaw in its basic design, namely, its misperception about 'universal language'. In the electronic community as it exists now, English is essentially accorded the status of the 'universal language', the one indispensable for access to the Internet. It is argued here, however, that such a state of things, if allowed to continue, is bound to hinder fruitful communications among various societies with diverse natural languages and social and cultural values peculiar to them, and to create, for the non-English world, a problem of'language discrimination' by driving a wedge between the elite few proficient in English and the non-English-speaking majority

This paper investigates the advantages and disadvantages of a publicly supported government-industry research consortium as a vehicle for conducting basic research, taking Japan's Fifth Generation Computer Systems Project as a case. After discussing the organization, activities, and achievements of the project, we inquire whether the usual explanations of cooperative R&D apply to the project. Complementarity of resources from different companies may not have been as important as in previous research associations in Japan (e.g. the VLSI Association) but the establishment of a temporary research institute with seconded researchers was an effective way to deal with the limited mobility of researchers in the Japanese labour market. Most importantly, the project fostered the spillover of created knowledge. There remains a doubt, however, whether the project can be a model for future research projects.

Limitations in power and speed, coupled with the unprecedented levels of integrations in today's semiconductor industry have resulted in massive hardware parallelism in modern computer chips leading to multicore/manycore general purpose processors and graphical processing units (GPUs) with hundreds of simple cores. As a result, all computing is becoming parallel (high performance) computing. On the other hand, future work force (our students) are not well prepared for this new era in processing technology, where parallel programming will be necessary and parallel processing will be the norm rather than just a niche area. In addition, our students are often prepared in an abstract manner with little or no connection to real life applications. This paper seeks to revolutionize our computer science curricula through a comprehensive reengineering of core computer science subjects in order to address the aforementioned pressing issues.

Logic programming has been considered a viable solution for distributed computing since the Fifth Generation Computer Systems project [8]. Nowadays, this line of thought is gaining new verve, pushed by the need for new programming paradigms for addressing new emerging issues in distributed computing. We argue that a missing piece in the current state-of-the-art is the capability to express statements about the knowledge state of distributed nodes. In fact, reasoning about the knowledge state of (group of) nodes has been demonstrated to be fundamental in order to design and analyze distributed protocols [7]. To reach this goal, we designed Knowlog: Datalog¬ augmented with a set of epistemic modal operators, allowing the programmer to directly express what a node "knows" instead of low level communication details.

Many AI researchers, particularly in the USA, were surprised when Japanese AI scientist Hiroaki Kitano was chosen to receive the 1993 'Computers and Thought' award for his work in AI and parallelism. Their astonishment derived from the pervasive perception in the USA that the Japanese Fifth Generation Project, seemingly the major Japanese effort in AI and parallelism, was a failure. "How," they asked, "was a Japanese scientist able to become a leader in this area?" This confusion about the state of parallel AI research in Japan is primarily due to lack of knowledge about Japanese efforts. >

One of the successful outcomes of the Fifth Generation Computer Systems Project is the development of Model Generation Theorem Provers (MGTPs). MGTPs have solved previously open problems in finite algebra, produced rapid proofs of condensed detachment problems, and are providing an inferential infrastructure for knowledge-processing research at ICOT. They successfully exploit the Fifth Generation Project's KLl logic programming language and parallel inference machines to achieve high performance and parallel speedup. This paper describes some of the key properties of MGTPs, reasons for their successes, and possible areas for future improvement.

Projecting a great vision of intelligent systems in the service of the economy and society, the Japanese government in 1982 launched the national Fifth Generation Computer Systems (FGCS) project. The project was carried out by a central research institute, ICOT, with personnel from its member-owners, the Japanese computer manufacturers (JCMs) and other electronics industry firms. The project was planned for ten years, but continues through year eleven and beyond. ICOT chose to focus its efforts on language issues and programming methods for logic programming, supported by special hardware. Sequential ‘inference machines’ (PSI) and parallel ‘inference machines’ (PIM) were built. Performances of the hardware-software hybrid was measured in the range planned (150 million logical inferences per second). An excellent system for logic programming on parallel machines was constructed (KL1). However, applications were done in demonstration form only (not deployed). The lack of a stream of applications that computer customers found effective, and the sole use of a language outside the mainstream, Prolog, led to disenchantment among the JCMs.

An overview will be given in this paper of the original plans of the Japanese Fifth Generation Computer Systems project and they will be compared with what has actually been achieved in the project's ten years of work. This is done by looking at the results demonstrated at the 1992 FGCS Conference, in the form of some 20 application programs all running on the PIM (Parallel Inference Machine) hardware and written in (systems written on top of) KL1. KL1 is the language developed for the FGCS project for the operating system controlling hundreds of processors, as well as for applications. It is derived from Prolog and could be called a Logic Programming language. Also the results of a team of international experts, asked by ICOT to evaluate the success of the FGCS project, will be given in a condensed form.

Japan's Ministry of International Trade and Industry's (MITI's) Superspeed project, which investigated high-speed devices and computer architecture, algorithms, and languages for parallel computing, is reviewed. The supercomputing industries in Japan and the United States are compared. The architecture and performance of current supercomputers and the current states of supercomputer technology and supercomputer software are discussed. >

Changes in the large computer market during 1992 are discussed. Competition for supercomputers from workstation networks is noted. New machines that debuted, despite the continuing mainframe decline, are described. Also examined are the push by massively parallel processors for larger market share, Cray's announced three-year collaboration with the US Department of Energy's Lawrence Livermore National Laboratory and the Los Alamos National Laboratory, the conclusion of Japan's Fifth Generation Computer Systems Project, popularly branded a spectacular failure, and Japan's first massively parallel processor. >

The Real World Computing (RWC) program (also called the New Information Processing Technology program), a 10-year program sponsored by Japan's Ministry of International Trade and Industry (MITI), is described. The goal of the program is to lay the technological foundations for the information society that Japan anticipates in the next century. The RWC program replaces the Fifth Generation project. The author presents a summary of the draft prepared by the RWC program's Feasibility Study Committee as presented in Tokyo, March 2-3, 1992, at what was called RWC92, the Second NIPT Workshop. MITI expects future information systems to be based on a flexible integration of massively parallel computing, optical computing, neural computing, and logic programming. The Real World Computing program aims to establish theoretical foundations for these technologies, explore applications, and study how they can be integrated. >

Research consortia in the United States have proliferated over the past decade. By the beginning of 1992, over 250 consortia had registered with the U.S. Government under the National Cooperative Research Act (NCRA) of 1984. The Microelectronics and Computer Technology Corporation (MCC) in Austin, Texas, was one of the first consortia to register under the Act. This article describes my first-hand experience in heading up the technology transfer function in the Computer Aided Design (CAD) Program--one of MCC's original four technology programs. While much has been written about MCC in the general and business press (1,2,3,4), and researchers from the University of Texas at Austin and the IC2 Institute have postulated several theories based on their extensive surveys and interviews with MCC staff and shareholders(5), this is one of the few papers describing MCC from an insider's perspective. In a recent article in Issues in Science and Technology (6), my associate Jack Bremer and I described five lessons that consortia must learn in the 1990s if they are to be successful. Those lessons are: 1. Maintain a coherent strategic vision. 2. Choose research problems that have significant payoffs. 3. Optimize communication with members. 4. Make technology transfer a continual process. 5. Strengthen membership commitment. While all five lessons are vital to consortium management and members, lessons 3 and 4 are particularly important if consortia are to realize the potential leverage that their supporters often cite. As I shall explain later, understanding these two challenges may lead to a new model of research consortia for the 1990s that overcomes the shortcomings of earlier collaborative efforts. PREDICTED VS REAL PROBLEMS I have been involved with U.S. research consortia (as a consultant, technology transfer manager, strategic planner, or journalist) for over a decade, and have visited with officials of Japanese consortia, including the former director of the VLSI Program and administrators of the Fifth Generation Computer Systems Project, on three occasions. During this period, I have discussed the merits and drawbacks of consortia with technologists and managers in many organizations who were already members of consortia or considering participation. Originally, many observers predicted that research consortia in the U.S. would fail for three reasons: 1. Competitors in domestic markets would not cooperate, even in the face of foreign competition supported by foreign-government-sponsored research consortia. 2. The concept of "pre-competitive research" was too ill-defined, and would not result in meaningful research results. 3. Consortia participants would not assign their best people to the venture, so the resulting research would not be first-rate. While there is some truth to each of the above predictions, numerous examples from personal experience contradict those predictions. By and large, the "competitor cooperation" syndrome is a non-issue. Only when a participant's particular product line or manufacturing process comes into play, do competitive stresses emerge, and several successful mechanisms have been used to diffuse those pressures. Pre-competitive research, often called generic research, has been well understood for many years, and does not limit the usefulness of research results so long as the research is at least taken to the prototype stage. Finally, while it has been true that companies usually and understandably assign non-critical researchers to a consortium, company "assignees" in MCC's case made up only a small percentage of the overall staff (less than 15 percent as of mid-1990). If the above-postulated problems with research consortia in the U.S. are red herrings, then what are the real problems? There are three: 1. The consortium, and especially its members, may treat "technology transfer" in too narrow a sense, leading to a situation in which the technology lies fallow after it becomes available to participants. …

The fifth generation computer project was launched in 1982, aiming at developing parallel computers dedicated for knowledge information processing. It was commonly believed that knowledge processing based on symbolic computation is very difficult to parallelize. We made a conjecture that logic programming technology will solve the difficulty. During the project, we succeeded in inventing a parallel logic programming language, GHC, and its realistic extention, KL1. We devoted our effort in developing parallel hardware called PIM, dedicated for KL1. We are now developing software technology on top of KL1 and PIM. We succeeded in developing PIM's operating system totally in KL1, which runs on Multi-PSI, a current experimental parallel hardware for developing KL1 programming. We are now concentrating our effort on developing mapping techniques, such as load balancing and priority control, for deriving real spead up in parallel execution. We are also trying to develop more realistic and complicated application programs such as VLSI-CAD systems, case based reasoning systems and parallel theorem provers which, we hope, will run efficiently on PIM.

Japanese supercomputers (e.g. Hitac S-3800, NEC SX-3R, Fujitsu VP2000) are described and compared with US machines. The Japanese approach has been to develop high-speed single processor systems with vector processing functions which are driven by the fastest technologies. Ease of programming and of use are taken into account. The result has been first-rate applications for supercomputers, but software, in general, lags behind that of the US. Academic access to supercomputers is also behind that in the US. The ambitious Real-World Computing Program (RWC), a 10-year Ministry of International Trade and Industry (MITI) program with a budget of about 60 billion yen (almost US $500 million), focuses on R&D for what is called flexible or intuitive information processing-the way human beings absorb information and make decisions. >

This paper proposes a programming style named co-programming for the design of basic software for massively parallel computers. This programming style well suited for computers with MIMD and Multi-SIMD control strategy. It allows programmers to design the software as a set of C fonctions, communicating by objects (whenever necessary), which are to be executed in parallel on the real machine and concurrently on the machine simulator. Example applications on the SPHINX Multi-SIMD pyramid machine are presented.

A brief background on British involvement in computer science is given, focusing on the Alvey Project. This was a five-year, $1-billion research-and-development program for information technology begun in response to Japan's Fifth Generation Computer Systems initiative, which highlighted expert systems as a major opportunity area. Britain's manpower in the area is examined. Marketplace factors are cited to explain why the British are particularly prominent in logic and human-computer interaction. Interactions among people in developing expert systems and the influence of hiring practices are examined. British projects and European Economic Community practices in which Europe is involved are reviewed. >

This article consists of a collection of slides from the author's conference presentation on NVIDIA's CUDA programming model (parallel computing platform and application programming interface) via graphical processing units (GPU). Some of the specific topics discussed include: the special features of GPUs; the importance of GPU computing; system specifications and architectures; processing capabilities; parallel memory sharing; CUDA programming models; transparent scalability; and major applications supported.

The FGCS project, aiming at developing a new generation of general purpose computers, began in 1982 and is now almost half way through the ten years of the project. The project was established to explore a qualitatively new approach to computer science in order to solve many difficult problems which have accumulated in the field. We considered that the two single most important and difficult problems are opening up a new market for computers to keep the computer industry growing, and overcoming the low productivity of software compared with hardware. We spent three years developing the target of a new broad market for next generation computers, and concluded that knowledge information processing will be the answer.

OpenMP is the industry standard technology for writing explicitly parallel programs for shared memory computers. OpenMP is easy to use with most of the constructs consisting of simple compiler directives. These directives create threads, split work between them, and manage how data is shared. In this tutorial, I will provide a comprehensive introduction to OpenMP. We will discuss the origins of OpenMP, the programming model behind it, and the basic syntax of the language. While we assume the students know Fortran or C in a Unix environment, we do not require any background in parallel programming. Finally, at this point, I can't predict how the content of the tutorial will be divided between advanced and introductory material. I will determine this at the beginning of the tutorial based on the backgrounds of the students. My goal is to make the experience valuable regardless of the parallel computing experience of the students. Proceedings of the 1st International Symposium on Cluster Computing and the Grid (CCGRID ’01) 0-7695-1010-8/01 $10.00 © 2001 IEEE

Many institutions (e.g., universities, banks) are moving towards clusters. These clusters consist of commodity components such as PCs connected by fast networks. However, the single factor limiting the harnessing of the enormous computing power of these clusters for parallel computing is the lack of appropriate software. Present operating systems are not built to support parallel computing – they do not provide services to manage parallelism. Managing the available parallelism in a cluster means managing parallel processes and computational resources, in order to achieve high performance and use computational resources efficiently, and to make programming and use of the parallel system easy. Parallelism management in parallel programming tools, distributed shared memory and enhanced operating system environments has been in the majority of trials left to application programmers. Programmers must deal not only with programming of communication and coordination of parallel processes to achieve the correct execution of an application, but also with the problems of initialisation and control of the execution on the cluster. Users do not see a cluster as a single powerful computer. Furthermore, parallel systems are seen as being user unfriendly, due to their complexity. An analysis of the existing clusters shows that parallelism management systems are being developed and offered using two different approaches: middleware, at the application level; and underware, at the kernel level. Both of them have advantages and disadvantages. The problem is how to take advantage of both of them to create the best solution from the point of view of the execution performance, programmers and efficient use of resources. In the majority of execution environments programmers of parallel application cannot make a choice between the message passing and distributed shared memory communication paradigm. Both paradigms have advantages and disadvantages. The former is fast but difficult to use, the latter is easy to use but demonstrates reduced performance. These communication paradigms and the systems supporting them are treated independently of an operating system, rather than to be parts of a comprehensive operating system as they manage system resources. Parallel processing can be divided into the initialisation, execution and termination phases. Currently, researchers and manufacturers mainly concentrate their work on the execution phase in order to achieve the best performance. Ease of use of parallel systems and programmer’s time are neglected. This approach discourages application programmers from parallel processing, as they have to program many activities, which are of an operating system nature, in particular those of the initialization and termination phases. These activities should be carried out automatically by a cluster operating system to relieve programmers from error prone and time-consuming activities. In this talk an overview of our work carried out toward the development of cluster operating systems that automatically and dynamically support parallel processing is presented. There are a number of aims of this talk. The first aim is to identify and discuss the basic issues of and solutions to the problem of the management of parallel processing on clusters. The second aim is to propose a new class of cluster operating systems that provide these services. In particular, these operating systems should: guarantee high performance of parallel processing on clusters and the efficient use of resources; support execution on a cluster of both message passing and shared memory based parallel applications; relieve programmers from error prone and time consuming work of allocation of processes to computers, management of interprocess communication and process synchronisation; provide transparency; and make the whole cluster based parallel system easy to use. The third aim is to introduce and discuss the architecture and services of a cluster operating system, called GENESIS, that allow the above specified goals to be achieved. The fourth

1 This work is supported in part by Defense Advanced Research Projects Agency (DARPA contract number F30602-00-2-0586) Abstract Researchers have long focused on dramatically increasing the size of problems that can be solved by exploiting massive parallelism. Considerable work was done on special purpose hardware for building concurrent computers and parallelizing conventional sequential programs. On the other hand, models such as Actors were inspired by the idea that using diversity and parallelism, a community of agents could collectively solve much larger problems than sequential programs. Early work on the Actor model studied the semantics of concurrency and distribution. Later work provided novel ways of implementing actors for efficiently executing and migrating them [1].

summaraze and amplzfy the dascussaon of research darectzons an compalzng for massave parallelasm that took place at'hew Frontiers: A Workshop on Future Directions in Massively Parallel Processind whzch preceded the Frontzer's '92 Conference. In addatzon to those problems specafic to compalataon for masszvely parallel machanes, problems dzscussed zncluded assues an language desagMthe success of current vectorzzang compalers and the amplacataons for compalang for massavely parallel machansand - the necesszty of nptamazzng for unzprocessor performance. 2

The article considers the theoretical basis and structural solutions of Galua's basis for the construction of microprocessor computer machine elements. The methodology for realization of vertical information technology is presented.

Properties of a multiple tree-based network called KYKLOS are presented and examined. The general form of KYKLOS is an interconnection network consisting of r sets of m-ary trees joined so that they share a common set of leaf nodes. It is shown that this structure effectively supports the parallel executions of database operations such as the join. A brief description of the hardware of a single I/O node is included. >

Current operational UK sonars use processors with throughputs in excess of five hundred million arithmetic operations per second. Several orders of magnitude increase in computing power are required to maintain long range surveillance capabilities in the 1990s and, within the next decade, typical applications will need throughputs approaching one million, million arithmetic operations per second, significantly greater than that currently achieved with fifth generation computers. This paper discusses some of the problems in realising systems with this level of performance. >

The areas of algorithms applications, architectures, and system software are discussed with reference to massively parallel computation. >

We are pursuing research on programming languages for massively parallel processing. The objective of the research is the following two points according to the top level research objective of our Massively Parallel Processing Principle Research Project: firstly to develop a prototype of a massively parallel programming language and compiler system, which is competitive to commercial language systems like data parallel C, Fortran D or HPF; and secondly to explore a massively parallel computation model, and design an experimental language as an implementation of the newly explored massively parallel computation model. >

Consider the serial emulation of a parallel algorithm. The thesis presented in this informal paper is rather broad. It suggests that such a serial emulation has the potential advantage of running on a serial machine faster than a standard serial algorithm for the same problem.The main concrete observation is very simple: just before the serial emulation of a round of the parallel algorithm begins, the whole list of memory addresses needed during this round is readily available; and, we can start fetching these addresses from secondary memories .This enables to prefetch the data that will be needed in the next "time window", perhaps by means of pipelining and have them ready at the fast memories when requested by the CPU. Thereby, utilization of commonly used memory management systems will be theoretically enhanced. Another (considerably less orthodox) implementation approach will result in abandoning locality of references , in favor of a drastically different approach: distribute memory addresses at random over memory modules, as has been proposed in the context of implementing the parallel-random-access machine (PRAM) design space.It is too early to reach conclusions regarding the significance of the thesis of this paper. Serious attempts to follow it up are needed to determine its usefulness. Parts of this paper are intentionally written in an informal way, suppressing issues that will have to be resolved in the context of a concrete implementation. My general intention is to encourage colleagues to: (i) suggest other specific approaches, and (ii) explore opportunities for taking advantage of the fine structure of parallel programs well beyond concrete suggestions made in the paper. The full paper is available as UMIACS TR 91-145, from the Institute for Advanced Computer Studies, University of Maryland, College Park, Maryland 20742--3251 .

Recognising parallel processing as a leap-frog path for supercomputing as well as the destiny of future generation supercomputers, C-DAC was launched by the Government of India as a national initiative with a first 3-year mission of designing, and bringing into commercial a state-of-the-art parallel supercomputer with peak performance exceeding 1 GFLOPS, proportionate primary and secondary storage, along with parallel programming environment, and application kernels of several applications of national importance running on the target machine. The launch of C-DAC marked the beginning of the era of high-performance computing in India. The prime delivery of the mission, accomplished with an investment of 300 manyears of effort, has been the PARAM parallel supercomputer along with the advanced parallel programming environment PARAS and a spectrum of application kernels in science and engineering. This paper presents a summary of the delivery of C-DAC mission in terms of PARAM parallel supercomputer, its architecture, programming environment and applications. A brief preview is also given of C-DAC's Second Mission to deliver a teraflops architecture Massively Parallel Supercomputer by 1997-98. >