Since 1993, NACSIS has been conducting a research and development project entitled ÒDevelopment of High-Perform-ance Communication Network for Scientific ResearchÓ, which is categorized in ÒCreative Basic ResearchÓ (commonly known as the ÒNew ProgramsÓ) of Grant-in-Aid for Scientific Research of the Ministry of Education, Science, Sports and Culture.
The objectives of this project are 1) to develop the future operation both of the high-performance LAN for campus networks whose installation has been increasingly pursued by universities and inter-university research institutions, and of the ATM based high-speed Science Information Network maintained by the NACSIS, 2) to demonstrate applications on high-performance communications networks, and 3) to conduct experimental studies to verify those feasibilities.
The rapid expansion of the Internet and the increasing awareness of the Internet as a basic tool for scientific research are trends which hardly need to be mentioned here. As a result, the configuration of campus networks is drastically changed in such a way that the existing FDDI is giving way to the ATM-LAN where greater development in terms of functions is expected. At the same time, even the Science Information Network (SINET), which is a wide-area backbone network, is moving toward a high-speed and flexible communications network configuration which is enabled by ATM technology. In spring, 1996, these latest facilities were introduced into all the com-munications nodes of the Science Information Network, and more than 40 universities.
ATM (Asynchronous Transfer Mode) is similar to the packet communications used by computer networks in the past. However, it enables transfer speeds beyond 100Mbps, it handles not merely data but voice and moving images as well, and it is believed to best suit the needs of todayÕs multimedia com-munications. Adoption of the ATM technology in campus networks enables work stations to be connected via high-speed interfaces, increases the speed at which information is accessed, and provides assured processing of high-quality sound and moving images. This is likely to promote the diversification of science information, stimulate joint experimentation between researchers and increase the level of sophistication in campus networks which can be used widely for lectures and even for tele-medicine and tele-surgery at university hospitals. In 1994, the first operation of ATM began to be applied in wider areas with the Science Information Network and, by way of adding to this, preparations are underway to introduce a new ATM technology with functions conducive to connection with ATM-LAN and to the stable operation of high-performance networks. However, ATM is still incomplete as a technology. Some examples will be given below.
The modes of data transfer for multimedia communications are many and varied, and the handling involved is not simple. With WWW or file transfer, data transfer rate is affected by significant variations (called ÒburstsÓ). This does not pose a problem if a communications capacity (bandwidth) sufficient for the peaks of these bursts to be transferred is provided at all times, but the entire bandwidth is limited. In order for multiple researchers to share the capacity, either individual bandwidth, therefore, must be restricted or interference between com-munications must be suppressed. This necessitates a technology known as Òtraffic control,Ó and this technology is still far from being perfected.
ATM, which is designed for multiplexed transmission, does not know the details of the communications applications. For instance, when lecture images are to be transmitted, it is desirable for the images to be free from distortion and have the definition which makes them suitable for showing on a large screen; and in tele-medicine in which medical equipment is operated from a remote location, delays must be kept to the absolute minimum so that the sense of operating the equipment is not impaired. Since the requirement for communication like these which need to be considered at the time of transmission differ according to the application, operators must be provided with indication from the application in the form of Òquality of serviceÓ (QoS), and a means must be made available to make it all possible. This means that we need quick development of QoS control technology to accommodate with requirements.
The Internet must operate even top on high-speed communication. As it stands also today, the Internet is beset with a large number of outstanding issues concerning how it will shape up in the future: these issues include the very urgent need to improve operation of the Internet, the validation and assessment of new LAN operations called virtual LANs, and the ÒIP over ATMÓ technology which has attracted attention as the next-generation Internet.
As far as applications are concerned, a wide range of research fields remain to be explored starting with the validations conducted in an high-speed network environment for those applications, such as electronic libraries, whose development is being pursued by NACSIS, the development and assessment of a prototype which will efficiently store and retrieve the still or motion images, and the validation of the future architecture of distributed processing. These are areas to be developed by NACSIS with the cooperation of frontier researchers from all parts of the country. The experiment of each of these issues in turn and the feasibility demonstration of the results are the focus of the research and development project mentioned at the beginning.
Fortunately, Nippon Telegraph and Telephone Corporation (NTT) has planned experiments on the joint use of multimedia testbed. It has urged universities, national research institutes and the research institutes in the private sector to participate, and the objective is to sound out the kind of high-speed services which NTT will offer in the future. Starting from the conceptual stage of its research and development project, NACSIS has been consulting with NTT, and it has drawn up research themes and put together a plan for utilizing the communications testbed provided by NTT. The figure shows the configuration of the current experimental network which is completely separate from the Science Information Network. A series of developments and feasibility demonstrations are being conducted using this experimental network.
The research and development program is conducting under a 5-year plan until March, 1997. This year marks the midpoint of the plan, and the research is being assessed by the Ministry of Education, Science Sports and Culture. The result of the assessment has been quite favorable. At this stage, interested parties were invited to participate, and on November 29 and 30, 1995 an International Symposium was held at the NTTÕs Musashino Research Center. Researchers pursuing the same research goals in the United States and United Kingdom were also invited to this event, and details of each research project were compared. The symposium brought together 230 individuals including attendees from four countries overseas. On the two days concerned, research findings were presented, and this attracted a great deal of interest and won high critical praise.
As was mentioned previously, the universities and NACSIS have teamed up together and started work on constructing the high-speed network of the next generation. The results achieved by this research project will also be used in this work. For the time being, they will be employed in traffic control. In the future, however, the plan is to put forward a wide range of presentations for application through the staging of trial services, etc.
More than 50 researchers and twenty research institutions are involved in our research project. We aim to urge others to join in wherever possible to bring about even better results. We plan to report on our progress in our publications.
Finally, I would like to express my sincere appreciation for the assistance and support given to us by the many researchers who are participating in the research work and all those concerned at the Ministry of Education, Science, Sports and Culture, and NTT.