Lapas attēli
PDF
ePub

Session M3 Model

M3-1 Self-Modifiable Computation Models for Massively Parallel/Distributed Systems

Tanaka, Jiro (Fujitsu Laboratories Ltd., Japan) (JIRO@IIAS.FLAB.FUJITSU.CO.JP)

M3-2 Development, Implementation, and Validation of a Programming Model and Programming System for Massively Parallel Computers

Giloi, Wolfgang (GMD, Germany) (GILOI@KMX.DBP.DE)

Session M4 System and Environment

M4-1 Network Supercomputing

Muraoka, Yoichi (Waseda Univ., Japan) (MURAOKA@WUCC.WASEDA.AC.JP)

M4-2 A Programming Environment Generator Parameterized with Architectures and User Interfaces OR How to Generate Dedicated Interactive Systems for Existing and Future Languages and Architectures

Kilnt, Paul (Centrun Voor Wiskunde en Informatice, Netherlands) (PAULK@CWI.NL)

M4-3 A Parallel Processing System for Multiple Computer Paradigms

Lee, Hoon Bock (Electronics and Telecommunications Research Institute, Korea) (HBLEE@KIET.ETRI.RE.KR)

M4-4 Architectures for Massively Parallel Computing

Maresca, Massino (Dist-Univ. of Genoa, Italy) (MM@DIST.UNIGE.IT)

Session M5 Language

M5-1 Design and Implementation of Programming Languages for Massively Parallel Computing Systems--Concurrent Object-Based Layered Approach

Yonezawa, Akinori (Univ. of Tokyo, Japan) (YONEZAWA@IS.S.U-TOKYO.AC.JP)

M5-2 Research on Object-Oriented Model for Massively Parallel Processing

Nakashina, Hiroshi (Mitsubishi Electric Corp., Japan) (HIROSHI@ISL.MELCO.CO.JP)

Session M6 Application

M6-1 Abstract Learning Machine for Super Parallel Computers

Refenes, Paul (Univ. College London, U.K.) (REFENES@CS.UCL.AC.UK)

M6-2 Dynamical-Simulation of Microscopic Worlds on Massively Parallel Systems

Ihara, Shigeo (Hitachi Ltd., Japan) (IHARA@HCRLGW.CRL.HITACHI.CO.JP)

M6-3 PART-DBS: PArallel and Real-Time DataBase Systems

Teranaka, Katsuni (Nippon Telegraph and Telephone Corp., Japan) (TERANAKA@SYRINX.NTT.JP)

M6-4 Research of Direct Mapping Paradigm

Tanaka, Hidehiko (Univ. of Tokyo, Japan) (TANAKA@MTL.T.U-TOKYO.AC.JP)

Session N1 Neural Model (1)

N1-1 Feedback Learning of Recurrent Neural System

Okabe, Yoichi (Univ. of Tokyo, Japan)

N1-2 Achieving a System That Grows Autonomously

Asakawa, Kazoo (Fujitsu Laboratories Ltd., Japan) (ASAKAWA@FLAB.FUJITSU.CO.JP)

N1-3 Distributed Learning Control Mechanisms for Modularized Neural Networks

Shinohara, Katsunori (Nippon Telegraph and Telephone Corp., Japan) (KATSU@NTTHIF.NTT.JP)

Session N2 Neural Model (2)

N2-1 Integrated Information Processing Technology

Nishiaura, Kazuo (Toshiba Corp., Japan) (KAZUO@SSEL.TOSHIBA.CO.JP)

N2-2 Neural Logic Networks

Shen, Zuliang (National Univ. of Singapore, Singapore) (ZULIANG@ISS.NUS.SG, ISS@NUSVM.BITNET)

N2-3 Machine Learning for Optimal Self-Organization

Refenes, Paul (Univ. College London, U.K.) (REFENES@CS.UCL.AC.UK)

Session N3 Neural Application

N3-1 Cooperative Problem Solving Based on Symbolic Information Processing System and Neural Network System Kinoshita, Tetsuo (Oki Electric Industry, Japan) (KINO@OKILAB.OKI.CO.JP)

N3-2 Multi Agents Using Neural Systems

Hoshino, Seizo (NKK Corp., Japan) (HOSHINO@CSRD.NKK.CO.JP)

N3-3 Recognition and Understanding Based on Large-Scale Neural Network Architecture--Realization of Flexible Human Interface

Togawa, Funio (Sharp Corp., Japan) (TOGAWA@CRL-SP.SHARP.CO.JP)

Session N4 Neural Hardware

N4-1 Million Neuron Parallel Processor

Yamada, Hinoru (Hitachi Ltd., Japan) (YAMADA@CRL.HITACHI.CO.JP)

N4-2 Modular Architectures for Electronic Neuro-Computers

Lee, Son-Young (Korea Institute of Science and Technology, Korea) (SYLEE@EEKAIST.KAIST.AC.KR,
SYLEE@KUMGANG.KAIST.AC.KR)

N4-3 Parallel Distributed Processing in Neural Systems Composed of Analog Elements with Chaotic Dynamics Aihara, Kazuyuki (Tokyo Denki Univ., Japan) (AIHARA@DENDAI.AC.JP)

Session 01 Optical Interconnection (1)

01-1 Electro-Photonic Processor Network

Kubota, Keiichi (NEC Corp., Japan) (KUBOTA@OEL.CL.NEC.JP)

01-2 A Research and Development of Devices for Intelligent Optical Interface Ishida, Koji (Hitachi Ltd., Japan)

01-3 2-Dimensional Parallel Optical Interconnection for Massively Parallel Computing Systems Nakamura, Masaru (Toshiba Corp., Japan)

Session 02 Optical Interconnection (2)

02-1 Multi-Dimensional Opto-Electronic Interconnection: A Hardware for Parallel and Distributed Optoelectronic Processing

Kamijoh, Takeshi (Oki Electric Industry Co. Ltd., Japan) (KKAMIJO@HILAB.OKI.CO.JP)

02-2 Research on Super-Multiplexing Optical Interconnections

Ishikana, Hiroshi (Fujitsu Laboratories Ltd., Japan)

02-3 Optical Control of High Speed Digital Electronics for Interconnecting Interfaces and Massively Parallel Systems Kaniya, Takeshi (Univ. of Tokyo, Japan)

Session 03 Optical Digital System (1)

03-1 Research on Integrated Parallel Optical Computers and Parallel Programming Ichioka, Yoshiki (Osaka Univ., Japan)

03-2 Stacked Optical Computing System

Takeuchi, Yoshinori (Matsushita Research Inst. Tokyo Inc., Japan) (TAKE@TRL.MEI.CO.JP)

03-3 Multi-Dimensional Optical Computer

Yatagi, Toyohiko (Univ. of Tsukuba, Japan) (YATAGI@OPTLAB.BK.TSUKUBA.AC.JP)

Session 04 Optical Digital System (2)

04-1 Optically Coupled Three-Dimensional Common Memory

Koyanagi, Mitsunasa (Hiroshima Univ., Japan) (KOYANAGI@SXSYS.HIROSHIMA-U.AC.JP)

04-2 Research and Development of 3D-Integrated Stacked Optical Devices for Surface-Emitting Lasers Ibaraki, Akira (Sanyo Electric Co. Ltd., Japan) (S53IBARAKI@HR105.AL.SANYO.CO.JP)

04-3 Digital Optical Image Processing

Digital Optical Numerical Computers

Micro-Optics

3D-Integration of Optical Components

Brenner, Karl-Heinz (Univ. of Erlangen, Germany)

Session 05 Optical Neural System

05-1 Optical Neurocomputing - Device and Architecture Technologies

Kyuma, Kazuo (Mitsubishi Electric Corp., Japan) (KYUMA@QUA.CRL.MELCO.CO.JP)

05-2 Real-Time Optical Neurocomputing System with Learning Capabilities

Ishikawa, Masatoshi (Univ. of Tokyo, Japan) (ISHIKAWA@K2.T.U-TOKYO.AC.JP)

05-3 Modular Architectures for Electro-Optic Neural Networks

Lee, Son-Young (Korea Institute of Science and Technology, Korea) (SLEE@EEKAIST.KAIST.AC.KR,
SYLEE@KUMGANG.KAIST.AC.KR)

05-4 Adaptive Optical Processing for 3D Vision

Chrostowski, Jacek (Institute for Information Technology, Canada)

COMPUTER SECURITY IN JAPAN

The state of progress in computer security activities in Japan is assessed. Attendance at the Tokyo Symposium on Information Security and industrial visits showed that the Japanese are at an early stage of development in this area, probably because of different social conditions from the West. However, as part of their vision of "global informatization," they are now making efforts to bring computer security up to a similar level to the West.

INTRODUCTION AND DISCUSSION

This is a report on a short trip to Japan to study progress in Japanese computer security. On 17 and 18 October 1991 I attended the International Symposium on Information Security in Tokyo, organized by the Japan Information Processing Development Center (JIPDEC). In addition, I arranged visits to the research laboratories of Fujitsu, NEC, and Sony.

I found the state of development of computer security at an earlier stage than I had expected from progress in other areas of computing. To illustrate this, a survey of 217 members of the Fujitsu Large System Group (mainframe users) showed that only 44% at present control access to files using an access control system, with a further 31% planning to do so soon. So 25% of this sample of mainframe users have no plans to introduce file access control. This is only one statistic, selected from a wider survey, but it is consistent with a number of conversations that I had and with the tone of the Information Security Symposium. This was a very high level meeting, but its program showed that Japan is a country that has not advanced very far in the development of computer security. The program contents were rather general

by Jonathan D. Moffett

and 9 out of 13 speakers were Westerners.

An explanation for these features can be found in the introduction to the announcement of this symposium: "Social stability cannot be maintained in an age of global informatization unless all countries uniformly adopt the same level of minimum security measures." The Japanese believe that they have less of a computer security problem than Western countries, but that they cannot isolate themselves from global systems, and so they must bring their own security up to an adequate level. Some facts are clear: Japan has a high rate of natural disasters (earthquakes and typhoons) and a very low reported crime rate. (Skepticism about the rate of unreported crime should be tempered by the fact that even the murder rate, which we can assume to be reasonably well reported, is only about 25% of that in Europe.) There has, therefore, probably been much less need in the past for computer security than in Western countries, except for physical measures to deal with natural disasters.

The Ministry of International Trade and Industry (MITI) issued Computer Systems Security Guidelines in 1977 and has updated them twice since. I was told that they are mainly concerned with physical security. They issued

System Auditing Guidelines in 1985. The first overt official response to technical computer insecurity problems was in 1990, when Computer Virus Measures Guidelines were published. Viruses have been the main publicly discussed "technical" problem, as can be seen from the program of the symposium. However, there appears to be a low level of installed security to deal with network hackers, and it would be surprising if international hackers had not attempted to penetrate Japanese networks, and succeeded. Indeed, there is reason to think that they have.

There is one exception to the relatively low level of security awareness and progress, and that is in the area of encryption. Every single one of the papers that I have been given is concerned with the design or use of encryption systems. I can only speculate why this should be the case, since it is inconsistent with the drive of the symposium, which was towards a more rounded, management attitude to security. Perhaps it only reflects the tradition of all research establishments to pursue what interests the researchers rather than what the market immediately requires.

If this were a Western symposium, one might be tempted to view the expression of unfulfilled intentions as an alternative to action rather than a plan for it. However, the Japanese have a

formidable track record of achieving their stated intentions, and it is reasonable to assume that they will do so in this case also. The plans were most clearly stated by Toshio Hiraguri's "Fujitsu Direction for Computer Security":

• Supply of products and functions which satisfy the demand for security

• Training

as for their real threat to computing),
all of the talks were related to security
management rather than technical
issues. The subjects covered were those
that are natural for a country that is
about to come to terms, in a serious
way, with computer security: risk assess-
ment, security management, and pro- INDUSTRIAL VISITS
motion of security awareness. Third, 9

My overall impression was of an event that is designed to motivate activity in this area and to leave none of the delegates in any doubt that this is an area of work that they must take seriously from now on.

out of 13 speakers were Westerners. Fujitsu
Japan has not yet gained the expertise
to be able to provide enough speakers
with the requisite experience for a

• Contribution to standardization meeting of this kind. True, this was an
activities
international symposium, but there can
be no doubt that there would have been
a higher proportion of Japanese giving
talks if this had been possible.

Certification and evaluation of products

The achievement of these goals, by Fujitsu and others, will put Japan back on the mainstream of computer security in developed countries.

INTERNATIONAL SYMPOSIUM
ON INFORMATION
SECURITY IN TOKYO

At this symposium about 1,000 delegates attended. The great majority were Japanese. The symposium consisted of a first day with four keynote speeches after the introductions and a second day with three parallel sessions, each of three speeches and a panel session.

The symposium had several notable features. First, it was a very high level meeting. The Minister of International Trade and Industry was slated to open it, although in the event he sent his deputy. The President of NEC and the Managing Director of Fujitsu gave full-length talks. An impressive array of Americans and Europeans was brought in. Second, a glance at the program showed that this was a symposium for a country that has not gotten very far along the road of development of computer security. Apart from viruses (presumably brought in as much because they are a fashionable security concern, easy for an outsider to relate to,

There were three Japanese speeches, by the President of NEC, the Director of MITI's Information Service Industry Division, and the Managing Director of Fujitsu Ltd., which were useful for assessing their view of the current situation and future plans.

There was a high degree of overlap in their contents, and I got the following message from each. Japan has historically been a country with a low crime rate and a high vulnerability to natural disasters (earthquakes and typhoons). The computer security measures in the past have been addressed to these weaknesses. However, the "Information Society" is becoming global in scope, and it is therefore necessary for the Japanese to address the computer security problems that have up to now been mainly of concern to the West. This is because global communications mean that global systems must have a consistently high level of security, and in any case it is impossible to isolate Japan from problems imported from the West, such as viruses. Japan will therefore study what needs to be done, including Western experience. It will create standards that are compatible with, but not necessarily identical to, those in the West, and will take the necessary actions to provide security in its products.

On 14 October 1991 I visited Fujitsu Laboratories in Kawasaki. My host was Ryota Akiyama, a senior researcher. I also spoke with Naoya Torii (researcher) and Haruki Tabuchi (a computer systems development manager). Akiyama and Torii gave me copies of two published papers (Ref 1 and 2) and told me about their work on an IDbased file security system. Its aim is to integrate login control, file access control, and encryption of data communication by an ID-based key management system. It is an interesting attempt to take an integrated approach, but it seems to have potentially severe performance problems when used in conjunction with large files. Also, there had been little consideration at this stage of how it would fit into a heterogeneous open-system world. It is a research project that appears at present to be some distance from the market. Akiyama and Torii also gave me a brief description of a network security module, to be connected to the extension bus of a Fujitsu PC, for network security and data protection. It is to be demonstrated at Asiacrypt '91. [Asiacrypt '91 is an English language symposium on encryption to be held 11-14 November 1991 in Fujiyoshida (at the foot of Mt Fuji).]

Tabuchi took a lively part in the discussions, and I had further conversations with him at the security symposium. He takes a direct practical interest in how to deliver security to customers and raise their awareness of its importance. He expressed interest in risk analysis methods, data protection

« iepriekšējāTurpināt »