Kyushu University Academic Staff Educational and Research Activities Database
List of Papers
Yoshiaki Kasahara Last modified date:2024.04.10

Assistant Professor / Section of Advanced Network and Security / Research Institute for Information Technology


Papers
1. A Preliminary Study on Gathering SMTP Responses for Email Delivery on the Internet.
2. Yoshiaki Kasahara, End of Basic Authentication and Migration to Modern Authentication for Exchange Online, Proceedings of the 2023 ACM SIGUCCS Annual Conference, 10.1145/3539811.3579560, 2023.03, At Kyushu University, Information Infrastructure Initiative provides an email se
rvice named "Primary Mail Service" for students and staff members with Microsoft
Office 365 Exchange Online. On September 20th, 2019, Microsoft announced the en
d of support for Basic Authentication for Exchange Online, which is considered v
ulnerable to identity leakages such as phishing and malware attacks. Microsoft w
ould require users to use Modern Authentication such as Exchange protocol or OAu
th 2.0 authorization with IMAP, POP, and SMTP. Historically we had instructed ou
r users to use IMAP or POP and SMTP protocols for their email applications, incl
uding Microsoft Outlook and Mozilla Thunderbird, so disabling Basic Authenticati
on would significantly impact our user population. In September 2021, Microsoft
announced the end of September 2022 as the hard deadline for disabling Basic Aut
hentication. Based on available information, we prepared migration documents fro
m Basic Authentication to Modern Authentication and started to notify users to a
bandon Basic Authentication. Sending messages to users did not seem to be effect
ive after a couple of notifications, so we tried to temporarily disable Basic Au
thentication to realize the remaining users through authentication failures. In
this paper, we would like to share our experiences about the effect of retiring
Basic Authentication for Exchange Online on our service and users..
3. A Consideration on a Sender-Side Spam Scoring Header of E-mail Messages.
4. , [URL].
5. A Supplementation Method for Delegation of Address Management on a Cloud Email Service
In providing organizational email service, the use of cloud email services contributes to reducing operation and management costs, generally. In most cloud email services, the management of a certain part of email addresses is not delegatable to administrators of organizational units, but this causes operational load concentration on organizational service administrators in large-scale organizations. For solving this issue, this study aims to enable delegations of email address management to administrators of organizational units. This article introduces a method for delegation of email address management and a supplemental system onto a cloud email service. The proposed method divides the management of email addresses within internet subdomains for organizational units from account administration and delegates the email address management to subdomain administrators. The present system for email address management is designed based on microservice architecture and configured in serverless. This microservice architecture and serverless configuration reduces operation and management costs and contributes to the maintainability and extensibility of the present system. Furthermore, the use of an identity-as-a-service for authentification, authorization and storing privilege information ensures confidentiality and security of system accounts., [URL].
6. , [URL].
7. Migrate Legacy Email Services in Kyushu University to Exchange Online.
In Kyushu University, Information Infrastructure Initiative provides an email service for students and staff members, called “Primary Mail Service”. We had operated an on-premises system for this service, and the lifetime of this system would end in early 2019. We needed to reduce costs for replacing this system because our university had just finished a major campus migration. We compared some options such as building a yet another on-premise system and migrating to a cloud-based email service and finally gave up the on-premise option because we couldn’t afford replacement and operational costs of another on-premises system anymore. We selected Microsoft Exchange Online as the new service mainly because we already had a contract with Microsoft and been operating an Office 365 tenant. We had additional requirements for user provisioning and services which were not available in Exchange Online, so we had to implement and maintain additional systems on top of it. On December 18th, 2018, we successfully migrated the email service to Exchange Online. By coincidence, Kyushu University Administration Bureau decided to migrate their in-house Exchange server to Exchange Online. After some discussions, they concluded to migrate their domain to the same tenant with Primary Mail Service. Other than that, there are more than a hundred legacy email servers inside our campus network operated by various departments as subdomains of kyushu-u.ac.jp. We are designing a plan to consolidate them into our tenant of Exchange Online to reduce a budget and human resource costs, and to improve security. In this presentation, we share our experiences about migrating our campus-wide email services to Exchange Online. We also discuss why we want to consolidate other legacy email servers and how to implement the plan..
8. Renovation of the Office 365 environment in Kyushu University: Integration of Account Management and Authentication.
Office 365 Education is a suite of cloud services for students and educators. Kyushu University has provided Office 365 accounts for all students and staff. The first generation of an environment for Office 365 provisioning in the university had several issues about associating between Office 365 accounts and member identifications of the university. All university members are randomly assigned unique identifiers by the central ID management system for using commonly in university-wide information services. Since the IDs are for internal use only, the first environment authenticated a user with another ID and password specific for Office 365. In addition, processes for assigning licenses and giving privilege to users of Office 365 depending on modifications to member information in the ID management system were not fully automated. This paper shows how we resolved problems integrating Office 365 into the ID management of the university by rebuilding the infrastructure. We configured a federated authentication system and developed a system for processing in events of the account life cycle..
9. 個人情報保護と情報セキュリティを考慮したOffice365Educationの環境構築方法.
10. A Study on Confidential Information Protection in Kyushu University.
11. Restructuring of Service Infrastructure for Office 365 in Kyushu University
九州大学では2016 年からOffice 365 を全学サービスとして正式に提供してきたが、その基盤システムやテナント運用などについて様々な課題が生じていた。そこで、基盤システムとテナントを含めて環境を新たに再構築し、2018 年4 月から新しいサービス環境の提供を開始した。本稿では、従来の環境における課題とその解決策、さらに、新しい環境の構築過程で生じた問題とその解決方法などについて紹介する。.
12. 精緻に制御可能な恒常性のある高集積マルチアカウント型のメール基盤.
13. Flexible Hosting/Cloud Platform Based on Light-Weight Containers and its Evaluation and Stress Test Environment.
14. A Design of Elaborately Analyzable Homeostatic E-mail System.
15. Flexible Hosting/Cloud Platform Based on Light-Weight Containers and its Evaluation and Stress Test Environment
As basic components of ever-expanding internet services, hosting services and cloud computing platforms are expected to satisfy various requirements, such as efficiency, fault-tolerance, resiliency against high load, flexibility, and security. We are trying to solve these problems with a light-weight container based architecture (called FastContainer). To improve FastContainer (especially its auto-scaling function), we are creating an environment for performance and stress tests on a public cloud platform. In this paper, we describe the overview of FastContainer and details of the test environment, its current status, and future plans..
16. Yoshiaki Kasahara, Takao Shimayoshi, Masahiro Obana, Naomi Fujimura, Our experience with introducing microsoft office 365 in Kyushu University, Proceedings ACM SIGUCCS User Services Conference, 10.1145/3123458.3123491, 131713, 109-112, 2017.10, Information Infrastructure Initiative of Kyushu University started serving Office 365 Education for all students and staff members at Kyushu University in November 2016. Since 2007, the university had signed Microsoft EES (Enrollment for Education Solutions) including licenses for the latest Microsoft Windows and Office suite. The EES agreement includes an advantage to provide Office 365 Education to the university members with minimum investments, and there was a demand for Skype for Business which is included in Office 365. To deploy Office 365 for our users, we first needed to configure our on-premises user authentication infrastructure to coordinate with Office 365. During trials, we had a couple of difficulties attributed to some disagreements between Microsoft's and our policy on whether the user identifier, namely the user principal name in Active Directory, was open or private. Additionally, we had to consider which services should be applied to the users, because we have been operating an on-premises email service which is competing with Microsoft Exchange mail service. In this presentation, we share our experiences in Office 365 deployment..
17. A study of disk usage distribution analysis to set optimal quota value
To prevent shortage of storage space in a service system, an administrator usually set per-user quota as an upper limit of usable space for each user. To prevent service failure caused by resource exhaustion, an administrator tends to set a conservative quota value such as the storage capacity divided by the expected maximum number of users. In this research, authors analyzed long-term storage usage history of our email system and file sharing (archiving) system in Kyushu University. Mostly through the analyzed period, the usage pattern showed a long-tailed distribution similar to lognormal distribution. Also the overall storage consumption slowly increased during the analyzed period. Based on these analysis, the authors defined "storage utilization ratio" to evaluate how the storage is effectively used. By approximating a disk utilization pattern as a power-law distribution, we proposed a method to calculate an optimal quota value to maximize the utilization ratio. We applied the method to the actual data to calculate the best quota value to maximize the utilization ratio..
18. A study of disk usage distribution analysis to set optimal quota value
ストレージの容量不足を防ぐため,システム管理者は各利用者の使用上限値 (クォータ値) を設定する.システム管理者はサービスの安定提供のために,少なめのクォータ値を設定しがちである.また,クォータ値に単純に容量を利用者数で均等割りした値を設定することも多い.本研究では,筆者らが所属する九州大学の電子メールとファイル共有 (蓄積) システムを対象に,長期間のディスク使用量を分析した.ほとんどの時点で,利用者のディスク使用量はロングテール型の分布をしており,その分布は対数正規分布に近いことが分かった.長期間のディスク使用量の推移を調べた所,全体の使用量が微増していたことも分かった.ストレージの活用度を定義し,それに基づく評価方法を考案した.さらに,ディスク使用量分布の冪分布近似により,活用度が高くなるクォータ値の算定方法を考案した.実際のデータに基づき,ストレージの活用度を最高にするクォータ値も算出した.To prevent shortage of storage space in a service system, an administrator usually set per-user quota as an upper limit of usable space for each user. To prevent service failure caused by resource exhaustion, an administrator tends to set a conservative quota value such as the storage capacity divided by the expected maximum number of users. In this research, authors analyzed long-term storage usage history of our email system and file sharing (archiving) system in Kyushu University. Mostly through the analyzed period, the usage pattern showed a long-tailed distribution similar to lognormal distribution. Also the overall storage consumption slowly increased during the analyzed period. Based on these analysis, the authors defined "storage utilization ratio" to evaluate how the storage is effectively used. By approximating a disk utilization pattern as a power-law distribution, we proposed a method to calculate an optimal quota value to maximize the utilization ratio. We applied the method to the actual data to calculate the best quota value to maximize the utilization ratio..
19. Yoshiaki Kasahara, Eisuke Ito, Naomi Fujimura, Introduction of new Kyushu university primary mail service for staff members and students, Proceedings ACM SIGUCCS User Services Conference, 10.1145/2661172.2662965, 103-106, 2014.11, In the end of fiscal year 2013 (March 2014), Kyushu University Information Infrastructure Initiative introduced new Primary Mail Service for Staff Members and Students. The previous service for staff members had been built using proprietary mail and LDAP appliance, but tight user license and the nature of proprietary system caused several troubles. The previous service for students had been built using open source software (Postfix and Dovecot), but there were some issues with the old implementation of LDAP authentication. With these experiences, we decided to design a new system by merging both system using open source software including Postfix, Dovecot, OpenLDAP, and Squirrelmail. We also extended alias address setting service (previously available for students only) to all the members including staff members. In this paper, we explain the design and implementation, user migration, current status and future works of our new mail service..
20. Introduction of new student-long user ID for intra-institutional information services
Integrated user authentication platform realizes secure and easy use intra-institutional services. Most universities and academic institutions have an integrated user account database, and construct an authentication platform. In Kyushu University, the user ID of a student was the same with his/her student ID. However, there were some problems around security and availability in user ID based on student ID. Since student ID was used for the mail address of student, it is easy to leak outside. Additionally, student ID includes a serial number, and then it is easy to guess other IDs from one ID. Student ID is issued to a student at the day of entrance ceremony, and then it is impossible to use university information services before entrance. ID continuity is also the problem. Student IDs for a person are deferent in between graduate school and in graduate school. Then, personal data for an account cannot continue during the student life. To solve these problems, Kyushu University decided to introduce another student-long user ID service. This paper reports the new user ID, ID management system, and the effect of introduction of new user ID..
21. Yoshiaki Kasahara, Eisuke Ito, A study of network issues for implementing large-scale academic cloud service, IEICE Technical Report, 113, 240, 35-40, 2013.10, The virtualization technology of computer resources has matured enough for production services, and cloud services became popular in various fields. Cloud services are also gradually incorporated in universities' research and education activities. There is a demand for cloud infrastructure suitable for academic activities. To implement practical and user-friendly services on private, community, or public cloud services, we need to consider about network design. This paper tries to clarify network issues for implementing cloud services considering use cases, the number of instances, utilization of IP addresses, virtual private network, client access networks, and so on..
22. A feasibility study of educational system on IaaS cloud
With advancement of virtualization technology, various kinds of cloud services are widely available such as Infrastructure-as-a-Service (IaaS) type services. By using IaaS cloud infrastructure, it is possible to outsource educational information system in a university to an external service provider and reduce operational and maintenance costs. We are investigating feasibility of implementing PC-equipped classroom in educational information system over IaaS cloud infrastructure. In this paper, we discuss problems when using virtual machines remotely, and report the results of experiments measuring RDP performance and bandwidth usage..
23. A Bot Detection Method Using Hierarchical Clustering Based on Mechanical Communication Behavior Model
ネットワーク上に存在するボットに感染した端末を特定するために,ネットワークアプリケーションが送信するアプリケーションプロトコルメッセージの送信間隔に着目したボット検知手法を提案する.人間がアプリケーションを操作した場合,そのアプリケーションプロトコルメッセージの送信間隔はばらつくのに対し,ボットの場合はその挙動がコードによって規定されているため,アプリケーションプロトコルメッセージの送信間隔の分布に偏りが起きる.この分布の違いを利用して,人間とそうでないものを区別することで,ボットを発見する.はじめにネットワークアプリケーションに対する入力と出力の関係をモデル化し,IRCクライアントのIRCメッセージ送信間隔のモデル化を行う.続いて,提案モデルに従ったボット検知アルゴリズムを設計する.アルゴリズムでは,IRCメッセージの送信間隔の列に対して階層型クラスタリングを適用することで人間と機械のモデルの区別を行う.評価では,モデルに基づく擬似データを用いてアルゴリズムのパラメータを設定した後,実際の人間が操作するIRCクライアントで観測されたIRCトラフィックならびにIRCボットのトラフィックに対して提案手法を適用した.その結果,IRCボットのトラフィックは機械が生成したトラフィックとして正しく判定された.In this paper, we propose a bot detection method which focuses on application protocol message transmission intervals of applications in order to find bot infected machine on a network. Our method predicts who is controlling the application by monitoring its network behavior, especially application protocol message transmission intervals. An application which is operated by a human has random behaviors due to the human operation, while a bot has a mechanical behavior since its behavior is written in its own code. First, we build a model of network behavior of human and non-human operated application and we find that several samples follow the model. Then we design a bot detection algorithm using a hierarchical clustering. In evaluation phase, we set parameters in the algorithm with artificial data based on our proposed model and then we evaluate our method with real human IRC traffic and IRC bot traffic. Our method correctly judges bot traffic as machine-generated one..
24. A Study of VCL in Graduate School of ICT.
25. Google makes a chronic big load to university mail server
九州大学では,従来から大学ドメインのメールサーバを学内に構築し,構成員へメールサービスを提供してきた. 2012 年 1 月頃より,学内の情報サービスに対し利用者認証機能等を提供する全学認証サーバの負荷の高さが問題となり,その原因の一つが学生向けメールサーバであることが明らかになった.詳細な分析の結果, Google 社の Gmail から本学のメールサーバへ持続的なアクセスがあり,中でも既に卒業などで消滅したアカウントへのアクセスが多数あることが分かった.本稿では,本学の学生メールサーバのアクセスログ解析に基づいて, Gmail がメールサーバへ与える負荷状況の分析とその理由について述べ,対応策について検討する.Traditionally, Kyushu University has been providing email service using its own domain name for staff members and students of the university. Around January 2012, we noticed that the high load of the university authentication server, and we realized that one of causes was the access from the mail server for students (called Student Primary Mail Service). Detailed analysis showed that there was chronic big load produced by Google's Gmail, especially toward nonexistent accounts removed due to graduation. In this paper, we explain the current situation and reasons of the big load induced by Gmail and its possible countermeasures based on the analysis of access logs for Student Primary Mail Service..
26. DNS traffic analysis to detect the botnet based on the query of the same domain
近年、ネットワークを利用した脅威の中で、多数のホストが連携したボットネットによる攻撃が問題視されている。ボットネットは金銭目的で使われる組織化された攻撃プラットフォームであるため、この検知は重要課題となっている。
我々はボットネットの早期検知を目標とし、本研究ではDNSトラヒックを用いて、複数の始点が同一のドメインに関する情報の問い合わせを行うパケットのうち、その応答パケットを抽出する。そして、その問い合わせ対象のドメイン名を用いて提供しているサービスの悪性について調査する。悪性については、既存の、動的アクセスに基づく悪性サイト判定システムと連携し、評価を行う。.
27. A study of LDAP load balancing for University ICT services
近年,大学では全学的な認証基盤構築が進んでいる.筆者らが所属する九州大学でも全学共通 ID の発行および認証基盤を構築し,学内向け情報サービスの利用者認証を一元化している.認証基盤を利用するサービスの増加に伴い, LDAP 認証サーバの負荷が上昇している.中でも電子メール利用時の認証と,無線LAN接続時の認証が認証サーバの負荷を増大させている.本論文では,九州大学認証基盤における負荷状況の解析結果を述べる.また,負荷分散の方法を提案し,実際に負荷分散を適用した効果について,短期間ではあるものの,その結果を示す.Integrated user authentication platform may realize secure and easy use intra-institutional services. So, most institutions have integrated user account, and constructed internal authentication platform. The more clients depend on the central authentication server. The loads of the authentication server will become much higher. In Kyushu University, which authors belong, the load of the authentication server becomes very high because of authentication of e-mail and wireless network service. In this paper, we analyze the load condition of the authentication server, and study some load balancing..
28. A study of LDAP load balancing for University ICT services
Integrated user authentication platform may realize secure and easy use intra-institutional services. So, most institutions have integrated user account, and constructed internal authentication platform. The more clients depend on the central authentication server. The loads of the authentication server will become much higher. In Kyushu University, which authors belong, the load of the authentication server becomes very high because of authentication of e-mail and wireless network service. In this paper, we analyze the load condition of the authentication server, and study some load balancing..
29. Primary Mail Service for Students Based on their Names
九州大学情報統括本部では学生に提供しているメールサービスとして、従来は学生番号をベースとしたアドレスで提供していた。しかしながら大学院に進学時にアドレスが変わるなどの問題があると考え、姓名をベースにしたアドレスも使えるようにメールシステムを再構築して 2011 年 4 月からサービスを開始した。ここではシステムの基本的な考え方、実装、運用に至るまでの問題点、運用の実際について報告する。Information Infrastructure Initiative provided the e-mail service based on Student ID to all the students in Kyushu University. However, the e-mail address based on the student ID changes when a student graduates from the department and enters the graduate school because a student gets a new student ID at that time. We implemented the new e-mail service system to improve the problem, and started the service in April 2011. Students can use the same e-mail address while they are the students in Kyushu University. We describe the purpose, operation, and experience with the new e-mail service..
30. Seiichiro Mizoguchi, Yoshiro Fukushima, Yoshiaki Kasahara, Yoshiaki Hori, Kouichi Sakurai, Darknet Monitoring on Real-Operated Networks, 2010 International Conference on Broadband, Wireless Computing, Communication and Applications, 10.1109/bwcca.2010.82, 2010.11.
31. Seiichiro Mizoguchi, Yuji Kugisaki, Yoshiaki Kasahara, Yoshiaki Hori, Kouichi Sakurai, Implementation and evaluation of bot detection scheme based on data transmission intervals, 2010 6th IEEE Workshop on Secure Network Protocols, 10.1109/npsec.2010.5634446, 2010.10.
32. BS-7-7 Attribute control in IdP for federated web services.
33. Masanori Nakakuni, Hiroshi Dozono, Eisuke Ito, Yoshiaki Kasahara, Hideaki Nakakuni, A Method of Personal Authentication by Shape Recognition of the Lips and Front Teeth, MACMESE'09: Proceedings of the 11th WSEAS international conference on Mathematical and computational methods in science and engineering, 203-207, 2009.11, In this paper we propose a method for carrying out personal authentication by recording a photo of a computer user's face, recognizing the shape of their lips and front teeth, and performing image matching. Using this method, personal authentication is carried out by comparing the previous and current shape of a user's lips and front teeth when they log on to a computer. This method has a number of merits. As teeth are normally hidden behind the user's lips, user authentication is not carried out automatically as is the case with facial recognition. This means that it is possible for the user to decide for themselves whether or not to carry out user authentication. As it is not possible for a user to be recognized simply by their face being recorded on camera, their privacy is protected. Additionally, as the shape of lips and front teeth vary between individuals, there is a high probability that a user can be correctly identified. Furthermore, there is a low probability of significant changes to the shape of adult teeth over a period of months or years unless major dental work is carried out, so this method is potentially suited to personal authentication. Another merit of this method is the fact that it is possible to perform user authentication using an inexpensive webcam. In this paper, we will detail experiments using this method to perform personal authentication as well as the subsequent results..
34. A method of automatic user authentication by fulltime monitoring of keystroke timings
This paper will propose a method of user authentication that constantly monitors the keystroke timings of a computer user, identifies a user based on the characteristics of keystrokes and discovers spoofing. This method authenticates users by comparing past and current keystroke timings when a user is working on a computer. This method has the following two main advantages. One is that it discovers spoofing even when users change while logged in. When a user pushes keys to operate a computer, the computer automatically performs personal authentication. Users do not have to take any special action for personal authentication. This paper will also propose how to enhance security using these characteristics in combination with existing user authentication methods..
35. Nakakuni Masanori, Ito Eisuke, Kasahara Yoshiaki, Inoue Sozo, Dozono Hiroshi, Construction and Use Examples of Private Electronic Notary Service in Educational Institutions, WSEAS Transactions on Advances in Engineering Education, 5, 10, 676-686, Issue 10, Vol. 5, 2008.10, [URL], People have many documents. For example, a variety of documents are prepared and used in public institutions. As the internet becomes widely available in recent years, paper documents are being replaced with electronic data, which are often distributed in the form of electronic data without being printed out. Similarly, in educational institutions, increasing number of documents are distributed in the form of electronic data. Such data are distributed through various routes and means, and prone to the risk of alteration in the process. Data may be protected against alteration, but it is difficult to completely prevent data alteration in the distribution process. Data can be generated with electronic signature that allows for the identification of data creator and possible alterations by third parties. This method is, however, not valid if the data becomes separated from the electronic signature, making the validation of data creator or data alterations difficult or impossible. In this paper, we describe the invention of a system that, even in cases where data is separated form the electronic signature, enables easy identification of possible data alterations by the electronic signature management. And we describe here an exploratory construction of private electronic notary service in university. We also add a review on the utilization method of private electronic notary service in universities..
36. Masanori Nakakuni, Eisuke Ito, Yoshiaki Kasahara, Hiroshi Dozono, Private Electronic Notary Service in Universities and Its Utilization in Education, The 4th WSEAS/IASME International Conference on EDUCATIONAL TECHNOLOGIES (EDUTE '08), 170-+, 2008.10, A variety of documents are prepared and used in universities and colleges. As the internet becomes widely available in recent years, paper documents are being replaced with electronic data, which are often distributed widely through websites and other electronic resources. Such data are distributed through various routes and means, and prone to the risk of alteration in the process. Data may be protected against alteration, but it is difficult to completely prevent data alteration in the distribution process. Data can be generated with electronic signature that allows for the identification of data creator and possible alterations by third parties. This method is, however, not valid if the data becomes separated from the electronic signature, making the validation of data creator or data alterations difficult or impossible. In this paper, we describe the invention of a system that, even in cases where data is separated form the electronic signature, enables easy identification of possible data alterations by the electronic signature management. Construction of a prototype system is also described. We also add a review on the utilization method of private electronic notary service in universities..
37. A Fusion of PKI and OpenID for inter-domain membership services
Recently, demand for inter-domain service is arising, such as web mash-up services. For inter-domain services, it needs inter-domain user authentication mechanism, because most services are membership oriented, and those services must implement user authentication mechanism. The UPKI project, which is started by NII since 2005, tries to realize PKI-like trust chain in universities. PKI based user authentication is strong and reliable, but it is not flexible. The OpenID mechanism is developed for distributed ID management and web user authentication. Although OpenID is a flexible user management and authentication, OpenID doesn't guarantee trustiness of user identity. We consider a fusion of PKI and OpenID user authentication and ID federation mechanism for inter-domain membership services..
38. 釘崎裕司, 笠原 義晃, 堀 良彰, 櫻井 幸一, トラフィック解析に基づくボット検知手法, 情報処理学会研究報告コンピュータセキュリティ(CSEC), 10.1109/IPC.2007.91, 2007, 48, 303-+, 2007.10, 近年、ボットの感染が拡大しボットネットが形成され、社会問題となっている。ボットネットに対抗するためには、脆弱性を有するコンピュータの排除が理想である。そのため、ボットに感染しているコンピュータを検知し、注意を促す情報システムが重要である。そこでボットの動作の特徴を利用する検知手法が考えられている。本稿では IRC を利用したボットが、IRC サーバに接続する際の動作の特徴について調べた。実際に、IRC サーバでよく使われるポートのトラフィックを観測した。それにより、ボットが IRC サーバへの接続を拒否された際に、ある一定の時間間隔で再接続を試みることが確認できた。また、その時間間隔の分布を調べたところ、別の IP アドレスからの通信も類似の挙動を示していることを確認した。Recently, botnet becomes a social problem due to the expansion of bot infection. Ideally, all the vulnerable computers should be fortified to counteract botnet. To do that, it is important to implement an information system which detects bot-infected computers and alerts them. Therefore, there are various studies ongoing to detect existence of bots based on characteristics of bot behavior. In this paper, we focused on bots using IRC to communicate, and examined the behavior of such bots when they connected to an IRC server. We observed the actual traffic of some ports which were often used by the IRC protocol. As a result, we confirmed that a bot tried to reconnect to an IRC server at certain intervals when the server refused the connection from the bot. Moreover, we examined the distribution of the intervals and confirmed that the communication from other IP addresses showed similar behavior..
39. Wireless LAN roaming on ID-Federation environment -A case study for UPKI and eduroam in Kyushu University-
ID federation is one of important topic for information services. User authentication mechanism must be implemented for secure and personalized services. For inter-institutional services or nationwide services,it need inter-institutional user authentication,but it is very difficult to manage inter-institutional identity data base. ID federation is a solution for this problem. Federated institutions exchange user ID data each other. UPKI project were started by NII and some universities for Japan area ID federation. UPKI team joined eduroam for an application of ID federation. Eduroam is a RADIUS-based infrastructure that uses 802.1X security technology to allow for inter-institutional roaming. In this paper,the authors describe a case study for UPKI and eduroam in Kyushu University..
40. A Study of Identifier Naming Conventions Suitable for User Authentication
User authentication plays an important role in information services such as e-mail, PC login, and web databases. They need an identity database to realize user authentication, and identifiers to refer the database. Identifier is a special name to uniquely identify an entity (a user, in this case). We considered six possible identifier strings: serial number, random string, concatenated string, localization of global ID, derivation from real name, and self selection. We evaluated those types of identifiers with three aspects: manageability, usability, and robustness..
41. Exchange and Distribution of 3D Multimedia Data via Distributed 3D Shared Spaces
The authors regard text data, figures, audio, video data and 3D shape modeling data as 3D multimedia data. If such data are represented as visible and manually operable software components, it would be possible to exchange and distribute them among users through direct manipulations on a computer screen. The authors have been studying collaborative virtual environments and already proposed the mechanism to easily and rapidly build distributed 3D shared spaces by sharing software components that manage user-operation events through network. In this paper, the authors propose an exchange and distribution environment of 3D multimedia data via distributed 3D shared spaces using that software component. This paper also explains its realization mechanism and performance..
42. Management and Evaluation of a WWW Cache Server in Kyushu University
Recently, an explosive increase of population and traffic of the Internet became a severe problem. The WWW (World Wide Web) is the major source of the explosion, and now the traffic on the backbone of the Internet is mostly dominated by the WWW traffic. There are a lot of duplication in the WWW traffic, and WWW cache servers may decrease traffic on the backbone. This paper analyzes the log files of a WWW cache server which has been running for WWW users in Kyushu University since August 1996. It is acceptable that the hit rate on the WWW server is about 60%, but the severe increase of a response time is observed during busy time..
43. Task Dependence Net of Concurrent Ada Programs and Its Automatic Generation
プログラム中のある文の実行結果がその後実行される文の実行に影響を与えるように,文の間には従属関係が存在する.このような,プログラムの各文間に暗黙的に存在する従属関係のことをプログラム従属性と呼ぶ.プログラムの最適化,並列化,テスト,デバッグ,保守など,多くのソフトウェア開発活動の際にプログラム従属性を把握する必要がある.並行プログラムにおけるプログラム従属性は逐次プログラムのものより複雑なものとなるため,並行プログラムにおけるプログラム従属性の明示化は並行処理ソフトウェア開発の場面で重要な意味をもつ.本論文では,並行プログラミングが可能な実用言語Adaを具体例として,Ada並行プログラムのプログラム従属性を明示的に表現するモデルであるタスク従属ネットと,対象 Ada並行プログラムからそのタスク従属ネットを自動的に生成する方法について述べる.copyright(c)1996 IEICE許諾番号:08RB0010.
44. Constructing an Integrated Development Supporting Environment for Concurrent Software based on Program Dependence Theory
Explicit representations of program dependences have many applications in software development activities including program understanding, testing, debugging, maintenance, and complexity measurement. We proposed the Process Dependence Net to explicitly represent basic program dependences in concurrent programs. It can be used as a unified representation of program dependences in programs written in various sequential/concurrent programming languages. Based on this representation we can implement various tools to support software development, and construct an integrated development supporting environment for concurrent software as well as sequential software..
45. Constructing an integrated development supporting environment for concurrent software based on program dependence theory
Explicit representations of program dependences have many applications in software development activities including program understanding,testing,debugging,maintenance,and complexity measurement.We proposed the Process Dependence Net to explicitly represent basic program dependences in concurrent programs .It can be used as a unified representation of program dependences in programs written in various sequential, concurrent programming languages.Based on this representation we can implement various tools to support software development,and construct an integrated development supporting environment for concurrent software as well as sequential software..
46. Yoshiaki Kasahara, Jingde Cheng, Kazuo Ushijima, A Task Dependence Net Generator for Concurrent Ada Programs, Proceedings of the IPSJ & KISS Joint International Conference on Software Engineering '93, 315-322, 1993.11.
47. Jingde Cheng, Yoshiaki Kasahara, Masayuki Kamachi, Yoshinari Nomura, Kazuo Ushijima, Compiling programs to their dependence-based representations, Proceedings of the 1993 IEEE Region 10 Conference on Computer, Communication, Control and Power Engineering (TENCON '93), 374-377, 1993.10, A dependence-based program representation has many applications in various software development activities, and therefore, it can be used as a unified internal representation of various target programs in an integrated software development environment. This paper presents an overview of a general-purpose system to compile target programs written in various programming languages including C, Pascal, Ada, and Occam 2 into their dependence-based representations..
48. Some Considerations on Dependences in SR Concurrent Programs
SRはアリゾナ大学で開発された新しい並行処理プログラミング言語である。特徴としては次のようなものが挙げられる。・ハードウェアやOSにおける並列処理・並行処理の実現方法の違いを意識することなく並行プログラムを記述することができる。・プロセス間の同期・通信のための機構の種類が豊富である。本論文ではSRで記述されたプログラムにおける依存関係を考えるときに問題となる点について述べ、それに対する考察を行う。.
49. An integrated software development environment based on program dependences
プログラム中には本質的に必ず何らかの依存関係が存在する。プログラムの各文の間に存在する依存関係を把握することは、多くのソフトウェア開発活動に要求される。我々の研究室では、プログラム依存表現に基づいて、ソフトウェアの解析、テスト、デバッグ、保守、複雑さ計測を支援する統合的ソフトウェア開発支援環境の構築を目指して研究を進めている。本論文では、我々が構築している統合的開発環境の概要と特徴について述べる。.
50. Process Dependence Net Generator for Concurrent Programs
If the execution of a statement in a program affects the execution of another statement, there is a dependence relationship between the two statements. There are five types of basic program dependences in concurrent programs. Process Dependence Net(PDN) is an arc-classified digraph to explicitly represent the five types of basic program dependences in the concurrent programs. This paper describes algorithms to compute PDNs for a class of concurrent Ada programs, and shows the structure and implementation of PDN generator for concurrent Ada programs based on these algorithms. The paper also discusses some applications of the PDN generator to development of concurrent Ada programs..
51. A Task Dependence Net Generator for Concurrent Ada Programs
近年、マルチプロセッサシステムやLANの普及によって、並行処理プログラムが多方面で使用されるようになってきた。これに伴ない、並行処理プログラムのデバッギングの因難さを克服することが緊急の課題となっている。しかし、これまでの所並行処理プログラムに対する系統的なデバッギング手法は確立されていない。プログラムをテストしたり、実際に使用している時に、エラーが発生したり、仕様と異なるプログラムの振舞いが発見されることがある。プログラムのデバッギングとは、そのようにして発見された異常な振舞いの原因となるプログラム中の誤りの位置を特定し、修正する作業である。一般に誤りの位置を特定するために要する期間は実際のデバッギング期間の90%以上を占めると言われており、誤りの位置特定を支援することはデバッギング作業の負担を大きく軽減する。プログラムの誤りをみつける際に、異常な振舞いが検出された実行文に関係のある部分のみを調査することができれば、関係のない部分を調べないですむためデバッギング作業の期間を短縮できる。このためには、プログラム中の各実行文間にある依存関係を把握する必要がある。逐次処理プログラムについては、すでに依存関係を表現するモデルが提案されており、依存関係抽出を行なうツールについての研究が行なわれている。本研究では、我々が提案したプロセス依存ネットをAdaに適用したタスク依存ネットに基づいて、Ada並行処理プログラムにおける実行文間の依存関係を抽出し明示的に表現するツールを開発するとともに、このツールを用いたデバッギング作業の支援について考察する。以下、第2章でタスク依存ネットについて述べる。第3章ではタスク依存ネットの生成について述べる。第4章では生成したタスク依存ネットの応用について考察する。第5章で現在の状況と今後の課題について述べる。.
52. Jingde Cheng, Yoshiaki Kasahara, Kazuo Ushijima, A tasking deadlock detector for Ada programs, Proceedings The Fifteenth Annual International Computer Software and Applications Conference, 10.1109/cmpsac.1991.170152, 56-63, 1991.01.