WEKO3
アイテム
フォールトトレラントグループ通信
https://ipsj.ixsq.nii.ac.jp/records/126774
https://ipsj.ixsq.nii.ac.jp/records/126774f8365caa-67ba-4e17-aa8b-d47947abe8a2
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
KJ00001340264.pdf (233.7 kB)
|
|
| Item type | National Convention(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 1994-09-20 | |||||
| タイトル | ||||||
| タイトル | フォールトトレラントグループ通信 | |||||
| タイトル | ||||||
| 言語 | en | |||||
| タイトル | Fault-Tolerant Group Communication | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||
| 資源タイプ | conference paper | |||||
| 著者所属 | ||||||
| 東京電機大学 | ||||||
| 著者所属 | ||||||
| 東京電機大学 | ||||||
| 著者所属 | ||||||
| 東京電機大学 | ||||||
| 著者所属 | ||||||
| 東京電機大学 | ||||||
| 著者所属(英) | ||||||
| en | ||||||
| Tokyo Denki University | ||||||
| 著者所属(英) | ||||||
| en | ||||||
| Tokyo Denki University | ||||||
| 著者所属(英) | ||||||
| en | ||||||
| Tokyo Denki University | ||||||
| 著者所属(英) | ||||||
| en | ||||||
| Tokyo Denki University | ||||||
| 論文抄録(英) | ||||||
| 内容記述タイプ | Other | |||||
| 内容記述 | In distributed applications, a group of processes have to be communicated. Multiple system processes have to support the application processes with the atomic and ordered delivery of messages by using the network. By using the group communication supported by the system processes in the group, the application processes can send messages to the others atomically and in some delivery order. The distributed systems suffer from failures, i.e. system process and network faults. In this paper, we discuss a fanlt-tolerace group communication in the presence of process faults including Byzantine fault. We assume that the network is dependable, i.e. messages are delivered to all the destinations atomically and in some specified order in the presence of message loss. From here, let processes mean system processes. Even if the processes in the group fault, the group has to support the application with the group communication. In order to support the fault-tolerance group communication in the presence of process faults,the processes are replicated. That is, each process is realized by a replica group which is a set of multiple replices of the process. Even if the replicas of the process fault, if at least one replica is operational, the process is considered to be operational. There are various replication strategies, pessive, active, and semi-active replications. In this paper, we present a hybrid replication method for replicating processes in orderto support fevlt-tolerancy.In section 2,we present a model of the system. In section 3,we discuss kinds of replications. In section4,we present how to construct a group of replicas. | |||||
| 書誌レコードID | ||||||
| 収録物識別子タイプ | NCID | |||||
| 収録物識別子 | AN00349328 | |||||
| 書誌情報 |
全国大会講演論文集 巻 第49回, 号 信頼性と安全性, p. 353-354, 発行日 1994-09-20 |
|||||
| 出版者 | ||||||
| 言語 | ja | |||||
| 出版者 | 情報処理学会 | |||||
