@techreport{oai:ipsj.ixsq.nii.ac.jp:00103253,
 author = {宮座, 菜緒 and 桝田, 秀夫 and Miyaza, Nao and Masuda, Hideo},
 issue = {14},
 month = {Oct},
 note = {ネットワーク技術の発展により，組織のネットワークは仮想化などといった物理構成に縛られにくい柔軟性のあるサービス基盤を用い，様々なコンポーネントを連携させながら我々の生活に必要不可欠なサービスを提供している．このようなサービスが広まるとともに，利用者にサービスを継続して提供する重要性も高まり，常に安定した運用管理が求められているが，サービス基盤内の構成が複雑であったり，監視対象や監視機器の種類の増加に伴い，互いのコンポーネントの依存関係や関連性の把握が難しく，管理者が障害の原因を突き止めることが困難となっている．そこで本稿では，複数の機器の監視項目に対して，グラフの表示方法や時系列の類似性に着目し，障害の引き金を探し当てることを支援するシステムの検討を行ったので，現状を報告し考察する．, With the development of network technology, the organizations use the network which is based on flexible service infrastructure such as virtualization, and provide us with essential services which are linked up with various components. With the spread of such services, organizations should be able to continuously provide the services to the users. However, because of the complicated constitution of the service infrastructure and the increase of the types of surveillance devices, the organization's network managers can't understand the relevance of various monitoring components. Therefore it is difficult for the organization's network managers to find the cause of the fault occurrences. This paper explain the consideration of a prototype to support them to find the cause of the fault occurrences by examining the expression method of graph and the similarity of time series regarding the various components monitored by the systems.},
 title = {監視項目の時系列類似度に基づく障害把握を支援するシステムの検討},
 year = {2014}
}