| 大会名称 |
| 2020年 総合大会 |
| 大会コ－ド |
| 2020G |
| 開催年 |
|
2020
|
| 発行日 |
|
2020-03-03
|
| セッション番号 |
|
BS-1
|
| セッション名 |
|
In-Network Intelligence for Design, Management, and Control of Future Networks and Services
|
| 講演日 |
|
2020/3/18
|
| 講演場所(会議室等) |
|
工学部 講義棟１F 104講義室
|
| 講演番号 |
|
BS-1-4
|
| タイトル |
|
Explained Decisions in Automated Network Management
|
| 著者名 |
|
○Pedro Martinez-Julia, Ved P. Kafle, Hitoshi Asaeda,
|
| キーワード |
|
Network, Resource, Control, Decision, Explanation
|
| 抄録 |
|
In this paper we discuss how the CBR methodology allows ARCA to provide explanations to its decisions, so that administrators are able to understand and validate the behavior of the automated management solution. Moreover, it exploits the metrics obtained from the VN together with the notifications provided by external event detectors [3]. We formalize external events and related information as causes and the ulterior performance measurements derived from them as effects. Both will form cases that are studied and learned by the intelligent solution to anticipate future undesired situations (undesired effects) and avoid them by taking the required countermeasures beforehand.
|
| 本文pdf |
|
PDF download
|
