| Presentation | 2018-11-16 Trend analysis method of operational data based on component decomposition using Non-negative Matrix Factorization Yuji Saitoh, Tetsuya Uchiumi, Yukihiro Watanabe, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In the ICT system, Understanding the trend of operational data such as the CPU usage rate is useful for system improvement. In this case, it is more useful if it is possible to understand the trend in not only the operational data of each device but also the business process executed on the device. However, in general, the measured operational data is often for each device, so it cannot be analyzed for each business process. In the proposed method, using Non-negative Matrix Factorization, which is used for decomposing sound signals into signals of musical instruments and vocals, the operational data acquired for each device is decomposed for each component with different business process characteristics, and the long-term trend for each component is estimated. As a result, it is possible to detect trends such as a long-term increase in load of a specific business process that could not be detected in the past. Experimental results for artificial data and actual data show that the operational data can be decomposed for each business process by the proposed method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Non-negative Matrix Factorization / component analysis / capacity management / resource provisioning / IT operations management |
| Paper # | ICM2018-32 |
| Date of Issue | 2018-11-08 (ICM) |
| Conference Information | |
| Committee | CQ / ICM / NS |
|---|---|
| Conference Date | 2018/11/15(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Takanori Hayashi(Hiroshima Inst. of Tech.) / Kiyohito Yoshihara(KDDI Research) / Yoshikatsu Okazaki(NTT) |
| Vice Chair | Hideyuki Shimonishi(NEC) / Jun Okamoto(NTT) / Yoichi Yamashita(NTT-N) / Takumi Miyoshi(Shibaura Inst. of Tech.) / Akihiro Nakao(Univ. of Tokyo) |
| Secretary | Hideyuki Shimonishi(NTT) / Jun Okamoto(Nippon Inst. of Tech.) / Yoichi Yamashita(KDDI Research) / Takumi Miyoshi(NTT) / Akihiro Nakao(NTT) |
| Assistant | Chikara Sasaki(KDDI Research) / Yoshiaki Nishikawa(NEC) / Ryo Yamamoto(UEC) / Yunchen Zhu(Hitachi) / Kenichi Kashibuchi(NTT) |
| Paper Information | |
| Registration To | Technical Committee on Communication Quality / Technical Committee on Information and Communication Management / Technical Committee on Network Systems |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Trend analysis method of operational data based on component decomposition using Non-negative Matrix Factorization |
| Sub Title (in English) | |
| Keyword(1) | Non-negative Matrix Factorization |
| Keyword(2) | component analysis |
| Keyword(3) | capacity management |
| Keyword(4) | resource provisioning |
| Keyword(5) | IT operations management |
| 1st Author's Name | Yuji Saitoh |
| 1st Author's Affiliation | FUJITSU LABORATORIES LTD.(FUJITSU LAB.) |
| 2nd Author's Name | Tetsuya Uchiumi |
| 2nd Author's Affiliation | FUJITSU LABORATORIES LTD.(FUJITSU LAB.) |
| 3rd Author's Name | Yukihiro Watanabe |
| 3rd Author's Affiliation | FUJITSU LABORATORIES LTD.(FUJITSU LAB.) |
| Date | 2018-11-16 |
| Paper # | ICM2018-32 |
| Volume (vol) | vol.118 |
| Number (no) | ICM-303 |
| Page | pp.pp.45-50(ICM), |
| #Pages | 6 |
| Date of Issue | 2018-11-08 (ICM) |