| Presentation | 2013-06-17 Automatic configuration for cloud computing infrastructure using decision tree Tetsuya UCHIUMI, Shinya KITAJIMA, Shinji KIKUCHI, Yasuhide MATSUMOTO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recently, cloud venders are required to construct cloud computing infrastructures quickly to keep up with the increase in the users' demands. However, since large-scale cloud datacenters have a great number of configuration parameters (parameters of OS or middleware, e.g., IP address), it is difficult for administrators to configure these parameters correctly. Some approaches for automatic parameter configuration have been proposed so far. These methods are mainly focusing on identifying common configurations in existing infrastructures as "design patterns" and apply these patterns for the development of another cloud infrastructure. However, these methods still require manual configuration for the parts which do not have common patterns. Therefore, a method to reduce manual configuration is demanded, because the manual configuration is one of the most dominant causes of misconfigurations. Based on this background, we propose an automatic configuration method which only requires minimum manual inputs by generating parameter configuration patterns. In our approach, we identify the linear relation between the generations (time when they were constructed) of existing infrastructures and the values assigned to the parameters in the infrastructures. By constructing the configuration patterns from the relationship, we reduce the manual configuration. We evaluated our method in actual cloud infrastructures and show the result that the proposed approach could configure 91.3% parameters of new infrastructure automatically. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Cloud datacenter / parameter generation / clustering / decision tree / machine learning / parameter estimation |
| Paper # | SC2013-7 |
| Date of Issue |
| Conference Information | |
| Committee | SC |
|---|---|
| Conference Date | 2013/6/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Services Computing (SC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Automatic configuration for cloud computing infrastructure using decision tree |
| Sub Title (in English) | |
| Keyword(1) | Cloud datacenter |
| Keyword(2) | parameter generation |
| Keyword(3) | clustering |
| Keyword(4) | decision tree |
| Keyword(5) | machine learning |
| Keyword(6) | parameter estimation |
| 1st Author's Name | Tetsuya UCHIUMI |
| 1st Author's Affiliation | System Software Laboratories, FUJITSU LABORATORIES LTD.() |
| 2nd Author's Name | Shinya KITAJIMA |
| 2nd Author's Affiliation | System Software Laboratories, FUJITSU LABORATORIES LTD. |
| 3rd Author's Name | Shinji KIKUCHI |
| 3rd Author's Affiliation | System Software Laboratories, FUJITSU LABORATORIES LTD. |
| 4th Author's Name | Yasuhide MATSUMOTO |
| 4th Author's Affiliation | System Software Laboratories, FUJITSU LABORATORIES LTD. |
| Date | 2013-06-17 |
| Paper # | SC2013-7 |
| Volume (vol) | vol.113 |
| Number (no) | 86 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |