| Presentation | 2013-01-17 Improvement of Inter-site Replicate Transaction for High Load Object Storage Kazuyoshi JINNAI, Minoru KATO, Shinji KONO, Mitsutomo IMAZAKI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In the present days of cloud computing, users can handle their own data anytime anywhere from cloud storage provided by cloud resource providers using mobile client. Storage amount of the cloud resource providers who provide such mass services continues to grow. These cloud resource providers are now beginning to use distributed object storage instead of traditional mainstream Storage. Area Network (Using big virtual volumes created by many cheap PC servers on networks). Since the distributed object storage is easy to maintain performance under high load dispersing huge amount and has no constraints of geographically distributed locations, some enterprise systems are now beginning to consider adopting this for disaster recovery. However, the object storage needs complex synchronous method for replication between clusters because it has unique structures replacing their own files. In case of a disaster, business continuity of backup cluster gets an impact from slow replicating transactions of main cluster particularly under high load. In case the main cluster gets shut down, there remains high load of un-transferred data, backup cluster would become useless, and too much data would be lost. This paper discusses ways to shorten the time to replicate transactions with simple synchronous method using SWIFT, a component of Open Stack, and proposes a way to improve, based on the findings from the actual test conducted as its verification. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Open Stack / Swift / Distributed Object Storage / Distributed File System / Disaster Recovery |
| Paper # | ICM2012-44,LOIS2012-54 |
| Date of Issue |
| Conference Information | |
| Committee | LOIS |
|---|---|
| Conference Date | 2013/1/10(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 | Life Intelligence and Office Information Systems (LOIS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Improvement of Inter-site Replicate Transaction for High Load Object Storage |
| Sub Title (in English) | |
| Keyword(1) | Open Stack |
| Keyword(2) | Swift |
| Keyword(3) | Distributed Object Storage |
| Keyword(4) | Distributed File System |
| Keyword(5) | Disaster Recovery |
| 1st Author's Name | Kazuyoshi JINNAI |
| 1st Author's Affiliation | NTT COMWARE CORPORATION() |
| 2nd Author's Name | Minoru KATO |
| 2nd Author's Affiliation | NTT COMWARE CORPORATION |
| 3rd Author's Name | Shinji KONO |
| 3rd Author's Affiliation | NTT INTERNET CORPORATION |
| 4th Author's Name | Mitsutomo IMAZAKI |
| 4th Author's Affiliation | NTT COMWARE CORPORATION |
| Date | 2013-01-17 |
| Paper # | ICM2012-44,LOIS2012-54 |
| Volume (vol) | vol.112 |
| Number (no) | 379 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |