Enhanced model to minimize future downtime : case study of Malaysia cloud providers towards near-zero downtime
In providing tremendous access to data and computing power of thousands of commodity servers, large-scale cloud systems must address a new challenge: they must detect and recover from a growing number of failures, in both hardware and software components. The growing complexity of technology scaling...
| 要約: | In providing tremendous access to data and computing power of thousands of commodity servers, large-scale cloud systems must address a new challenge: they must detect and recover from a growing number of failures, in both hardware and software components. The growing complexity of technology scaling, manufacturing, design logic, usage, and operating environment increases the occurrence of failures. Unfortunately, downtime handling has proven to be problematic in today’s cloud systems. The downtime recovery path is often complex, under-specified, and tested less frequently than the normal path. As indicated by recent cloud outage incidents, existing large-scale cloud systems are still fragile and error-prone. The purpose of this study to identify the issues causing cloud downtime, to investigate the recovery ability
of the database during cloud downtime and to propose enhanced model that can be used to minimize the future downtime. |
|---|