Abstract:
Cloud environments prioritize VM data and service reliability. Continuous Snapshots are preferred for long-running systems because they reduce data loss in failures. Existing methods have long VM downtime, snapshot intervals, and performance loss.
iConSnap takes fine-grained continuous snapshots of virtual machines without affecting performance. First, iConSnap uses copy-on-write (COW) to save memory pages on-demand, reducing VM downtime to 200 milliseconds.
Second, we extend COW and propose a lazily incremental approach to save the delta data between two snapshots only once, greatly reducing snapshot duration and data. Third, a scheduling mechanism reduces the VM performance penalty.
Finally, we use compression and time-aware multi-granularity reclamation to reduce storage costs without sacrificing performance or availability. We test iConSnap on QEMU/KVM. iConSnap outperforms other methods in VM downtime, snapshot duration, storage costs, and performance.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
