Abstract:
In cloud-deployed distributed data stores, server performance fluctuates. Thus, the reading request replica directly impacts response latency. Key-value stores are latency-critical, making replica selection difficult. Small data stores require clients to choose replicas. Even the best replica selection algorithm (C3) can be made faster.
Compared to C3, Oracle reduces 99th latency by 34-60%. NetRS addresses the fundamental factors that make replica selection algorithms ineffective. NetRS enables distributed data store in-network replica selection. It selects replicas using programmable switches and network accelerators. NetRS supports multiple replica selection algorithms and fits modern data center networks.
NetRS reduces mean latency by 50.3 percent and 99th latency by 56.3 percent compared to clients selecting replicas for requests, according to our extensive evaluations. Even during workload changes, network device failures, and network congestion, NetRS could reduce response latency.
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
