Reduction of response time by data placement reflecting co-occurrence structures in structured overlay networks

Area of Science:

• Computer Science
• Network Engineering
• Distributed Systems

Background:

• Structured overlay networks face challenges in efficiently handling multi-queries.
• Optimizing data placement is crucial for reducing query latency.

Purpose of the Study:

• To develop a method for accelerating structured overlay network responses.
• To reduce the number of hops required for multi-queries through intelligent data placement.

Main Methods:

• Formulating data placement as an integer programming optimization problem.
• Proposing a greedy approach to efficiently solve the optimization problem.
• Utilizing redundant storage spaces for data item copying to reflect co-occurrence structures.

Main Results:

• Achieved a maximum reduction of approximately 30% in the average number of hops for multi-queries.
• Demonstrated that the reduction rate is dependent on the level of data co-occurrence.
• Evaluated the computational time reduction offered by the greedy approach.
• Confirmed no adverse impact on the load balancing of structured overlay networks.

Conclusions:

• The proposed data placement method effectively accelerates structured overlay network responses.
• The greedy approach provides an efficient solution for the data placement optimization problem.
• The method offers significant performance improvements without compromising network load balance.