a symmetric load balancing algorithm
TRANSCRIPT
A Symmetric Load Balancing Algorithm with PerformanceA Symmetric Load Balancing Algorithm with Performance
Guarantees for Distributed Hash TablesGuarantees for Distributed Hash Tables
Abstract—Abstract—
Peers participating in a distributed hash table (DHT) may host different numbers ofPeers participating in a distributed hash table (DHT) may host different numbers of
virtual servers and are enabled to balance their loads in the reallocation of virtualvirtual servers and are enabled to balance their loads in the reallocation of virtual
servers. Most decentralized load balance algorithms designed for DHTs based onservers. Most decentralized load balance algorithms designed for DHTs based on
virtual servers require the participating peers to be asymmetric, where some servevirtual servers require the participating peers to be asymmetric, where some serve
as the rendezvous nodes to pair virtual servers and participating peers, therebyas the rendezvous nodes to pair virtual servers and participating peers, thereby
introducing another load imbalance problem.While state-of-the-art studies intendintroducing another load imbalance problem.While state-of-the-art studies intend
to present symmetric load balancing algorithms, they introduce significantto present symmetric load balancing algorithms, they introduce significant
algorithmic overheads and guarantee no rigorous performance metrics. In thisalgorithmic overheads and guarantee no rigorous performance metrics. In this
paper, a novel symmetric load balancing algorithm for DHTs is presented bypaper, a novel symmetric load balancing algorithm for DHTs is presented by
having the participating peers approximate the system state with histograms andhaving the participating peers approximate the system state with histograms and
cooperatively implement a global index. Each peer independently reallocates in ourcooperatively implement a global index. Each peer independently reallocates in our
proposal its locally hosted virtual servers by publishing and inquiring the globalproposal its locally hosted virtual servers by publishing and inquiring the global
index based on their histograms. Unlike competitive algorithms, our proposalindex based on their histograms. Unlike competitive algorithms, our proposal
exhibits analytical performance guarantees in terms of the load balance factor andexhibits analytical performance guarantees in terms of the load balance factor and
the algorithmic convergence rate, and introduces no load imbalance problem due tothe algorithmic convergence rate, and introduces no load imbalance problem due to
the algorithmic workload.the algorithmic workload.
Existing system :Existing system :
To tackle the load imbalance problem in DHTs, prior proposals have presentedTo tackle the load imbalance problem in DHTs, prior proposals have presented
centralized algorithms that rely on a few rendezvous nodes to balance the loads ofcentralized algorithms that rely on a few rendezvous nodes to balance the loads of
peers in a DHT.peers in a DHT.
In contrast to the centralized approach, some studies, e.g., [9], [10], suggest In contrast to the centralized approach, some studies, e.g., [9], [10], suggest
organizing rendezvous nodes in a hierarchical manner. Virtual servers are firstorganizing rendezvous nodes in a hierarchical manner. Virtual servers are first
matched with peers through the rendezvous nodes in the lower layer of thematched with peers through the rendezvous nodes in the lower layer of the
hierarchy; for unpaired virtual servers, the rendezvous peers relay them to thehierarchy; for unpaired virtual servers, the rendezvous peers relay them to the
rendezvous in the upper layer to seek reallocation. This process repeats iterativelyrendezvous in the upper layer to seek reallocation. This process repeats iteratively
until an unpaired virtual server reaches a rendezvous in the highest layer.until an unpaired virtual server reaches a rendezvous in the highest layer.
Demerits :Demerits :
considering large-scale and dynamic DHT networks, the centralized algorithmsconsidering large-scale and dynamic DHT networks, the centralized algorithms
may introduce the performance bottleneck and the single point of failure.may introduce the performance bottleneck and the single point of failure.
Consequently, the rendezvous nodes may experience skewed workloads,Consequently, the rendezvous nodes may experience skewed workloads,
introducing another load imbalance problem. They may also become theintroducing another load imbalance problem. They may also become the
performance bottleneck and the single point of failure. Moreover, the hierarchicalperformance bottleneck and the single point of failure. Moreover, the hierarchical
networks (e.g., the tree-shaped network in ) facilitating the load balancingnetworks (e.g., the tree-shaped network in ) facilitating the load balancing
algorithms are prone to node/communication failure, thus demanding sophisticatedalgorithms are prone to node/communication failure, thus demanding sophisticated
maintenance for the networks.maintenance for the networks.
Proposed system :Proposed system :
In this paper, we present a fully decentralized load balancing algorithm for DHTs.In this paper, we present a fully decentralized load balancing algorithm for DHTs.
Our proposal, is essentially different from the previous rendezvous-based solutionsOur proposal, is essentially different from the previous rendezvous-based solutions
in that each peer estimates and represents the system state with histograms. Basedin that each peer estimates and represents the system state with histograms. Based
on the histograms, each peer reallocates its load independently without introducingon the histograms, each peer reallocates its load independently without introducing
an asymmetric rendezvous in collecting the system state as well as matchingan asymmetric rendezvous in collecting the system state as well as matching
virtual servers and participating peers. virtual servers and participating peers.