Load Balance with Imperfect Information in Structured Peer-to-Peer Systems

Hsiao, Hung‐Chang; Liao, Hao; Chen, Ssu-Ta; Huang, Kuo-Chan

doi:10.1109/tpds.2010.105

Cited by 34 publications

(24 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Therefore, Hsiao [18] proposed a decentralized algorithm for load-distribution. Although the proposed method uses a dedicated server, it would be possible to adopt a decentralized algorithm and remove the dedicated server.…”

Section: Related Workmentioning

confidence: 99%

Dynamic Load-Distribution Method of uTupleSpace Data-Sharing Mechanism for Ubiquitous Data

Arakawa

Kashiwagi

Nakamura

et al. 2014

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThe number of networked devices of sensors and actuators continues to increase. We are developing a data-sharing mechanism called uTupleSpace as middleware for storing and retrieving ubiquitous data that are input or output by such devices. uTupleSpace enables flexible retrieval of sensor data and flexible control of actuator devices, and it simplifies the development of various applications. Though uTupleSpace requires scalability against increasing amounts of ubiquitous data, traditional load-distribution methods using a distributed hash table (DHT) are unsuitable for our case because of the ununiformity of the data. Data are nonuniformly generated at some particular times, in some particular positions, and by some particular devices, and their hash values focus on some particular values. This feature makes it difficult for the traditional methods to sufficiently distribute the load by using the hash values. Therefore, we propose a new load-distribution method using a DHT called the dynamichelp method. The proposed method enables one or more peers to handle loads related to the same hash value redundantly. This makes it possible to handle the large load related to one hash value by distributing the load among peers. Moreover, the proposed method reduces the load caused by dynamic load-redistribution. Evaluation experiments showed that the proposed method achieved sufficient load-distribution even when the load was concentrated on one hash value with low overhead. We also confirmed that the proposed method enabled uTupleSpace to accommodate the increasing load with simple operational rules stably and with economic efficiency.

show abstract

Section: Related Workmentioning

confidence: 99%

Dynamic Load-Distribution Method of uTupleSpace Data-Sharing Mechanism for Ubiquitous Data

Arakawa

Kashiwagi

Nakamura

et al. 2014

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Previous studies have assumed only one bottleneck resource in the system (Wang et al, 2004;Zhu and Hu, 2005). The study carried out in (Hsiao et al, 2011) also assumed only one bottleneck resource in the system, but it failed to assume a particular resource constraint. We have considered a specific resource as a constraint, namely the number of active TCP ports.…”

Section: Jcsmentioning

confidence: 99%

“…In addition, they did not try to balance the load over the whole system so that every peer reached its threshold level. Hsiao et al (2011) tried to reduce the load imbalance factor by transferring virtual servers from overloaded to underloaded peers. They tried to reduce the movement cost, but did not try to minimize the load imbalance factor.…”

Section: Related Workmentioning

confidence: 99%

“…They tried to reduce the movement cost, but did not try to minimize the load imbalance factor. Their method used the virtual server concept for load balancing, by assigning load to each object entering the system (Hsiao et al, 2011;Godfrey et al, 2004). However, their method assumed that there was only one bottleneck resource in the system and left the multiple resources as future work.…”

Section: Related Workmentioning

confidence: 99%

“…In this case, each peer hosts a number of virtual servers. To balance the load of the system, the virtual servers are moved from an overloaded to an underloaded peer (Rao et al, 2003;Li and Shao, 2011;Wang and Vanninen, 2006;Zou et al, 2002;Hsiao et al, 2011;Godfrey et al, 2004). In earlier work, the virtual servers were transferred from overloaded to underloaded peers in only one direction (Hsiao et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Using Mobile Agents for Load Balancing in Peer-to-Peer Systems Hosting Virtual Servers

Gopalakrishnan¹,

Saibharath²,

Uthariaraj³

2014

Journal of Computer Science

View full text Add to dashboard Cite

This study proposes a novel load-balancing algorithm for managing virtual servers in a Peer-To-Peer (P2P) system through mobile agents. The proposed algorithm is implemented in a fully decentralized manner for a structured P2P system. It uses mobile agents and is independent of the geometry of the P2P auxiliary networks. The load-balancing algorithm effectively reduces the load imbalance of the system using the load per unit capacity derived by the mobile agents. A unique feature of the proposed algorithm is the mutual swapping of virtual servers between overloaded and underloaded peers to efficiently use the available resources. The proposed solution has been verified in a P2P environment consisting of peers and embedded glassfish server instances, created dynamically to act as virtual servers.

show abstract

Cloud Computing, Internet of Things (IoT), Edge Computing, and Big Data Infrastructure

Hsiao¹,

Hung²,

Chen³

et al. 2021

Industry 4.1

View full text Add to dashboard Cite

Load Balance with Imperfect Information in Structured Peer-to-Peer Systems

Cited by 34 publications

References 18 publications

Dynamic Load-Distribution Method of uTupleSpace Data-Sharing Mechanism for Ubiquitous Data

Dynamic Load-Distribution Method of uTupleSpace Data-Sharing Mechanism for Ubiquitous Data

Using Mobile Agents for Load Balancing in Peer-to-Peer Systems Hosting Virtual Servers

Cloud Computing, Internet of Things (IoT), Edge Computing, and Big Data Infrastructure

Contact Info

Product

Resources

About