Hash-based proximity clustering for efficient load balancing in heterogeneous DHT networks

Shen, Haiying; Xu, Cheng‐Zhong

doi:10.1016/j.jpdc.2007.10.005

Cited by 36 publications

(27 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Section 6.1 presented a way of clustering in Cycloid. Readers are referred to [16] for its generalization to other DHT networks.…”

Section: Discussionmentioning

confidence: 99%

Locality-Aware and Churn-Resilient Load-Balancing Algorithms in Structured Peer-to-Peer Networks

Shen

2007

IEEE Trans. Parallel Distrib. Syst.

Self Cite

View full text Add to dashboard Cite

Abstract-Structured peer-to-peer overlay networks, like distributed hash tables (DHTs), map data items to the network based on a consistent hashing function. Such mapping for data distribution has an inherent load balance problem. Data redistribution algorithms based on randomized matching of heavily loaded nodes with light ones can deal with the dynamics of DHTs. However, they are unable to consider the proximity of the nodes simultaneously. There are other methods that rely on auxiliary networks to facilitate localityaware load redistribution. Due to the cost of network construction and maintenance, the locality-aware algorithms can hardly work for DHTs with churn. This paper presents a locality-aware randomized load-balancing algorithm to deal with both the proximity and network churn at the same time. We introduce a factor of randomness in the probing of lightly loaded nodes in a range of proximity. We further improve the efficiency by allowing the probing of multiple candidates (d-way) at a time. Simulation results show the superiority of the locality-aware two-way randomized algorithm in comparison with other random or locality-aware algorithms. In DHTs with churn, it performs no worse than the best churn-resilient algorithm. It takes advantage of node capacity heterogeneity and achieves good load balance effectively even in a skewed distribution of items.

show abstract

“…Section 6.1 presented a way of clustering in Cycloid. Readers are referred to [16] for its generalization to other DHT networks.…”

Section: Discussionmentioning

confidence: 99%

Locality-Aware and Churn-Resilient Load-Balancing Algorithms in Structured Peer-to-Peer Networks

Shen

2007

IEEE Trans. Parallel Distrib. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The second method is to measure the service load, including, e.g. the number of service requests, routed messages, the size of stored items, maintenance signalling or the total volume of the traffic (Shen & Xu 2008, Bianchi et al 2006. The relation between the load types can be described as follows; service load generates hardware load and the impact depends on the hardware capabilities of the device in question.…”

Section: Basic Load Balancing Conceptsmentioning

confidence: 99%

“…This is done in order to maximize performance (Ferrari & Zhou 1987) and efficiency (Zhu & Hu 2004 of a distributed computer system by harnessing all available resources in the system, and to eliminate traffic bottlenecks (Shen & Xu 2008). The need for load balancing grows along with the overall load of a distributed computer system, since the probability of nodes getting overloaded grows.…”

Section: Basic Load Balancing Conceptsmentioning

confidence: 99%

Energy-efficient peer-to-peer networking for constrained-capacity mobile environments

Harjula

Ojala

Ylianttila

2017

2017 IFIP/IEEE Symposium on Integrated Network and Service Management (IM)

View full text Add to dashboard Cite

Energy efficiency is a powerful measure for promoting sustainability in technological evolution and ensuring feasible battery life of end-user devices in mobile computing. Peer-to-peer technology provides decentralized and self-organizing architecture for distributing content between devices in networks that scale up almost infinitely. However, peer-to-peer networking may require lots of resources from peer nodes, which in turn may lead to increased energy consumption on mobile devices. For this reason, peer-to-peer networking has so far been considered unfeasible for mobile environment.This thesis makes several contributions towards enabling energy-aware peer-to-peer networking in mobile environments. First, an empirical study is conducted to understand the energy consumption characteristics of radio interfaces and typical composition of traffic in structured peer-to-peer networks. This is done in order to identify the most essential obstacles for utilizing peer-to-peer technology in mobile environments. Second, the e-Aware model for estimating the energy consumption of a mobile device is developed and empirically verified to achieve 3-21% error in comparison to real-life measurements. Third, the e-Mon model for the energy-aware load monitoring of peer nodes is developed and demonstrated to improve the battery life of mobile peer nodes up to 470%. Fourth, the ADHT concept of mobile agent based virtual peers is proposed for sharing the peer responsibilities between peer nodes in a subnet so that they can participate in a peer-to-peer overlay without compromising their battery life.The results give valuable insight into implementing energy-efficient peer-to-peer systems in mobile environments. The e-Aware energy consumption model accelerates the development of energy-efficient networking solutions by reducing the need for time-consuming iterations between system development and evaluations with real-life networks and devices. The e-Mon load monitoring model facilitates the participation of battery-powered devices in peer-to-peer and other distributed networks by enabling energy-aware load balancing where energy-critical mobile nodes carry less load than other nodes. The ADHT facilitates the participation of constrained-capacity wireless devices, such as machine-to-machine nodes, in a peer-to-peer network by allowing them to sleep for most of their time.Keywords: battery life, constrained-capacity devices, energy-efficiency, green computing, internet of things, machine-to-machine networks, mobile networks, peer-topeer networks Harjula, Erkki, Energiatehokkaat vertaisverkkoratkaisut rajoitetun kapasiteetin mobiiliympäristöissä.Oulun yliopiston tutkijakoulu; Oulun yliopisto, Tieto-ja sähkötekniikan tiedekunta; Centre for Wireless Communications; Infotech Oulu Acta Univ. Oul. C 573, 2016 Oulun yliopisto, PL 8000, 90014 Oulun yliopisto TiivistelmäEnergiatehokkuus on kustannustehokas tapa vähentää päätelaitteiden käytön aiheuttamia kasvihuonepäästöjä sekä parantaa niiden akunkestoa. Vertaisverkkoteknologia ta...

show abstract

“…In the case of peer-to-peer systems, data migration can be further classified in the node virtualization [9], [10], [11], [12], [13], [14] and one ID per server [1], [14], [15], [16], [17], [18], [19], [20], [21] strategies. In the former, every actual server can host a number of virtual servers which are responsible for small disjoint ranges of its items, and balancing is performed by moving virtual servers between actual ones.…”

Section: Related Workmentioning

confidence: 99%

“…In current bibliography, a variety of methods exists focusing on achieving efficient load-balancing for such structures, whether they utilize the notion of "virtual servers" [9], [10], [11], [12], [13], [14] or not [1], [14], [15], [16], [17], [18], [19], [20], [21]. Yet, they can be categorized in two general strategies: Node Migration and Neighbor Item Exchange.…”

mentioning

confidence: 99%

Fast and Cost-Effective Online Load-Balancing in Distributed Range-Queriable Systems

Konstantinou

Tsoumakos

Koziris

2011

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Abstract-Distributed systems such as Peer-to-Peer overlays have been shown to efficiently support the processing of range queries over large numbers of participating hosts. In such systems, uneven load allocation has to be effectively tackled in order to minimize overloaded peers and optimize their performance. In this work, we detect the two basic methodologies used to achieve load-balancing: Iterative key redistribution between neighbors and node migration. We identify these two key mechanisms and describe their relative advantages and disadvantages. Based on this analysis, we propose NIXMIG, a hybrid method that adaptively utilizes these two extremes to achieve both fast and cost-effective load-balancing in distributed systems that support range queries. We theoretically prove its convergence and as a case study, we offer an implementation on top of a Skip Graph, where we thoroughly validate our findings in a variety of static, dynamic and realistic workloads. We compare NIXMIG with an existing load-balancing algorithm proposed by Karger and Ruhl [1] and our experimental analysis shows that, NIXMIG can be as much as three times faster, requiring only one sixth and one third of message and item exchanges, respectively, to bring the system to a balanced state.Index Terms-Peer-to-peer systems, load-balancing, range queries.

show abstract

Hash-based proximity clustering for efficient load balancing in heterogeneous DHT networks

Cited by 36 publications

References 43 publications

Locality-Aware and Churn-Resilient Load-Balancing Algorithms in Structured Peer-to-Peer Networks

Locality-Aware and Churn-Resilient Load-Balancing Algorithms in Structured Peer-to-Peer Networks

Energy-efficient peer-to-peer networking for constrained-capacity mobile environments

Fast and Cost-Effective Online Load-Balancing in Distributed Range-Queriable Systems

Contact Info

Product

Resources

About