Results 1  10
of
20
Novel Architectures for P2P Applications: the ContinuousDiscrete Approach
 ACM TRANSACTIONS ON ALGORITHMS
, 2007
"... We propose a new approach for constructing P2P networks based on a dynamic decomposition of a continuous space into cells corresponding to processors. We demonstrate the power of these design rules by suggesting two new architectures, one for DHT (Distributed Hash Table) and the other for dynamic ex ..."
Abstract

Cited by 166 (8 self)
 Add to MetaCart
We propose a new approach for constructing P2P networks based on a dynamic decomposition of a continuous space into cells corresponding to processors. We demonstrate the power of these design rules by suggesting two new architectures, one for DHT (Distributed Hash Table) and the other for dynamic expander networks. The DHT network, which we call Distance Halving, allows logarithmic routing and load, while preserving constant degrees. Our second construction builds a network that is guaranteed to be an expander. The resulting topologies are simple to maintain and implement. Their simplicity makes it easy to modify and add protocols. We show it is possible to reduce the dilation and the load of the DHT with a small increase of the degree. We present a provably good protocol for relieving hot spots and a construction with high fault tolerance. Finally we show that, using our approach, it is possible to construct any family of constant degree graphs in a dynamic environment, though with worst parameters. Therefore we expect that more distributed data structures could be designed and implemented in a dynamic environment.
Envirostore: A cooperative storage system for disconnected operation in sensor networks
 In INFOCOM
, 2007
"... Abstract — This paper presents a new cooperative storage system for sensor networks geared for disconnected operation (where sensor nodes do not have a connected path to a basestation). The goal of the system is to maximize its data storage capacity by appropriately distributing storage utilization ..."
Abstract

Cited by 20 (1 self)
 Add to MetaCart
(Show Context)
Abstract — This paper presents a new cooperative storage system for sensor networks geared for disconnected operation (where sensor nodes do not have a connected path to a basestation). The goal of the system is to maximize its data storage capacity by appropriately distributing storage utilization and opportunistically offloading data to external devices when possible. The system is motivated by the observation that a large
A Scheme for Load Balancing in Heterogeneous Distributed Hash Tables
, 2005
"... We present a scheme for evenly partitioning the key space in distributed hash tables among the participating nodes. The scheme is based on the multiple random choices paradigm [3, 19], and handles both node joins and leaves. It achieves, with high probability, a ratio of at most 4 between the loads ..."
Abstract

Cited by 12 (0 self)
 Add to MetaCart
We present a scheme for evenly partitioning the key space in distributed hash tables among the participating nodes. The scheme is based on the multiple random choices paradigm [3, 19], and handles both node joins and leaves. It achieves, with high probability, a ratio of at most 4 between the loads of the most and least burdened nodes, in the face or arbitrary node arrivals and departures. Each join or leave operation incurs message cost that is, with high probability, O(log 2 n), where n is the number of nodes, and causes the relocation of keys from at most one node (for joins) or three nodes (for leaves). A version of our scheme is suitable for heterogeneous systems, where the capacities of nodes to serve keys can vary widely.
HistogramBased Global Load Balancing in Structured PeertoPeer Systems
"... Abstract—Over the past few years, peertopeer (P2P) systems have rapidly grown in popularity and have become a dominant means for sharing resources. In these systems, load balancing is a key challenge because nodes are often heterogeneous. While several loadbalancing schemes have been proposed in ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
(Show Context)
Abstract—Over the past few years, peertopeer (P2P) systems have rapidly grown in popularity and have become a dominant means for sharing resources. In these systems, load balancing is a key challenge because nodes are often heterogeneous. While several loadbalancing schemes have been proposed in the literature, these solutions are typically ad hoc, heuristic based, and localized. In this paper, we present a general framework, HiGLOB, for global load balancing in structured P2P systems. Each node in HiGLOB has two key components: 1) a histogram manager maintains a histogram that reflects a global view of the distribution of the load in the system, and 2) a loadbalancing manager that redistributes the load whenever the node becomes overloaded or underloaded. We exploit the routing metadata to partition the P2P network into nonoverlapping regions corresponding to the histogram buckets. We propose mechanisms to keep the cost of constructing and maintaining the histograms low. We further show that our scheme can control and bound the amount of load imbalance across the system. Finally, we demonstrate the effectiveness of HiGLOB by instantiating it over three existing structured P2P systems: Skip Graph, BATON, and Chord. Our experimental results indicate that our approach works well in practice. Index Terms—Peertopeer, framework, load balancing, histogram, DHT, overlay network. Ç
Dynamic load sharing in peertopeer systems: when some peers are more equal than others
 IEEE Internet Computing, Special Issue on Resource Allocation
"... AbstractIn the past few years, several DHTbased abstractions for peertopeer systems have been proposed. The main characteristic is to associate nodes (peers) with keys (objects) and to construct distributed routing structures to support an efficient location. These approaches address the load p ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
(Show Context)
AbstractIn the past few years, several DHTbased abstractions for peertopeer systems have been proposed. The main characteristic is to associate nodes (peers) with keys (objects) and to construct distributed routing structures to support an efficient location. These approaches address the load problem, and load balancing is achieved by moving the keys. However, the problem is still not properly covered. In this paper we present an analysis of structured peertopeer systems taking into consideration Zipflike requests distribution. Based on our analysis, we propose a novel approach for load balancing relying on object popularity. Our approach is based on routing table reorganization in order to balance the lookup traffic load. We have implemented this approach in a Pastrylike system. The obtained results demonstrate a better balance of load, which can lead to improved scalability and performance.
Random Choices for Churn Resilient Load Balancing in PeertoPeer Networks
"... Peertopeer (P2P) networks based on consistent hashing functions have an inherent load uneven distribution problem. Things are even worse in unstructured P2P systems. The objective of load balancing in P2P networks is to balance the workload of the network nodes in proportion to their capacity so a ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
(Show Context)
Peertopeer (P2P) networks based on consistent hashing functions have an inherent load uneven distribution problem. Things are even worse in unstructured P2P systems. The objective of load balancing in P2P networks is to balance the workload of the network nodes in proportion to their capacity so as to eliminate traffic bottleneck. It is challenging because of the dynamic nature of overlay networks and timevarying load characteristics. Random choices schemes can balance load effectively while incurring only a small overhead, making such schemes appealing for practical systems. Existing theoretical work analyzing properties of random choices algorithms can not be applied in the highly dynamic and heterogeneous P2P systems. In this paper, we characterize the behaviors of randomized search schemes in the general P2P environment. We extend the supermarket model by investigating the impact of node heterogeneity and churn to the load distribution in P2P networks. We prove that by using dway random choices schemes, the length of the longest queue in P2P systems with heterogeneous nodal capacity and node churn for d ≥ 2 is clog logn/logd + O(1) with high probability, where c is a constant. Keywords: Randomized probing; PeertoPeer networks; Load balancing; Heterogeneous and bounded
Ranged hash functions and the price of churn
"... Ranged hash functions generalize hash tables to the setting where hash buckets may come and go over time, a typical case in distributed settings where hash buckets may correspond to unreliable servers or network connections. Monotone ranged hash functions are a particular class of ranged hash functi ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
Ranged hash functions generalize hash tables to the setting where hash buckets may come and go over time, a typical case in distributed settings where hash buckets may correspond to unreliable servers or network connections. Monotone ranged hash functions are a particular class of ranged hash functions that minimize item reassignments in response to churn: changes in the set of available buckets. The canonical example of a monotone ranged hash function is the ringbased consistent hashing mechanism of Karger et al. [13]. These hash functions give a maximum load of Θ � n m log m � when n is the number of items and m is the number of buckets. The question of whether some better bound could be obtained using a more sophisticated hash function has remained open. We resolve this question by showing two lower bounds. First, the maximum load of any randomized monotone ranged hash function is Ω ( � n m ln m) when n = o(m log m). This bound covers almost all of the nontrivial case, because when n = Ω(m log m) simple random assignment matches the trivial lower bound of Ω(n/m). We give a matching (though impractical) upper bound that shows that our lower bound is tight over almost all of its range. Second, for randomized monotone ranged hash functions derived from metric spaces, there is a further tradeoff between the expansion factor of the metric and the load balance, which for the special case of growthrestricted metrics gives a bound of Ω � n m log m � , asymptotically equal to that of consistent hashing. These are the first known nontrivial lower bounds for ranged hash functions. They also explain why in ten years no better ranged hash functions have arisen to replace consistent hashing.
Efficient Load Balancing Techniques for Selforganizing Content Addressable Networks
"... Abstract — Balancing the load in a decentralized P2P system is a challenging problem due to the dynamic nature of such environment and the absence of global knowledge about the actual composition of the system.In this paper, we address the problem of load balancing in large scale and selforganizing ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Abstract — Balancing the load in a decentralized P2P system is a challenging problem due to the dynamic nature of such environment and the absence of global knowledge about the actual composition of the system.In this paper, we address the problem of load balancing in large scale and selforganizing P2P systems managing multidimensional data. We propose simple and efficient decentralized mechanisms to evenly distribute the data load among the participating nodes in Content Addressable Networks. The basic idea is to enable a new node that joins the system to share the load with a heavily loaded node which is already in the system, such that the load is still evenly distributed among all the participating nodes. In the multiple random choices method, the new node probes the load of some existing nodes selected uniformly at random, then chooses the heaviest node among
Passive/Active Load Balancing with Informed Node Placement in DHTs
"... Abstract. Distributed key/value stores are a basic building block for largescale Internet services. Support for range queries introduces new challenges to load balancing since both the key and workload distribution can be nonuniform. We build on previous work based on the power of choice to prese ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract. Distributed key/value stores are a basic building block for largescale Internet services. Support for range queries introduces new challenges to load balancing since both the key and workload distribution can be nonuniform. We build on previous work based on the power of choice to present algorithms suitable for active and passive load balancing that adapt to both the key and workload distribution. The algorithms are evaluated in a simulated environment, focusing on the impact of load balancing on scalability under normal conditions and in an overloaded system.