Results 1  10
of
27
Accessing Nearby Copies of Replicated Objects in a Distributed Environment
"... Consider a set of shared objects in a distributed network, where several copies of each object may exist at any given time. To ensure both fast access to the objects as well as efficient utilization of network resources, it is desirable that each access request be satisfied by a copy "clos ..."
Abstract

Cited by 549 (8 self)
 Add to MetaCart
Consider a set of shared objects in a distributed network, where several copies of each object may exist at any given time. To ensure both fast access to the objects as well as efficient utilization of network resources, it is desirable that each access request be satisfied by a copy &quot;close &quot; to the requesting node. Unfortunately, it is not clear how to efficiently achieve this goal in a dynamic, distributed environment in which large numbers of objects are continuously being created, replicated, and destroyed. In this paper
Approximation Algorithms for Data Placement in Arbitrary Networks
, 2001
"... We study approximation algorithms for placing replicated data in arbitrary networks. Consider a network of nodes with individual storage capacities and a metric communication cost function, in which each node periodically issues a request for an object drawn from a collection of uniformlength objec ..."
Abstract

Cited by 84 (4 self)
 Add to MetaCart
(Show Context)
We study approximation algorithms for placing replicated data in arbitrary networks. Consider a network of nodes with individual storage capacities and a metric communication cost function, in which each node periodically issues a request for an object drawn from a collection of uniformlength objects. We consider the problem of placing copies of the objects among the nodes such that the average access cost is minimized. Our main result is a polynomialtime constantfactor approximation algorithm for this placement problem. Our algorithm is based on a careful rounding of a linear programming relaxation of the problem. We also show that the data placement problem is MAXSNPhard. We extend our approximation result to a generalization of the data placement problem that models additional costs such as the cost of realizing the placement. We also show that when object lengths are nonuniform, a constantfactor approximation is achievable if the capacity at each node in the approximate solution is allowed to exceed that in the optimal solution by the length of the largest object.
A Framework for Evaluating Replica Placement Algorithms
, 2002
"... This paper introduces a framework for evaluating replica placement algorithms (RPA) for content delivery networks (CDN) as well as RPAs from other fields that might be applicable to current or future CDNs. First, the framework classifies and qualitatively compares RPAs using a generic set of primiti ..."
Abstract

Cited by 43 (1 self)
 Add to MetaCart
This paper introduces a framework for evaluating replica placement algorithms (RPA) for content delivery networks (CDN) as well as RPAs from other fields that might be applicable to current or future CDNs. First, the framework classifies and qualitatively compares RPAs using a generic set of primitives that capture problem definitions and heuristics. Second, it provides estimates for the decision times of RPAs using an analytic model. To achieve accuracy, the model takes into account disk accesses and message sizes, in addition to computational complexity and message numbers that have been considered traditionally. Third, it uses the "goodness" of produced placements to compare RPAs even when they have different problem definitions. Based on these evaluations, we identify open issues and potential areas for future research.
Distributed transactional memory for metricspace networks
 In Proc. International Symposium on Distributed Computing (DISC 2005
, 2005
"... Abstract. Transactional Memory is a concurrent programming API in which concurrent threads synchronize via transactions (instead of locks). Although this model has mostly been studied in the context of multiprocessors, it has attractive features for distributed systems as well. In this paper, we con ..."
Abstract

Cited by 38 (0 self)
 Add to MetaCart
(Show Context)
Abstract. Transactional Memory is a concurrent programming API in which concurrent threads synchronize via transactions (instead of locks). Although this model has mostly been studied in the context of multiprocessors, it has attractive features for distributed systems as well. In this paper, we consider the problem of implementing transactional memory in a network of nodes where communication costs form a metric. The heart of our design is a new cachecoherence protocol, called the Ballistic protocol, for tracking and moving uptodate copies of cached objects. For constantdoubling metrics, a broad class encompassing both Euclidean spaces and growthrestricted networks, this protocol has stretch logarithmic in the diameter of the network. 1
Bandwidth Constrained Placement in a WAN
 IN PODC
, 2001
"... In this paper, we examine the bandwidthconstrained placement problem, focusing on tradeos appropriate for wide area network (WAN) environments. The goal is to place copies of objects at a collection of distributed caches to minimize expected access times from distributed clients to those objects ..."
Abstract

Cited by 30 (11 self)
 Add to MetaCart
In this paper, we examine the bandwidthconstrained placement problem, focusing on tradeos appropriate for wide area network (WAN) environments. The goal is to place copies of objects at a collection of distributed caches to minimize expected access times from distributed clients to those objects subject to a maximum bandwidth constraint at each cache. We develop a simple algorithm to generate a bandwidthconstrained placement by hierarchically rening an initial percache greedy placement. We prove that this hierarchical algorithm generates a placement whose expected access time is within a constant factor of the optimal placement's expected access time. We then proceed to extend this algorithm to compute close to optimal placement strategies for dynamic environments.
Oblivious routing on nodecapacitated and directed graphs
 IN PROCEEDINGS OF THE 16TH ACMSIAM SYMPOSIUM ON DISCRETE ALGORITHMS (SODA), 2005
, 2005
"... Oblivious routing algorithms for general undirected networks were introduced by Räcke [17], and this work has led to many subsequent improvements and applications. Comparatively little is known about oblivious routing in general directed networks, or even in undirected networks with node capacities. ..."
Abstract

Cited by 15 (7 self)
 Add to MetaCart
(Show Context)
Oblivious routing algorithms for general undirected networks were introduced by Räcke [17], and this work has led to many subsequent improvements and applications. Comparatively little is known about oblivious routing in general directed networks, or even in undirected networks with node capacities. We present the first nontrivial upper bounds for both these cases, providing algorithms for kcommodity oblivious routing problems with competitive ratio O (√ k log(n)) for undirected nodecapacitated graphs and O (√ k n 1/4 log(n)) for directed graphs. In the special case that all commodities have a common source or sink, our upper bound becomes O ( √ n log(n)) in both cases, matching the lower bound up to a factor of log(n). The lower bound (which first appeared in [6]) is obtained on a graph with very high degree. We show that in fact the degree of a graph is a crucial parameter for nodecapacitated oblivious routing in undirected graphs, by providing an O(∆ polylog(n))competitive oblivious routing scheme for graphs of degree ∆. For the directed case, however, we show that the lower bound of Ω (√ n) still holds in lowdegree graphs. Finally, we settle an open question about routing problems in which all commodities share a common source or sink. We show that even in this simplified scenario there are networks in which no oblivious routing algorithm can achieve a competitive ratio better than Ω(log n).
A Data Tracking Scheme for General Networks
, 2001
"... Consider an arbitrary distributed network in which large numbers of objects are continuously being created, replicated, and destroyed. A basic problem arising in such an environment is that of organizing a distributed directory service for locating object copies. In this paper, we present a new data ..."
Abstract

Cited by 14 (0 self)
 Add to MetaCart
Consider an arbitrary distributed network in which large numbers of objects are continuously being created, replicated, and destroyed. A basic problem arising in such an environment is that of organizing a distributed directory service for locating object copies. In this paper, we present a new data tracking scheme for locating nearby copies of objects in arbitrary distributed environments. Our tracking scheme supports ecient accesses to data objects while keeping the local memory overhead low. In particular, our tracking scheme achieves an expected polylog(n) approximation in the cost of any access operation, for an arbitrary network. The memory overhead incurred by our scheme is O(polylog(n)) times the maximum number of objects stored at any node, with high probability. We also show that our tracking scheme adapts well to dynamic changes in the network.
Atomic routing games on maximum congestion
 In Proceedings of the Second Internation Conference on Algorithmic Aspects of Information and Management
, 2006
"... We study atomic routing congestion games in which each player chooses a path in the network from its strategy set (a collection of paths) with the objective to minimize the maximum congestion along any edge on its selected path. The social cost is the global maximum congestion on any edge in the net ..."
Abstract

Cited by 13 (4 self)
 Add to MetaCart
(Show Context)
We study atomic routing congestion games in which each player chooses a path in the network from its strategy set (a collection of paths) with the objective to minimize the maximum congestion along any edge on its selected path. The social cost is the global maximum congestion on any edge in the network. We show that for arbitrary routing games, the price of stability is 1, and the price of anarchy, PoA, is bounded by κ − 1 ≤ PoA ≤ c(κ 2 + log 2 n), where κ is the length of the longest cycle in the network, n is the size of the network and c is a constant. Further, any best response dynamic converges to a Nash equilibrium. Our bounds show that for maximum congestion games, the topology of the network, in particular the length of cycles, plays an important role in determining the quality of the Nash equilibria. A fundamental issue in the management of large scale communication networks is to route the packet traffic so as to optimize the network performance. Our measure of network performance is the worst bottleneck (most used link) in the system. The model we use for network traffic is that of finite, unsplittable packets (atomic flow), and each packet’s path is controlled independently
Oblivious Routing on Geometric Networks
 Proceedings of the 17th Annual ACM Symposium on Parallelism in Algorithms and Architectures
, 2005
"... We study oblivious routing algorithms in which packet paths are constructed independently. ..."
Abstract

Cited by 11 (5 self)
 Add to MetaCart
(Show Context)
We study oblivious routing algorithms in which packet paths are constructed independently.