L.W. Dowdy and D.V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):287-313, June 1982.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....are connected to the backbone via direct links. The major requirement is that the total access cost for the users is within a specified bound, whereas the construction cost of the backbone network should be minimized. Similar problems arise in the efficient maintenance of distributed databases [28, 6, 5, 20, 9]. Other applications of BCCMED include location theory and manufacturing logistics (see [25, 27] and the references cited therein) In this paper, we study the complexity and approximability of BCCMED. The paper is organized as follows. In Section 2 we formally define the problem under study and ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287--313, June 1982.

.... network called a Content Delivery Network (CDN) The problem of information location to maximize performance or minimize the cost of a system that uses this information has been studied extensively since the early 1980s in several fields, including distributed databases, storage systems and CDNs [6, 1, 7]. This previous work solved the following decision problem: given a set of objects (files, pages of memory) and a set of potential locations, decide in which locations to place copies of the objects in order to optimize some objective and satisfy some additional constraints. A practical example of ....

L. Dowdy and D. Foster. Comparative models of the file assignment problem. Computing Surveys, 14, 1982.

....in a data center. It uses simple metrics for performance such as number of active connections and overall response time; it is similar in nature to the automatic loop in section 5.2.3 in its management of compute and network resources. Existing solutions to the file assignment problem [DF82, Wol89] use heuristic optimization models to assign files to disks to get improvements in I O response times. The work described on file allocation in [GWS91, SWZ98] will automatically determine an optimal stripe width for files, and stripe those files over a set of homogeneous disks. They then ....

Lawrence W. Dowdy and Derrel V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287--313, June 1982.

....and the CPU cost for processing the information such as updating, retrieving, and storing. Network cost are decomposed into communication and transmission costs. This problem of determining a solution to this problem under varying system conditions has been addressed in a number of publications [ChL87, DoF82, MoL77, Mur83, NoA87, Pol94, Pu86, RoS96, SeS79, Seg76, Yu85 and ZSA92]. Solutions to the file assignment problem can be divided into two classes: static file assignment and dynamic file assignment. In static file assignment [MoL77, Mur83, and Pu86] the design of the file assigner is based upon a set of time invariant system parameters. Therefore, in solutions ....

Dowdy L.W., Foster D.V. (1982); "Comparative Models of the File Assignment Problem"; ACM Computing Surveys Vol. 14 No. 2 (1982)

....placement of replicas across multiple servers in order to increase the overall system performance. Previous work on data replication problems mainly focused on networked systems, distributed database systems, and most recently, content delivery networks (CDNs) The File Allocation Problem (FAP) [13] studies how to assign files to different nodes in a distributed environment in order to optimize certain performance metric (e.g. total network traffic) The files can have multiple copies and they are allocated to different nodes with respect to the available storage capacity of these nodes. ....

....given metric. FAP has been proved to be NP complete. Many previous research projects assumed that access patterns are known and remain unchanged. Some solutions for dynamic environments were proposed in [21, 4, 39] A complete survey on models for FAP in computer network systems can be found in [13]. The document distribution problem of GDWS systems can be seen as an instance of the FAP, in which each document can have multiple copies and storage space limitation is considered. However, the solutions for the FAP cannot be used directly in a GDWS system for the following reason. Firstly, ....

L.W. Dowdy and D.V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):287-313, June 1982.

....distributed algorithm that evaluates its placement after each single access, while an RPA is evaluated with the frequency of hours or days and can thus be much more elaborate and even centralized. The replica placement problem is an instance of the classical file allocation problem (FAP) [5], in which only algorithms that allow replicas to be created are considered. There have been a plethora of proposed RPAs [6, 7, 8, 9, 11, 12, 13, 16, 17, 20] see [10] for an extensive analysis) for possible use in CDNs, but no direct comparison between them or with caching has been performed. The ....

L. W. Dowdy and D. V. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287--313, 1982.

....storage costs. We are interested in finding a residence set of minimum cost. We call this problem the optimal residence set problem for trees with read, write, and storage costs. The optimal residence set (file allocation) problem has been studied extensively in the literature. Dowdy and Fos ter [2] survey a number of mixed linear programming models for the file allocation problem. Wolfson and Milo [1] consider the optimal residence set problem without stor age costs for various interconnection networks (completelyconnected, tree, and ring networks) They show that the optimal residence set ....

L.W. Dowdy and D. V. Foster, "Comparative models of the file assignment problem," ACM Computing Surveys, vol. 14, no. 2, pp. 287-313, 1982.

....proposed in the CDN literature, but until now, there has been no systematic way to classify and compare these algorithms. The replica placement problem has also been studied extensively in several other distributed computing fields, including the file assignment problem (FAP) in storage systems [13], distributed databases [26] network location [11] and data management [29] So, a question that arises is how all these approaches are related to each other and how can one systematically represent and evaluate new as well as existing algorithms within and between these fields. In this paper, ....

....depending on the threshold) 5 Related Work Despite the importance of the replica placement problem and its practical implications, little has been done to compare the various approaches and their applicability to different system models. A survey of the file assignment problem by Dowdy et al.[13] was published in 1982. However, it focuses mostly on problem definitions and their classifications and comparison on a feature by feature basis. Levin and Morgan [26] introduced a framework for replica placement problems, in the context of distributed databases. They categorize algorithms ....

L. W. Dowdy and D. V. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287--313, 1982.

....[28] bin packing heuristic. Minerva [2] extends Forum to handle disk arrays. The comparison with Minerva, and hence a superset of Forum can be found in Section 4.1 where we show that Ergastulum is faster and generates as good or better designs. Existing solutions to the file assignment problem [11, 35] use heuristic optimization models to assign files to disks to get improvements in I O response times. The file allocation schemes described 23 in [12, 24] automatically determine an optimal stripe width for files, and stripe those files over a set of homogeneous disks. They then balance the load ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287--313, June 1982.

....write policy used in [10] the Minimum Spanning Tree write policy (or MST write policy) and the optimal replica set computed in [10] the MST optimal replica set for a tree network. The optimal residence set (file allocation) problem has been studied extensively in the literature. Dowdy and Foster [5] survey a number of mixed linear programming models for the file allocation problem. Wolfson and Milo [15] consider the optimal residence set problem without storage costs for various interconnection networks (completely connected, tree, and ring networks) They show that the optimal residence set ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. A CM Computing Surveys, 14(2):287-313, 1982.

....quality service with QoS guarantee by efficiently utilizing the total storage capacity and communication bandwidth in the server network. As one of the key problems in parallel and distributed processing, the mapping problem that we study here is also related to the File Allocation Problem (FAP) [13]. FAP is a classical optimization problem. However, it always considers the files that have to be placed onto a network as a fixed parameter without being able to change their size, as it is the case for media streams. For the mapping of media assets onto the server network, the restrictions ....

L.W.Dowdy, D.V.Foster. Comparative Models of the File Assignment Problem. Computing Surveys, Vol.14, No.2, 1982.

.... not only it is difficult to construct a realistic cost function due to the dynamic, distributed nature of the system, and the fact that tokens are perishable resources (i.e. they cease to exist after being used) but also it is long known that even simpler formulations of the problem are NP hard [4, 5]. Since optimal solutions are impractical, we focus on heuristics based solutions in the rest of the paper. In particular we avoid global strategies that require tight synchronization, and concentrate only on decentralized strategies that make progress using only pair wise synchronization. Note ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287-313, 1982.

....It is also based on an iterative loop, like Hippodrome, but it focuses on allocating computational resources. Its resource allocation framework is based on an economic model that factors in the tradeoffs between the service quality and the cost. Existing solutions to the file assignment problem [13, 34] use heuristic optimization models to assign files to disks to get improvements in I O response times. The file allocation schemes described in [16, 28] will automatically determine an optimal stripe width for files, and stripe those files over a set of homogeneous disks. They then balance the ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287--313, June 1982.

....general context of data organization, say, in a distributed object store [Lisk92, RD90, Stam84, LMW91, PZ91] distributed databases and services in computer networks. The amount of work on the subject in the systems community is overwhelming. For example, the 1981 survey paper by Dowdy and Foster [DF82] and 1990 survey paper by Gavish and Sheng [GS90] dealing with the file allocation problem, cite close to a hundred references. In theory community, only special restricted versions of the problem have been previously addressed, such as 1 ffl Uniprocessor paging [ST85, KMRS88, FKL 88, ....

D. Dowdy and D. Foster. Comparative models of the file assignment problem. Computing Surveys, 14(2), Jun 1982.

....on the same processor. The allocation problem of data in distributed systems has been widely analyzed in the context of the file allocation problem. Several analytical models which build the optimal allocation of files under very different assumptions and with different objectives are described in [5, 9]. In [9] 12 models for the file assignment are surveyed and compared. In [5] authors present a model for resource allocation in distributed systems which aims at distributing resources so that many applications become completely local. Grouping of rules according to subject or use in order to ....

....processor. The allocation problem of data in distributed systems has been widely analyzed in the context of the file allocation problem. Several analytical models which build the optimal allocation of files under very different assumptions and with different objectives are described in [5, 9] In [9], 12 models for the file assignment are surveyed and compared. In [5] authors present a model for resource allocation in distributed systems which aims at distributing resources so that many applications become completely local. Grouping of rules according to subject or use in order to assist ....

L.W. Dowdy and D.V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2), 1982.

.... not only it is difficult to construct a realistic cost function due to the dynamic, distributed nature of the system, and the fact that tokens are perishable resources (i.e. they cease to exist after being used) but also it is long known that even simpler formulations of the problem are NP hard [4, 5]. Since optimal solutions are impractical, we focus on heuristicsbased solutions. In particular we avoid global strategies that require tight synchronization, and concentrate only on decentralized, pair wise strategies that make progress using only pair wise synchronization. Note that previous ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287-313, 1982.

....has applied optimization techniques to the file assignment problem. Files, typically characterized by the probability that they will be accessed from particular nodes in a network, are mapped to storage devices with the aim of optimizing some objective such as communication overhead or reliability [3, 10, 19]. Our work is most closely related to the variant where files characterized by size and access rate are placed on a given set of storage devices, with the aim of minimizing access latency or maximizing throughput [24] However, these approaches do not design the system that will efficiently ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287--313, June 1982.

....quality service with QoS guarantee by efficiently utilizing the total storage capacity and communication bandwidth in the server network. As one of the key problems in parallel and distributed processing, the mapping problem that we study here is also related to the File Allocation Problem (FAP) [13]. FAP is a classical optimization problem. However, it always considers the files that have to be placed onto a network as a fixed parameter without being able to change their size, as it is the case for media streams. For the mapping of media assets onto the server network, the restrictions ....

L.W.Dowdy, D.V.Foster. Comparative Models of the File Assignment Problem. Computing Surveys, Vol.14, No.2, 1982.

....and the access patterns of the clients into account. This combinatorial problem can be formulated mathematically as a NP complete optimization problem and can therefore be solved by use of some heuristic optimization solutions. The mapping problem extends the File Allocation Problem (FAP) [6], a well known problem in parallel and distributed processing, using the nice feature of media assets that the assets are scalable in size as they are encoded in different bit rates. 3.2 Formalizing the Mapping Problem We assume the hierarchy of a server network is the binary tree with ring, ....

L.W.Dowdy, D.V.Foster. Comparative Models of the File Assignment Problem. Computing Surveys, Vol.14, No.2, 1982.

.... not only it is difficult to construct a realistic cost function due to the dynamic, distributed nature of the system, and the fact that tokens are perishable resources (i.e. they cease to exist after being used) but also it is long known that even simpler formulations of the problem are NP hard [4, 5]. Since optimal solutions are impractical, we focus on heuristics based solutions in the rest of the paper. In particular we avoid global strategies that require tight synchronization, and concentrate only on decentralized strategies that make progress using only pair wise synchronization. Note ....

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287-313, 1982.

....to the backbone via direct links. The major requirement is that the total access cost for the users be within a specified bound, whereas the construction cost of the backbone network should be minimized. Problems of similar nature arise in the efficient maintenance of distributed databases [3, 4, 7, 15, 22]. Other applications of the Budget Constrained Connected Median Problem studied in this paper include location theory and manufacturing logistics (see [20, 21] and the references cited therein) The above problems can be cast in a graph theoretic framework as follows: Given an undirected graph G ....

L. W. Dowdy and D. V. Foster, Comparative models of the file assignment problem, ACM Computing Surveys 14 (1982), no. 2, 287--313.

....Yair Bartal Amos Fiat Department of Computer Science School of Mathematics Tel Aviv University Tel Aviv 69978, Israel. Yuval Rabani Abstract We deal with the competitive analysis of algorithms for managing data in a distributed environment. We deal with the file allocation problem ([DF], ML] where copies of a file may be be stored in the local storage of some subset of processors. Copies may be replicated and discarded over time so as to optimize communication costs, but multiple copies must be kept consistent and at least one copy must be stored somewhere in the network ....

....(a generalization of mobile user tracking [AP1, AP3] to transform our competitive data management algorithms into distributed algorithms themselves. 1 Introduction The management of data in a multiprocessing environment has been extensively studied. The 1981 survey paper by Dowdy and Foster [DF], dealing with the file allocation (or assignment) problem, cites close to a hundred references. 1 The file allocation problem has a plethora of models, with differing design goals and assumptions. DF] compares studies on fourteen different models, and mentions several others. We deal with ....

[Article contains additional citation context not shown here]

D. Dowdy and D. Foster. Comparative Models of The File Assignment Problem. Computing Surveys, 14(2), June 1982.

....data management is the design of a dynamic allocation of file copies in a network in order to achieve low communication costs. Distributed file assignment problems have been extensively studied in management science, engineering, and computer systems. The 1981 survey paper by Dowdy and Foster [DF81] compares studies on different models, and cites close to a hundred references. The 1990 paper by Gavish and Sheng [GS90] gives a survey of recent research on dynamic file assignment problems, where the locations of file copies may be changed over time. All of these models rely heavily on prior ....

D. Dowdy and D. Foster. Comparative Models of The File Assignment Problem. Computing Surveys, 14(2), June 1982.

....algorithm. Our main result, shown in Section 5, is that the amortizing algorithm achieves a constant factor approximation for the hierarchical placement problem. In Section 6, we describe an efficient distributed implementation of the amortizing algorithm. 1.3 Related work. Dowdy and Foster [13] initiated the study of cooperative caching in the context of allocating files in a distributed network [13] A sequence of results [6, 20] obtained improved algorithms for centralized as well as distributed file allocation. These results, however, did not consider cache capacities at the ....

....approximation for the hierarchical placement problem. In Section 6, we describe an efficient distributed implementation of the amortizing algorithm. 1.3 Related work. Dowdy and Foster [13] initiated the study of cooperative caching in the context of allocating files in a distributed network [13]. A sequence of results [6, 20] obtained improved algorithms for centralized as well as distributed file allocation. These results, however, did not consider cache capacities at the individual machines. As mentioned earlier in this section, Awerbuch, Bartal, and Fiat [2] provide a polylog(n) ....

[Article contains additional citation context not shown here]

D. Dowdy and D. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14:287--313, 1982.

....recent results on the approximation of general metrics by tree metrics [5, 6, 12] imply that any hierarchical placement algorithm can be used to obtain a placement algorithm for general metrics giving up an extra O(log n log log n) factor in the approximation. 1. 3 Related work Dowdy and Foster [15] initiated the study of cooperative caching in the context of allocating files in a distributed network [15] A sequence of results [2, 7, 23] obtained improved algorithms for centralized as well as distributed file allocation. These results, however, did not consider cache capacities at the ....

....placement algorithm can be used to obtain a placement algorithm for general metrics giving up an extra O(log n log log n) factor in the approximation. 1. 3 Related work Dowdy and Foster [15] initiated the study of cooperative caching in the context of allocating files in a distributed network [15]. A sequence of results [2, 7, 23] obtained improved algorithms for centralized as well as distributed file allocation. These results, however, did not consider cache capacities at the individual nodes. As mentioned earlier in this section, Awerbuch, Bartal, and Fiat [3] provide a ....

D. Dowdy and D. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14:287--313, 1982.

....fiat math.tau.ac.il. 0 1 Introduction 1. 1 The Problem This paper is concerned with finding efficient strategy for replicating and migrating data in presence of a dynamic pattern of reads and writes (see figure 1, details in Section 2) This is a well studied problem in distributed systems [DF82] F F F F read write Figure 1: The file allocation problem. A subset of the network processors are holding copies of a file F . The file may be read or written by any of the processors. In order to achieve maximum efficiency, we need to maximally exploit the locality of reference in the ....

....information 2 and thus cannot be be efficiently implemented in a distributed setting. 1.3 Existing results. The management of data in a multiprocessing environment has been extensively studied, both from the theoretical and the practical standpoints. The 1981 survey paper by Dowdy and Foster [DF82] dealing with the file allocation (or assignment) problem, cites close to a hundred references. First competitive algorithms for special cases of the centralized version of the problem were found by Black and Sleator and by Westbrook [BS89, Wes91] The file allocation problem may be viewed as ....

D. Dowdy and D. Foster. Comparative models of the file assignment problem. Computing Surveys, 14(2), Jun 1982.

....the file distribution or reconstruction algorithms to be of any particular form, we aim at simple and efficient ones. The problem of file allocation in a network, i.e. of storing a file in a network so that every processor has easy access to the file, has been considered in many variants (see [4] for a survey) The specific version of reconstruction from adjacent nodes only has received attention in the form of file segmentation, where the task is to partition the file so that, for each node u in the network, the union of the file segments stored at nodes adjacent to u is the complete ....

.... for a survey) The specific version of reconstruction from adjacent nodes only has received attention in the form of file segmentation, where the task is to partition the file so that, for each node u in the network, the union of the file segments stored at nodes adjacent to u is the complete file [4][8] 13] As we shall see, allowing more general reconstruction procedures than simply taking the union of file segments at adjacent nodes can result in a considerable savings of the total amount of memory required: Letting Delta G denote the maximum degree of any node in G, the memory requirement ....

L.W. Dowdy, D.V. Foster, Comparative models of the file assignment problem, Comp. Surveys, 14 (1982), 287--313.

....in one of the servers in the distributed system. The servers have to determine the storage location of each data item. If a central decision maker exists, then the problem is called the file allocation problem, which is a well known problem in computer science and operations research (e.g. [13, 1]. We consider a similar problem for a non cooperative environment where there is no central controller and no global utility. Each server in the environment is independent and has its own commercial interests, but would like to cooperate with the other information servers in order to make ....

....to optimize the system performance. The various file allocation models are presented as optimization problems in terms of objective functions and constraints. The file allocation problem is known to be NP complete [18] and many 4 contributions have been made towards solving it. Dowdy and Foster [13] have given a brief and extended description, as well as mathematical formulations of the objective function and the constraints which may be considered in the file allocation problem, and they have described some known solution techniques. Apers [1] investigates the problem of the allocation of ....

[Article contains additional citation context not shown here]

L. W. Dowdy and D. V. Foster. Comparative models of the file assignment problem. Computing Survey, 14 (2):289--313, 1982. 46

....fiat math.tau.ac.il. 0 1 Introduction 1. 1 The Problem This paper is concerned with finding efficient strategies for replicating and migrating data in presence of dynamic patterns of reads and writes (see figure 1, details in Section 2) This is a well studied problem in distributed systems [DF82] F F F F read write Figure 1: The file allocation problem. A subset of the network processors are holding copies of a file F . The file may be read or written by any of the processors. In order to achieve maximum efficiency, we need to maximally exploit the locality of reference in the ....

....information and thus cannot be be efficiently implemented in a distributed setting. 1.3 Previous results. The management of data in a multiprocessing environment has been extensively studied, both from the theoretical and the practical standpoints. The 1981 survey paper by Dowdy and Foster [DF82] dealing with the file allocation (or assignment) problem, cites close to a hundred references. The file allocation problem that we deal with in this paper may be viewed as the combined solution to the two subproblems defined in [BS89] file migration and file replication. The file migration ....

D. Dowdy and D. Foster. Comparative models of the file assignment problem. Computing Surveys, 14(2), June 1982.

....the update and the lookup costs. If the information replicated is coarse then it needs to be updated less frequently in the expense of a higher lookup resolution cost. Choosing the network sites at which to maintain replicas of the current location of a mobile user resembles the file allocation [15] and the database allocation [46] problem. These classical problems are concerned with the selection of sites at which to maintain replicas of files or database partitions. The selection of sites is based on the read write pattern of each file or partition, that is the number of read and write ....

L. W. Dowdy and D. V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):288--313, June 1982.

....(FAP) We will now review the FAP. 177 5.11.1 File Allocation Problem The problem of replicating files in a distributed environment is similar to the File Allocation Problem . The general problem of allocating a file, or replica of it, on a network node is hard and has been studied extensively [42]. The major result [45] is its NP Completeness [50] although, depending on how the problem is formulated, some instances of this problem may be polynomial time computable. FAP consists of a graph whose nodes are computers, or storage facilities, and its edges are communication links. In the ....

....and communication are minimized. In other words, according to Wah [125] we are looking for an index set I = fY k = 0 or 1; k = 1; Delta Delta Delta ; ng such that the cost function is minimized, where Y k = 1 if node k has a copy and 0 otherwise. Variations of this problem exist. Dowdy [42] presents a survey of different interpretations and models for this problem in an increasing level of complexity and detail. He separates the models in two orthogonal categories: the first one dealing with cost and the second with performance. Most models constructed for solving the first set, ....

[Article contains additional citation context not shown here]

Lawrence W. Dowdy and Derrel V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):287--313, June 1982.

....i;j : expected number of calls from j to i over time T , and U i : number of moves made by i over T . Other Factors: database service capacity, storage Other Issues: ffl where to keep replication set ffl other applications ffl granularity of location replicas Comparison with file allocation [3]and database allocation [12] problem. E. Pitoura Summer School on Mobile Computing, Jyvaskyla, 1998 6 Replication 1. Per User Profile Replication [17] Problem Formulation Let M : the number of users and N : number of zones. Find a replication assignment of a user s profile P i to a ....

L. W. Dowdy and D. V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):288--313, June 1982.

....bounded by a fixed polynomial in n. 4.2 Data Management and Job Scheduling Problems Bartal, Fiat and Rabani [BFR92] have been the first to deal with the competitive analysis of algorithms for managing data in a distributed environment. More specifically, they consider the file allocation problem [DF82] where copies of a file may be stored in the local storage of some subset of processors, copies may be replicated and discarded over time so as to optimize communication costs, but multiple copies must be kept consistent and at least one copy must be stored somewhere in the network at all times. ....

D. Dowdy and D. Foste. "Comparative Models of the File Assignment Problem ". Computing Surveys, 14(2), 1982.

....provide insights for building high performance IR servers using symmetric multiprocessors. 1.2. 2 Partial Collection Replication Replication has long been used in the areas of distributed file systems and distributed database systems in order to improve system performance and availability [1, 2, 3, 16, 19, 24, 25, 30, 43, 44, 51, 65, 69, 76] (see Section 2.5 for detailed discussions) Database vendors such as Oracle, Informix, and Sybase offer replication techniques in their products [60] However, there are several distinctions that make this work unique. First, information retrieval is different from the focus of previous database ....

....areas of distributed file systems and distributed database systems in order to improve system availability and performance. Much work, especially early work, focuses on availability due to replication [2, 3, 16, 19, 25, 30, 51, 65, 69] Relatively few papers focus on performance due to replication [1, 24, 44, 76]. Replication strategies can be classified as static and dynamic replication, based on when the system decides what to replicate. The File Allocation Problem (FAP) is an example of static replication, where data is replicated at design time based on probable access patterns, and remains unchanged ....

[Article contains additional citation context not shown here]

Dowdy, D., and Foster, D. Comparative models of the file assignment problem. Computing Surveys 14, 2 (June 1982).

No context found.

L.W. Dowdy and D.V. Foster. Comparative Models of the File Assignment Problem. ACM Computing Surveys, 14(2):287-313, June 1982.

No context found.

L. W. Dowdy and Derrell V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287-- 313, 1982.

No context found.

L.W. Dowdy and D.V. Foster, "Comparative Models of the File Assignment Problem", ACM Computing Surveys, 14(2):287--313, 1982.

No context found.

L. Dowdy and D. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287-- 313, 1982.

No context found.

L.W. Dowdy and D.V. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2):287-- 313, 1982.

No context found.

L.W. Dowdy and D.V. Foster, "Comparative models of the file assignment problem," ACM Computer Sueveys, vol. 14(2), pp. 287-313, 1982.

No context found.

L. Dowdy and D. Foster, "Comparative models of the file assignment problem," ACM Comput. Surv., vol. 14, no. 2, pp. 287--313, June 1982.

No context found.

L. W. Dowdy and D. V. Foster, "Comparative models of the file assignment problem," ACM Computing Surveys, vol. 14, no. 2, pp. 287-- 313, June 1982.

No context found.

L. Dowdy and D. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287--313, 1982.

No context found.

L.W. Dowdy and D.V. Foster, "Comparative Models of the File Assignment Problem", ACM Computing Surveys, 14(2):287--313, 1982.

No context found.

L. W. Dowdy and D. V. Foster, "Comparative Models of the File Assignment Problem," ACM Computer Surveys, vol. 14, no. 2, pp. 287--313, 1982.

No context found.

D. Dowdy and D. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14:287--313, 1982.

No context found.

L. Dowdy and D. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287--313, 1982.

No context found.

L. Dowdy and D. Foster. Comparative Models of the File Assignment Problem. ACM Computer Surveys, 14(2):287--313, 1982.

No context found.

W. Dowdy and D. Foster. Comparative models of the file assignment problem. ACM Computing Surveys, 14(2), 1982.

No context found.

D. Dowdy and D. Foster, "Comparative models of the file assignment problem," ACM Computing Surveyv vol.14, no.2, pp.287--313, 1982.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC