A fleet of unmanned air vehicles undertakes a mission to disable an enemy airfield. Pre-mission intelligence indicates that the airfield is not defended, and mission planning proceeds accordingly. While the UAVs are en route to the target, new intelligence indicates that a mobile surface-to-air missile launcher now guards the airfield. The UAVs autonomously replan their mission, dividing into two groups—a SAM-suppression unit and an airfield-suppression unit—and proceed to accomplish their objectives. During the flight, specialized algorithms for detecting and recognizing SAM launchers automatically upload and are integrated into the SAM-suppression unit’s software. In this scenario, new software components are dynamically inserted into fielded, heterogeneous systems without requiring system restart, or indeed, any downtime. Mission replanning relies on analyses that include feedback from current performance. Furthermore, such replanning can take place autonomously, can involve multiple, distributed, cooperating planners, and where major changes are demanded and require human approval or guidance, can cooperate with mission analysts. Throughout, system integrity requires the assurance of consistency, correctness, and coordination of changes. Other applications for fleets of UAVs

Motivated by applications such as document and image classification in information retrieval, we consider the problem of clustering dynamic point sets in a metric space. We propose a model called incremental clustering which is based on a careful analysis of the requirements of the information retrieval application, and which should also be useful in other applications. The goal is to efficiently maintain clusters of small diameter as new points are inserted. We analyze several natural greedy algorithms and demonstrate that they perform poorly. We propose new deterministic and randomized incremental clustering algorithms which have a provably good performance. We complement our positive results with lower bounds on the performance of incremental algorithms. Finally, we consider the dual clustering problem where the clusters are of fixed diameter, and the goal is to minimize the number of clusters. 1 Introduction We consider the following problem: as a sequence of points from a metric...

At each point in time a decision maker must choose a decision. The payoff in a period from the decision chosen depends on the decision as well as the state of the world that obtains at that time. The difficulty is that the decision must be made in advance of any knowledge, even probabilistic, about which state of the world will obtain. A range of problems from a variety of disciplines can be framed in this way. In this

Caching (paging) is a well-studied problem in online algorithms, usually studied under the assumption that all pages have a uniform size and a uniform fault cost (uni- form caching). However, recent applications related to the web involve situations in which pages can be of different sizes and costs. This general caching problem seems more intricate than the uniform version. In particular, the offline case itself is NP-hard. Only a few results exist for the general caching problem [8, 17]. This paper seeks to develop good offline page replacement policies for the general caching problem, with the hope that any insight gained here may lead to good online algorithms. Our first main result is that by using only a small amount of additional memory, say O(1) times the largest page size, we can obtain an O(1)-approximation to the general caching problem. Note that the largest page size is typically a very small fraction of the total cache size, say 1%. Our second result is that when no add...

Abstract. Broadcast disks are an emerging paradigm for massive data dissemination. In a broadcast disk, data is divided into n equal-sized pages, and pages are broadcast in a round-robin fashion by a server. Broadcast disks are effective because many clients can simultaneously retrieve any transmitted data. Paging is used by the clients to improve performance, much as in virtual memory systems. However, paging on broadcast disks differs from virtual memory paging in at least two fundamental aspects: • A page fault in the broadcast disk model has a variable cost that depends on the requested page as well as the current state of the broadcast. • Prefetching is both natural and a provably essential mechanism for achieving significantly better competitive ratios in broadcast disk paging. In this paper, we design a deterministic algorithm that uses prefetching to achieve an O(n log k) competitive ratio for the broadcast disk paging problem, where k denotes the size of the client’s cache. We also show a matching lower bound of Ω(n log k) that applies even when the adversary is not allowed to use prefetching. In contrast, we show that when prefetching is not allowed, no deterministic online algorithm can achieve a competitive ratio better than Ω(nk). Moreover, we show a lower bound of Ω(n log k) on the competitive ratio achievable by any nonprefetching randomized algorithm against an oblivious adversary. These lower bounds are trivially matched from above by known results about deterministic and randomized marking algorithms for paging. An interpretation of our results is that in the broadcast disk paging, prefetching is a perfect substitute for randomization.

The concept of "real-time" has different meanings in the systems and theory communities. As a consequence, the currently available formal real-time models do not capture all the practically relevant aspects of such computations. We propose a new definition, and we believe that it allows a unfied treatment of all practically meaningful variants of real-time computations. Then, we use the developed formalism to model two important features of real-time algorithms, namely the presence of deadlines and the real-time arrival of input data. We also emphasize the expressive power of our model by using it to formalize aspects from the areas of real-time database systems and ad hoc networks. We offer formulations of the recognition problem for real-time database systems and the routing problem in ad hoc networks. Finally, we suggest a variant of our formalism that is suited for modeling parallel distributed real-time algorithms. We believe that the proposed formalism is a first step towards a unified and realistic complexity theory for real-time parallel computations.

We assume the multitape real-time Turing machine as a formal model for parallel real-time computation. Then, we show that, for any positive integer k, there is at least one language L k which is accepted by a k-tape real-time Turing machine, but cannot be accepted by a (k-1)-tape real-time Turing machine. It follows therefore that the languages accepted by real-time Turing machines form an infinite hierarchy with respect to the number of tapes used. Although this result was previously obtained in [1], our proof is considerably shorter, and explicitly builds the languages L k .

this paper because we assume throughout that the total initial wealth of all systems of agents is $1

Abstract—This paper considers the design of handoff rerouting algorithms for reducing the overall session cost in personal communication systems (PCS). Most modern communication systems that are used as an infrastructure for PCS networks are based on connection-based technologies. In these systems, the session cost is composed of two components. The setup cost represents the cost associated with the handoff operations, and the hold cost determines the expense related to the use of network resources held by the connection. This work introduces for the first time, rerouting algorithms for general graphs which are cost effective in terms of their worst-case analysis. The algorithms are analyzed using a competitive analysis approach, and it is proved that the competitive ratio of the proposed algorithms is a small constant of which the precise value depends on the ratio between the setup costs and the hold costs of the links. We also prove a lower bound of 2 on the competitive ratio of any online algorithm, which means that the proposed algorithms are close in terms of worst case behavior to the best possible rerouting algorithm. In addition, experimental results also show that the proposed algorithms indeed balance between the session setup cost and the hold cost, yielding overall lower cost when compared to other algorithms described in the literature. Index Terms—Competitive analysis, connection management, handoff rerouting algorithms, online algorithms, personal communication systems (PCS). I.

Parallel Vector Access (PVA) is a technique that exploits the regularity of vector or stream accesses to perform them efficiently in parallel on a multibank memory system. The performance of vector applications may be improved by a memory controller that performs scatter/gather operations, so that only the vector or stream elements that are accessed by the application are transmitted across the system bus. These scatter/gather operations can be accelerated by broadcasting vector operations in parallel to all memory banks, each of which implements an algorithm to determine which elements of the requested vector it contains. This thesis presents the mathematical foundations behind one such algorithm for efficient parallel access of base-stride vectors on both word interleaved and cache-line interleaved memory systems. The design of a memory controller subcomponent that uses the PVA algorithm to improve the performance of applications with strided access patterns is described. The hardware implementation issues behind such a memory controller are investigated. Gate level simulation on vector kernels demonstrates that this parallel approach allows the PVA to load elements up to 32.8 times faster than a conventional SDRAM memory system and 3.3 times faster than a pipelined vector unit, without hurting normal cache-line fill performance.