By Topic

Parallel Architectures, Algorithms and Networks, 2004. Proceedings. 7th International Symposium on

Date 10-12 May 2004

Filter Results

Displaying Results 1 - 25 of 112
  • Wavelength assignment of parallel FFT communication pattern on a class of regular optical WDM networks

    Page(s): 495 - 500
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1377 KB) |  | HTML iconHTML  

    Wavelength assignment is a key topic in WDM optical interconnection networks. How to realize different parallel communication patterns on optical interconnection networks is a hot research field. Based on the two embedding ways of sequential mapping and shift-reversal mapping, wavelength assignment of realizing parallel FFT communication pattern on a class of regular optical WDM networks is discussed in this paper. By sequential mapping, the numbers of wavelengths required to realize parallel FFT communication pattern with 2n nodes on WDM linear array, ring, 2-D mesh and 2-D torus are 2n - 1 , 2n - 1 , 2max(k,n - k) - 1 and 2max(k,n - k) - 1 respectively. By shift-reversal mapping, the numbers of wavelengths required is max(3×2n - 3,2) , 2n - 2 , max(3×2max(k,n - k) - 3,2) and 2max(k,n - k) - 2. These results show that shift-reversal mapping outperforms sequential mapping. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Matrix factorizations for parallel integer transforms

    Page(s): 254 - 259
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1424 KB) |  | HTML iconHTML  

    Integer mapping is critical for lossless source coding and the techniques have been used for image compression in the new international image compression standard, JPEG 2000. In this paper, from block factorizations for any nonsingular transform matrix, we introduce two types of parallel elementary reversible matrix (PERM) factorizations which are helpful for the parallelization of perfectly reversible integer transforms. With improved degree of parallelism (DOP) and parallel performance, the cost of multiplication and addition can be respectively reduced to O(logN) and O(log2N) for an N-by-N transform matrix. These make PERM factorizations an effective means of developing parallel integer transforms for large matrices. Besides, we also present a scheme to block the matrix and allocate the load of processors for efficient transformation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Space optimal packet classification for 2D conflict-free filters

    Page(s): 260 - 265
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1455 KB) |  | HTML iconHTML  

    In this paper, we study the 2D packet classification problem for a set of conflict-free filters in an IP network. We design a linear space data structure with O(min{logiu loglogn, √lognloglogn}) query time where n is the number of filters in the router and w is the number of bits in an IP address. This is the first optimal space data structure with poly-logarithmic query time for this problem. Our technique can also be extended to solve the binary dispatching problem in object-oriented programming. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On a region-of-interest based approach to robust wireless video transmission

    Page(s): 385 - 390
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1410 KB) |  | HTML iconHTML  

    This paper presents a scheme aiming at transmitting real-time video to wireless channel with vigorously varying quality, which is in practice the norm rather than the exception. Region of interest (ROI) is an efficient approach to making the video more adaptive to the wireless channel because ROI is the region that human eyes tend to put more attention to than the remainder region (RM). In our proposed scheme, we adopted this feature. The real-time source video stream is divided into two regions, the ROI and the RM regions. The two regions were encoded using H.263 standard codec such that the video transmission is adaptive to the current channel state, which is characterized by the effective data rate that varies from tens of kilobits per second to hundreds of kilobits per second. Channel state parameters are fed back to the source coder to adjust the compression ratio as well as the intra/inter options of the encoders. Results including frame loss probability, compression characteristics, peak signal to noise ratio (PSNR) against channel states are given, indicating that the resulting adaptive video codec can respond judiciously to time-varying channel quality. Our scheme is evaluated together with a ROI-enabled moving picture coding standard JPEG2000. Using the features provided in JPEG2000, we have made the JPEG2000 codec adaptive to the vigorously varying wireless channel and then compared it with the H.263 scheme. Our technique is suitable for a broad area of applications including real-time news reporting and video conferencing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Checkpointing in hybrid distributed systems

    Page(s): 136 - 141
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1407 KB) |  | HTML iconHTML  

    To provide fault tolerance to computer systems suffering from transient faults, checkpointing and rollback recovery is widely-used. Among other techniques, two primary checkpointing schemes have been proposed: independent and coordinated schemes. However, most existing work addresses only the need to employ a single checkpointing and rollback recovery scheme to a target system. In this paper, issues are discussed and a new algorithm is developed to address the need of integrating independent and coordinated checkpointing schemes for applications running in a hybrid distributed environment containing multiple heterogeneous subsystems. The required changes to the original checkpointing schemes for each subsystem and the overall prevented unnecessary rollbacks for the integrated system are presented. Also described is an algorithm for collecting garbage checkpoints in the combined hybrid system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimizing relay control in Wireless 1394 network

    Page(s): 54 - 59
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1360 KB) |  | HTML iconHTML  

    Relay control ought to be performed when a direct communication link cannot be established between the source station and the target station. We propose in this paper how such function can be accomplished effectively in Wireless 1394 based networks. Simulation results yield the fact that the proposed scheme not only can optimize the data relay control from the perspectives of the premium (the lowest cost) relay path and relay station selection but also can improve the system's performance in terms of its higher packet delivery ratio and lower packet end-to-end delay in a shadowing prone home environment. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Arm up administrators: automated vulnerability management

    Page(s): 587 - 593
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1311 KB) |  | HTML iconHTML  

    With the continuous flood of vulnerabilities of computers, vulnerability management is a very important task for administrators to keep systems as secure as possible. Facing numerous attackers armed with complicated, automated tools, current manual vulnerability management by administrators is so time-consuming, error-prone. Administrators also do need automated defensive tools. This paper proposes an open framework of automated vulnerability management that dramatically alleviates the burden of administrators and improves the security of systems. In this framework, we present three XML based markup languages, Common Vulnerability Markup Language (CVML), System Information Markup Language (SIML), Network System Markup Language (NSML) to express crucial information related to systems and vulnerabilities to facilitate automated exchange and processing. Host vulnerability managers (HVMs) running on the target host maintain the crucial system information in SIML, receive vulnerability advisories in CVML from various sources, decide what vulnerabilities exist, and try to fix vulnerabilities automatically if possible. Domain vulnerability managers (DVMs) are responsible for the vulnerability management in NSML of the local network. DVMs correlate reports from HVMs and scan for network-based vulnerabilities in this domain. We have implemented a prototype of the framework that shows the effectiveness and efficiency of our solution. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A study of concurrency control in Web-based distributed real-time database system using extended time Petri nets

    Page(s): 67 - 72
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1385 KB) |  | HTML iconHTML  

    The Web-based database system has both distributed and real-time features. In addition to satisfying data consistency requirements as in conventional distributed database systems, concurrency control in Web-based distributed real-time database systems (DRTDBS) must also satisfy timing constraints, such as deadlines associated with transactions. So, the concurrency control problems become more complex and extremely difficult in a Web-based DRTDBS. Petri nets are promising tools for modeling and analysis information processing systems that are characterized as being concurrent, asynchronous, parallel and distributed. An extended time Petri net model for describing and analyzing the concurrent control of the Web-based DRTDBS is presented in this paper. The model has features of deadlock-free and serializability and can prevent useless restarts and useless waits. We compose the reduced Petri net models of all sites into a Petri net model for globe concurrent transactions using synchronizing composition. Moreover, the paper largely reduces the scale of the Petri net model for every site with reduction technique before composing. It solves the state explosion problem in reachability analysis of Petri nets. Last, a sufficient and necessary condition for judging whether the whole system is deadlock is given in the paper. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal loop scheduling with register constraints using flow graphs

    Page(s): 180 - 185
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1349 KB) |  | HTML iconHTML  

    We present a novel loop scheduling approach using a generalized flow graph model of the resource constraints. From this model we derive a new flow graph to incorporate register constraints. Our linear programming implementation produces an optimum loop schedule, respecting the constraints on functional units and registers in a single optimization problem. Moreover, the iteration interval is treated as a rational number, and the approach supports heterogeneous processor architectures and pipelined functional units. Compared to earlier approaches, the solution can reduce the problem complexity and solution time, and provide faster loop schedules. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance and stability bounds for dynamic networks

    Page(s): 239 - 246
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1425 KB) |  | HTML iconHTML  

    In this work, we study the impact of dynamically changing link capacities on the delay bounds of LIS (longest-in-system) and SIS (shortest-in-system) protocols on specific networks (that can be modelled as directed acyclic graphs - DAGs) and stability bounds of greedy contention-resolution protocols running on arbitrary networks under the adversarial queueing theory. Especially, we consider the model of dynamic capacities, where each link capacity may take on integer values from [1,C] with C > 1, under a (w,p)-adversary. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computing the rupture degrees of graphs

    Page(s): 368 - 373
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1368 KB) |  | HTML iconHTML  

    The rupture degree of a noncomplete connected graph G is defined by r(G) = max{ω(G - X) - |X| - m(G - X) : X ⊂ V(G), ω(G - X) ≥ 2}, where ω(G - X) denotes the number of components in the graph G - X. For a complete graph Kn, we define r(Kn) = 1 - n. This parameter can be used to measure the vulnerability of a graph. To some extent, it represents a trade-off between the amount of work done to damage the network and how badly the network is damaged. In this paper, we prove that the problem of computing the rupture degree of a graph is NP-complete. We obtain the rupture degree of the Cartesian product of some special graphs and also give the exact values or bounds for the rupture degrees of Harary graphs. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A study of data fusion in Cayley graphs G(sn,pn)

    Page(s): 557 - 562
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2091 KB) |  | HTML iconHTML  

    In this paper, we examine a method for the fusion of ranked data in the context of a Cayley graph. We investigate this Cayley graph model for optimization of fusion by rank combination. We outline a method of data fusion by combination of weighted rankings. Information systems are represented as nodes in a Cayley graph. Our goal is to determine a metric of diversity and performance in this graph in order to build a model for optimizing fusion by rank combination. We use the Kendall distance between nodes in the Cayley graph of the symmetric group Sn as a measure of performance. In doing so we demonstrate that in S6 there is a quadratic relationship between the weights of the fusion of two information systems and the performance of the fusion in our abstract space. From such a relationship we propose a set of functions for extrapolating optimal fusion weights in the symmetric group Sn. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Particle swarm optimization algorithm in signal detection and blind extraction

    Page(s): 37 - 41
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1353 KB) |  | HTML iconHTML  

    The particle swarm optimization (PSO) algorithm, which originated as a simulation of a simplified social system, is an evolutionary computation technique. In this paper the binary and real-valued versions of PSO algorithm are exploited in two important signal processing paradigm: multiuser detection (MUD) and blind extraction of sources (BES), respectively. The novel approaches are effective and efficient with parallel processing structure and relatively feasible implementation. Simulation results validate either PSO-MUD or PSO-BES has a significant performance improvement over conventional methods. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Effects of nuisance parameters on the family-based linkage tests under general disease models

    Page(s): 614 - 618
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1338 KB) |  | HTML iconHTML  

    Family-based linkage tests are methods for detection of the genetic basis of a disease. For the case-parent triad data, it is known that tests of linkage interfere with the association parameter under recessive disease models. The purpose of the paper is to discuss the effects of nuisance parameters on the family-based linkage tests under general models. It is shown that the linkage tests do not only interfere with the association parameter, but also with the gene substitution effect in general models. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An embedded Web server for equipment

    Page(s): 345 - 350
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1392 KB) |  | HTML iconHTML  

    In this paper a design concept is proposed to accomplish a full function Web server on a strictly limited resource space. Through this embedded Web server user can access their equipments remotely. The equipment mentioned here could be home appliances and factory devices. This paper is focused on realization of the TCP/IP suite and user development platform for this embedded Web server. A key goal of the present paper is to provide an effective approach of access to traditional equipments that have no Internet interface and a reduction policy of the TCP/IP protocol suite. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance evaluation of data distributions with load-balancing for sparse arrays

    Page(s): 207 - 212
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1466 KB) |  | HTML iconHTML  

    In our previous work, we have studied three data distribution schemes, Send Followed Compress (SFC), Compress Followed Send (CFS), and Encoding- Decoding (ED), for sparse arrays based on the traditional matrix representation (TMR) scheme. Since multidimensional arrays can also be represented by the extended Karnaugh map representation (EKMR) scheme, in this paper, we first apply the SFC/CFS/ED schemes based on the EKMR scheme. Then, we compare the performance of these three schemes with those based on the TMR scheme. Both theoretical analysis and experimental test were conducted. In theoretical analysis, we analyze the SFC/CFS/ED schemes based on the TMR/EKMR schemes in terms of the data distribution time and the data compression time. In experimental test, we implement these three schemes on an IBM SP2 parallel machine. The theoretical analysis and experimental results first show that the ED scheme is superior to the CFS scheme that is superior to the SFC scheme. Second, these three schemes based on the EKMR scheme outperform those based on the TMR scheme. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distributed server selection with imprecise state for replicated server group

    Page(s): 73 - 78
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1394 KB) |  | HTML iconHTML  

    Server replication is an approach often used to improve service access efficiency and availability for a large number of clients. This paper adopts a simple service model and investigates a set of efficient algorithms for server selection in an anycast group under the condition of imprecise network delay. The algorithms use ticket-based probing approach to search the path with short delay and probe the server load at the same time and try to optimize the server and path delay at the same time. The simulation results demonstrate that our algorithms are efficient in terms of server response time for sever selection compared with several other server selection algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new routing table update and ant migration scheme for ant based control in telecommunication networks

    Page(s): 508 - 513
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1399 KB) |  | HTML iconHTML  

    Swarm intelligence based algorithms for routing in telecommunication networks have gained considerable amount of attention in recent years. Swarm intelligence, demonstrated by natural biological swarms, exhibits numerous powerful features that are desirable in communication systems. Mobile agents having swarm intelligence similar to the real life ants have been employed to achieve a general, flexible and modular solution for routing in communication networks. In this paper, we propose mobile agents that are more advanced than those found in the previous literatures. Our proposal includes a new migration scheme for the ant-like mobile agents and a new routing table management scheme. Simulation results show that our proposed schemes can ultimately result in better load balancing and better performance within telecommunication networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new method for independent task scheduling in nonlinearly DAG clustering

    Page(s): 187 - 192
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1350 KB) |  | HTML iconHTML  

    For the parallel tasks represented by the directed acyclic graph (DAG), if it is linearly clustered, the ordering of the execution time of the tasks in each cluster is based on their arrows in the DAG. But for nonlinearly clustering, the ordering of the independent tasks in each cluster is not easily decided. Improper ordering of these independent tasks will greatly increase the scheduling length of the DAG. We discuss the shortcomings of current scheduling algorithms and the reason behind poor performance, and then propose some new node information to be extracted which is used by a new independent tasks scheduling algorithm based on the maximized parallelism degree (MPD). Experimental results show that the MPD algorithm can yield better performance than the previous algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IDR: an intrusion detection router for defending against distributed denial-of-service (DDoS) attacks

    Page(s): 581 - 586
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1376 KB) |  | HTML iconHTML  

    Distributed denial-of-service (DDoS) attack has turned into one of the major security threats in recent years. Usually the only solution is to stop the services or shut down the victim and then discard the attack traffic only after the DDoS attack characteristics (such as the destination ports of the attack packets) are known. In this paper, we introduce a generic DDoS attack detection mechanism as well as the design and setup of a testbed for performing experiments and analysis. Our results showed that the mechanism can detect DDoS attack. This enables us to proceed to the next steps of packet classification and traffic control. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic update of shortest path tree in OSPF

    Page(s): 18 - 23
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1424 KB) |  | HTML iconHTML  

    The shortest path tree (SPT) construction is a critical issue to the high performance routing in an interior network using link state protocols, such as open shortest path first (OSPF) and IS-IS. In this paper, we propose a new efficient algorithm for dynamic SPT update to avoid the disadvantages (e.g. redundant computation) caused by static SPT update algorithms. The new algorithm is based on the understanding of the update procedure to reduce redundancy. Only significant elements that contribute to the construction of new SPT from the old one are focused on. The efficiency of our algorithm is improved because it only pay attention to the edges really count for the update process. The running time for the proposed algorithm is maximum reduced, which is shown through experimental results. Furthermore, our algorithm can be easily generalized to solve the SPT updating problem in a graph with negative weight edges and applied to the scenario of multiple edge weight changes. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Algorithms for the problem of K maximum sums and a VLSI algorithm for the K maximum subarrays problem

    Page(s): 247 - 253
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1450 KB) |  | HTML iconHTML  

    Given an array of positive and negative values, we consider the problem of K maximum sums. When an overlapping property needs to be observed, previous algorithms for the maximum sum are not directly applicable. We designed an O(K * n) algorithm for the K maximum subsequences problem. This was then modified to solve the K maximum subarrays problem in O(K * n3) time. Finally, we present a VLSI K maximum subarrays algorithm with O(K * n) steps and a circuit size of O(n2), which is cost-optimal in parallelisation of the sequential algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deterministic small-world graphs and the eigenvalue power law of Internet

    Page(s): 374 - 379
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1416 KB) |  | HTML iconHTML  

    Many relevant real-life networks like the WWW, Internet, transportation and communication networks, or even biological and social networks can be modelled by small-world scale-free graphs. These graphs have strong local clustering (vertices have many mutual neighbors), a small diameter and a distribution of degrees according to a power law. On the other hand, the knowledge of the spectrum of a graph is important for the relation which the eigenvalues and their multiplicities have with relevant graph invariants and topological and communication properties such as diameter, bisection width, distances, connectivity, expansion, partitions, edge-loading distribution etc. In this paper we introduce a new family of deterministic small-world graphs, we determine analytically their spectra and we show how these graphs can model the eigenvalue power-law of the Internet network. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Container problem in substring reversal graphs

    Page(s): 563 - 568
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1452 KB) |  | HTML iconHTML  

    In this paper, we propose an algorithm that solves the container problem in n-substring reversal graphs in polynomial order time of n. Its correctness is proved and estimates of time complexity and sum of path lengths are given. We also report the results of computer experiment conducted to measure the average performance of our algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Effect of soft handoff on packet transmissions in cellular CDMA downlinks

    Page(s): 42 - 47
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1378 KB) |  | HTML iconHTML  

    Soft handoff and packet transmission scheduling are studied for cellular CDMA downlinks. A power distribution law is developed first in order to guarantee the required signal-to-interference ratio for each connection in soft handoff. Based on this, packet transmission scheduling for integrated voice and data traffic is proposed. Simulation results show that the proposed packet scheduling scheme using the derived power distribution law can significantly improve the voice connection reliability, increase the system capacity, and achieve high data packet transmission throughput. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.