Network computing

Transmission across neocortical synapses depends on the frequency of presynaptic activity (Thomson & Deuchars, 1994). Interpyramidal synapses in layer V exhibit fast depression of synaptic transmission, while other types of synapses exhibit facilitation of transmission. To study the role of dynamic synapses in network computation, we propose a unified phenomenological model that allows computation of the postsynaptic current generated by both types of synapses when driven by an arbitrary pattern of action potential (AP) activity in a presynaptic population. Using this formalism, we analyze different regimes of synaptic transmission and demonstrate that dynamic synapses transmit different aspects of the presynaptic activity depending on the average presynaptic frequency. The model also allows for derivation of mean-field equations, which govern the activity of large, interconnected networks. We show that the dynamics of synaptic transmission results in complex sets of regular and irr...

Self-organizing computational models with specific intracortical connections can explain many functional features of visual cortex, such as topographic orientation and ocular dominance maps. However, due to their computational requirements, it is difficult to use such detailed models to study large-scale phenomena like object segmentation and binding, object recognition, tilt illusions, optic flow, and fovea-periphery interaction. This paper introduces two techniques that make large simulations practical. First, a set of general linear scaling equations for the RF-LISSOM self-organizing model is derived and shown to result in quantitatively equivalent maps over a wide range of simulation sizes. This capability makes it possible to debug small simulations and then scale them up to larger simulations only when needed. The scaling equations also facilitate the comparison of biological maps and parameters between individuals and species with different brain region sizes. Second, the equations are combined into a new growing map method called GLISSOM, which dramatically reduces the memory and computational requirements of large self-organizing networks. With GLISSOM it should be possible to simulate all of human V1 at the single-column level using existing supercomputers, making detailed computational study of large-scale phenomena possible.

A distributed system is one in which components located at networked computers communicate and coordinate their actions only by passing messages. The research areas of multi-agent systems and distributed systems coincide, and form the research area of distributed agent computing. The study of multi-agent systems (MAS) focuses on systems in which many intelligent agents interact with each other. In many situations, agents have to locate resources involving some kind of matchmaking. In this paper, we will give an overview of different approaches of matchmaking processes. We are interested in large-scale networks where certain nodes, called consumers, have to locate available resources and other nodes, called producers, are looking for tasks to execute. For matchmaking to occur, some issues must be considered, such as how to enable agents to decompose their tasks and goals, how to enable agents to communicate and how to enable agents to represent and reason about the actions, plans, and knowledge of other agents in order to interact with them. Criteria such as the maximum number of matched tasks and resources, task allocation efficiency, resource usage efficiency and matching time must be considered. The choice of the matchmaking architecture will evidently impact important aspects such as task allocation, load balancing and routing as well as interaction between agents. In this paper we provide an overview of existing approaches such as centralized matchmaking and peer-to-peer communications and propose a framework that compares the different approaches in terms of performance, efficiency, communication cost and task complexity.

Chromium Renderserver (CRRS) is software infrastructure that provides the ability for one or more users to run and view image output from unmodified, interactive OpenGL and X11 applications on a remote, parallel computational platform equipped with graphics hardware accelerators via industry-standard Layer 7 network protocols and client viewers. The new contributions of this work include a solution to the problem of synchronizing X11 and OpenGL command streams, remote delivery of parallel hardware-accelerated rendering, and a performance analysis of several different optimizations that are generally applicable to a variety of rendering architectures. CRRS is fully operational, Open Source software.

Chromium Renderserver (CRRS) is software infrastructure that provides the ability for one or more users to run and view image output from unmodified, interactive OpenGL and X11 applications on a remote, parallel computational platform equipped with graphics hardware accelerators via industry-standard Layer 7 network protocols and client viewers. The new contributions of this work include a solution to the problem of synchronizing X11 and OpenGL command streams, remote delivery of parallel hardware-accelerated rendering, and a performance analysis of several different optimizations that are generally applicable to a variety of rendering architectures. CRRS is fully operational, Open Source software.

Chromium Renderserver (CRRS) is software infrastructure that provides the ability for one or more users to run and view image output from unmodified, interactive OpenGL and X11 applications on a remote, parallel computational platform equipped with graphics hardware accelerators via industry-standard Layer 7 network protocols and client viewers. The new contributions of this work include a solution to the problem of synchronizing X11 and OpenGL command streams, remote delivery of parallel hardware-accelerated rendering, and a performance analysis of several different optimizations that are generally applicable to a variety of rendering architectures. CRRS is fully operational, Open Source software.

Computer architects increasingly depend on the use of software tools to evaluate and investigate the design of computer systems. It is therefore very important that educators in this eld promote extensive tool-based student experimentation in architecture classes. However, the integration of today's complex architecture tools into curricula poses several challenges to an instructor, including management o f p o werful computing resources, software installation and maintenance, and development of tool-speci c educational material. This paper describes how these challenges are addressed by a universally accessible network-computing infrastructure -NETCARE -that provides educators with a Web portal to access computing resources, executable tools and educational material.

The computational science community has long been at the forefront of advanced computing, because of its persistent need to solve problems that require resources beyond those provided by the most powerful computers of the day. Examples of such high-end applications range from financial modeling and vehicle simulation to computational genetics and weather forecasting. Over the years, such considerations have led computational scientists to become aggressive and innovative adopters of vector computers, parallel systems, clusters, and other novel computing technologies. Recently, the widespread availability of high-speed networks and a growing awareness of the new problem solving modalities, made possible when these networks are used to couple geographically distributed resources, have stimulated interest in so-called Grid computing . The term "the Grid" refers to an emerging persistent infrastructure providing security, resource access, information, and other services that enable controlled and coordinated resource sharing among "virtual organizations" formed dynamically from individuals and organizations sharing a common interest . A variety of ambitious projects are applying Grid computing concepts to such challenging problems as the distributed analysis of experimental physics data, community access to earthquake engineering facilities, and the creation of "science portals"-thin clients providing remote access to the collection of information sources and simulation systems supporting a particular scientific discipline. Underpinning both parallel and Grid computing is a common need for coordination and communication mechanisms that allow multiple resources to be applied in a concerted fashion to complex problems. Scientific and engineering applications have, for the most part, addressed this requirement in an ad hoc and low-level fashion, using specialized message-passing libraries within parallel computers and various communication mechanisms among networked computers. While low-level approaches have allowed users to meet performance goals, an unfortunate consequence is that the computational science community has not benefited to any great extent from the significant advances in software engineering that have occurred during the past ten years in industry. In particular, the various benefits of Java, which seems ideal for multiparadigm communication environments, are hardly exploited at all. Java's platform-independent bytecode can be executed securely on many platforms, making Java an attractive basis for portable Grid computing. In addition, Java's performance on sequential codes, which is a strong prerequisite for the development of such "Grande" applications (see sidebar on Java Grande), has increased substantially over the past years . Inspired originally by coffee house jargon, the buzzword "Grande" has become commonplace in order to distinguish this emerging type of high-end applications when written in Java. 1 Furthermore, Java provides a sophisticated graphical user interface framework, as well as a paradigm to

A novel simulation framework is described that uses a client-server paradigm combined with a discrete event simulator to allow the simulation modules to be distributed around a centralized core. The framework also simplifies the creation of new simulations by achieving the goals of being general purpose, easily extensible, modular, and distributed. A prototype of the Client-Server Discrete Event Simulator (CS-DEVS) has been developed as a proof of concept. Two examples are presented to demonstrate the power of the new paradigm and prove the correctness of CS-DEVS. The first example is a modeling of the classical bank queuing simulation. The second example models the simulation of a distributed computing system, with complete modules for workload generation, architecture description, and scheduling and load balancing algorithms.

The Supercomputer Toolkit constructs parallel computation networks by connecting processor modules. These connections are set by the user prior to a run and are static during the run. The Technion's Toolkit prototype was used to run a simplified version of the PSU/NCAR MM5 mesoscale model . Each processor is assigned columns of the grid points of a square in the (x,y) space. When n × n columns are assigned to each processor its computation time is proportional to n 2 and its communication time to n. Since the Toolkit's network computes in parallel and communicates in parallel, then, for a given n, the total time is independent of the size of the two dimensional array or the area over which the weather prediction takes place. A mesoscale forecast over the eastern Mediterranean was run and measured; it suggests that were the Toolkit constructed from ALPHA processors, 10 processors would do a 36 h prediction in only about 13 minutes. A 36 hours prediction with full physics for the whole earth will require 2 hours for 80 ALPHA processors.

This paper introduces an analytical method for approximating the performability of a firm realtime system modeled by a multi-server queue. The service discipline in the queue is earliestdeadline-first (EDF), which is an optimal scheduling algorithm. Real-time jobs with exponentially distributed relative deadlines arrive according to a Poisson process. All jobs have deadlines until the end of service and are served non-preemptively. An important performance measure to calculate is the loss probability. The performance of the system is approximated by a Markovian model in the long run. A key parameter, namely, the loss rate when there are n jobs in the system is used in the model, which is estimated by partitioning the system into two subsystems. The resulting model can then be solved analytically using standard Markovian solution techniques. The number of servers in the system may change due to failure or repair. The performability of the system is evaluated in the presence of such structural changes. The latter measure is approximated by a Markov reward model, considering the loss probability as the reward rate. Comparing numerical and simulation results, we find that the existing errors are relatively small.

Some of the costs associated with system virtualization using a virtual machine monitor (VMM) or hypervisor (HV) can be attributed to the need to virtualize peripheral devices and to then manage the virtualized resources for large numbers of guest domains. For higher end, 'smart' IO devices, it is possible to offload selected virtualization functionality onto the devices themselves, thereby creating self-virtualizing devices. A self-virtualizing device is aware of the fact that it is being virtualized. It implements this awareness (1) by presenting to the HV an interface that enables the HV to manage virtual devices on demand, and (2) by presenting to a guest OS a virtual device abstraction that permits it to interact with the device with minimal HV involvement. This paper presents the design of a self-virtualizing network device and its implementation on a modern multi-core network platform, an IXP2400 network processor-based development board. Initial performance results demonstrate that the performance of virtual interfaces is similar to that of regular network interfaces. The longer term goal of this work is to better understand how modern multi-core platforms can be used to realize lightweight , scalable virtualization for future server platforms.

Internet has been growing at an amazing rate and it becomes pervasive in all aspects of our l i f e On the other hand, the ubiquity of networked computers and their services has signijicantly increased their vulnerability to virus and worm attacks. To make pei-vasive systems and their services reliable and secure it becomes highly essential to develop on-line monitoring, analysis, and quantijication of the operational state of such systems and services under a wide range of normal and abnormal workload scenarios. In this paper, we prevent several abnormality metrics that can be used to detect abnormal behaviors and ulso can be used to quanti& the impact of attacks on pervasive system services. Our online monitoring approach is based on deploying software agents on selected routers, clients and servers to continuously monitor the measurement attributes and compute the abnormality metrics. Further, we use this metrics to quanti& the impact of attacks on the individual components and on the system as a whole. This analysis leads to identi& the most critical components in the system. We have built a test bed to experiment and evaluate the effectiveness of these metrics to detect several well-known network attacks such as MS SQL slammer worm attack, Denial of Service attack, and email worm spam.

As mobile and embedded computing devices become more pervasive, it is becoming obvious that the nature of interactions between users and computers must evolve. Applications need to become increasingly autonomous and invisible, by placing greater reliance on knowledge of context and reducing interactions with users. Moreover, applications must cope with highly dynamic environments in which resources, such as network connectivity and software services, frequently vary over time. This paper presents our vision of pervasive computing and enumerates the software engineering challenges involved in realizing this vision. It also evaluates the current state of research and presents an agenda for future investigations in pervasive computing.

As the cost of computing technology is decreasing and networks & internetworks of computers expand in scope, the discovery and location of resources becomes an essential function of the networked computing environment. In light of the exponential growth of the global Internet, with increasingly large and dynamic computing domains, comes the challenge of scalable resource discovery, where client applications search for resources (i.e. services or devices etc.) on the network by describing some attribute of interest, instead of particular resource location. This paper presents a comparative study of various resource discovery, naming & resolution schemes used for heterogeneous networked environments. The schemes we found interesting to compare are:

We discuss the use of Condor, a distributed resource management system, as a provider of computational resources for NEOS, an environment for solving optimization problems over the Internet. We also describe how problems are submitted and processed by NEOS, and then scheduled and solved by Condor on available (idle) workstations

We discuss a novel nano-electronic computing paradigm in which cells composed of interacting quantum dots are employed in a locally-interconnected architecture. We develop a network-theoretic description in terms of appropriate local state variables in each cell.

A honey pot is a technique of cloud computing that is proposed for capturing hackers or tracking unusual methods of attack. This technique will seize, recognize and duplicate the hacker behavior. It works in a Cloud environment where anything like technology, tool, and the result can be offered as a service. Purveyors offer and deliver such services to their customers via the network. This paper presents the concept of a high-interaction honeypot, Kerberos authentication system as a service in a cloud environment to implement the benefits of, such service to ably distinguish between hackers and users and to provide overall security to the data/network.

Large scale High Performance Computing and Communication (HPCC) applications (e.g. Video-on-Demand, and HPDC) would require storage and processing capabilities which are beyond existing single computer systems. The current advances in networking technology (e.g. ATM) have made high performance network computing an attractive computing environment for such applications. However, using only high speed network is not su cient to achieve high performance distributed computing environment unless some hardware and software problems have been resolved. These problems include the limited communication bandwidth available to the application, high overhead associated with context switching, redundant data copying during protocol processing and lack of support to overlap computation and communication at application level. In this paper, we propose a Multithreaded Message Passing system for Parallel/Distributed processing, that we refer to as NYNET Communication System (NCS). NCS, being developed for NYNET (ATM wide area network testbed), is built on top of an ATM Application Programmer Interface (API). The Multithreaded environment allows applications to overlap computations and communications and provide a modular approach to support e ciently HPDC applications with di erent quality of service (QOS) requirements.

Background Phylogenetic profiles record the occurrence of homologs of genes across fully sequenced organisms. Proteins with similar profiles are typically components of protein complexes or metabolic pathways. Various existing methods measure similarity between two profiles and, hence, the likelihood that the two proteins co-evolve. Some methods ignore phylogenetic relationships between organisms while others account for such with metrics that explicitly model the likelihood of two proteins co-evolving on a tree. The latter methods more sensitively detect co-evolving proteins, but at a significant computational cost. Here we propose a novel heuristic to improve phylogenetic profile analysis that accounts for phylogenetic relationships between genomes in a computationally efficient fashion. We first order the genomes within profiles and then enumerate runs of consecutive matches and accurately compute the probability of observing these. We hypothesize that profiles with many runs are...

Computer security is of growing importance in the increasingly networked computing environment.This work examines the issue of high-performance network security, specifically integrity, by focusing on integrating security into network storage system. Emphasizing the cost-constrained environment of storage, we examine how current software-based cryptography cannot support storage's Gigabit/sec transfer rates. To solve this problem, we introduce a novel message authentication code, based on stored message digests. This allows storage to deliver high-performance, a factor of five improvement in our prototype's integrity protected bandwidth, without hardware acceleration for common read operations. For receivers, where precomputation cannot be done, we outline an inline message authentication code that minimizes buffering requirements.

Wireless Mobile Communication systems will be more sophisticated and wide spread in future. This growth demands not only for capacity increase, but also high quality of service and better coverage without increase in the radio frequency spectrum allocated for mobile applications. Smart antenna is one of the promising technologies for achieving efficient networks that maximize capacity and improve quality and coverage. These Smart antennas dynamically adapt to changing traffic requirements. The core of smart antenna is the selection of smart algorithms in adaptive array. Using beam forming algorithms, the weight of antenna arrays can be adjusted to form certain amount of adaptive beam to track corresponding users automatically and at the same time to minimize interference arising from other users by introducing nulls in their directions Smart antennas are usually employed at the base stations and they radiate narrow beams to serve different users. This work looked at the Smart Antenna technology and its principles of operation, and the associated benefits obtainable from its implementation in a mobile cellular network.

From the 1990s on, one of the most important challenges facing computer science researchers has been the design and construction of software tools to exploit Internet computing. At the same time, the development of agent technology has gone hand in hand with the explosion of the Internet. As worldwide network computing environments become more and more complex, software agents are believed to have the potential to help present and manage the Internet in an autonomous or semi-autonomous way. Yet, to date, a number of fundamental questions about the theory and practice of this new software engineering paradigm have remained unanswered. Here we will explore the features that make the agent-based approach such an appealing and evolutionary computational model. In particular, we envision a global agent-based distributed computing architecture that provides a convenient programming abstraction and sufficient transparency. This paper gives a general introduction to the underlying concepts of our research and development both at the level of design philosophy and in practical implementation techniques. It will be argued that the shift from computer networks to agent networks is a significant extension of network programming technology because agents are well suited to modeling, designing and implementing scalable, flexible and secure distributed systems over a worldwide computing environment.

Morphometric analysis is one of the important study in the natural sciences in general, and geomorphology and hydrology in particular, to describe the nature of surface drainage networks. The study area lies in north of Iraq, in Al-sulaymaniya province. With an area of (1545.1) Km². And the geographic coordinates of the study area is (35°.28`-35°.52` N). (45°.20`-46°.24` E). DEM image were used with (90 m) resolution and the drawing tools in ARC GIS program to delineate the total basin of the study area and watersheds. We conclude that there were 3 main watersheds in the study area and that group of parameters were calculated such as (linear, areal and relief properties)). The study shows that the stream density was (1.77) Km/Km 2 which mean that the study area has a bad discharge of water and sediments, the elongation in the study area is (0.55) and that mean the study area is closer to a rectangular shape, and the relief in the study area is (25.1) m/Km and it reflects that the study area had rich effects of erosion and weathering.

This paper introduces an analytical method for approximating the performance of a two-class priority M/M/1 system. The system is fully non-preemptive. More specifically, the prioritized class-1 jobs are real-time and served with the non-preemptive earliest-deadline-first (EDF) policy, but despite their priority cannot preempt any non real-time class-2 job. The waiting class-2 jobs can only be served from the time instant that no class-1 job is in the system. The service discipline of the class-2 jobs is FCFS. The required mean service times may depend on the class of the jobs. The real-time jobs have exponentially distributed relative deadlines until the end of service. The system is approximated by a Markovian model in the long run, which can be solved numerically using standard Markovian solution techniques. The performance measures of the system are the loss probability of the class-1 jobs and the mean sojourn (waiting) time of the class-2 jobs. Comparing the numerical and simulation results, we find that the existing errors are relatively small.

We consider transmission policies for multiple users sharing a single wireless link to a base station. The noise, and hence the probability of correct transmission of a packet, depends on the state of the user receiving the packet. The state for each user is independent of the states of the other users and changes according to a two-state (good/bad) Markov chain. The state of a user is observed only when it transmits. We give conditions under which the optimal policy is the myopic policy, in which a packet is transmitted to the user that is most likely to be in the better of the two states. We do this by showing that the optimal value function is marginally linear in each of the users' probabilities of being in the good state. Our model also may be applied to flexible manufacturing systems with unreliable tools and networked computer systems.

This FY03 final report on Wet Etch Figuring involves a 2D thermal tool. Its purpose is to flatten (0.3 to 1 mm thickness) sheets of glass faster thus cheaper than conventional sub aperture tools. An array of resistors on a circuit board was used to heat acid over the glass Optical Path Difference (OPD) thick spots and at times this heating extended over the most of the glass aperture. Where the acid is heated on the glass it dissolves faster. A self-referencing interferometer measured the glass thickness, its design taking advantage of the parallel nature and thinness of these glass sheets. This measurement is used in close loop control of the heating patterns of the circuit board thus glass and acid. Only the glass and acid were to be moved to make the tool logistically simple to use in mass production. A set of 4-circuit board, covering 80x80-cm aperture was ordered, but only one 40x40-cm board was put together and tested for this report. The interferometer measurement of glass OPD was slower than needed on some glass profiles. Sometimes the interference fringes were too fine to resolve which would alias the sign of the glass thickness profile. This also caused the phase unwrapping code (FLYNN) to struggle thus run slowly at times taking hours, for a 10 inch square area. We did extensive work to improve the speed of this code. We tried many different phase unwrapping codes. Eventually running (FLYNN) on a farm of networked computers. Most of the work reported here is therefore limited to a 10-inch square aperture. Researched into fabricating a better interferometer lens from Plexiglas so to have less of the scattered light issues of Fresnel lens groves near field scattering patterns, this set the Nyquest limit. There was also a problem with the initial concept of wetting the 1737 glass on its bottom side with acid. The wetted 1737 glass developed an Achromatic AR coating, spoiling the reflection needed to see glass thickness interference fringes. In response to this dilemma, the acid wetting was moved to the top of the glass and the thermal diffusion layer under the glass replaced with rubber. The bottom side of the glass was then protected from etching with a vacuum grease layer on the glass, though many other ways were tried and need to be further developed. We switched our mission away from flattening the OPD of 1737 glass, as the Eyeglass project no longer wanted glass. We turned our attention to the NIF Borofloat Debris shields.

Internet has been growing at an amazing rate and it becomes pervasive in all aspects of our l i f e On the other hand, the ubiquity of networked computers and their services has signijicantly increased their vulnerability to virus and worm attacks. To make pei-vasive systems and their services reliable and secure it becomes highly essential to develop on-line monitoring, analysis, and quantijication of the operational state of such systems and services under a wide range of normal and abnormal workload scenarios. In this paper, we prevent several abnormality metrics that can be used to detect abnormal behaviors and ulso can be used to quanti& the impact of attacks on pervasive system services. Our online monitoring approach is based on deploying software agents on selected routers, clients and servers to continuously monitor the measurement attributes and compute the abnormality metrics. Further, we use this metrics to quanti& the impact of attacks on the individual components and on the system as a whole. This analysis leads to identi& the most critical components in the system. We have built a test bed to experiment and evaluate the effectiveness of these metrics to detect several well-known network attacks such as MS SQL slammer worm attack, Denial of Service attack, and email worm spam.

Network-centric computing aims to solve the problems associated with traditional client/server systems, namely the high performance requirements, and costly maintenance of, the computing resources. With the network-centric model, applications are stored on servers and downloaded to clients as required, where they are then executed locally. User data is saved back to the remote server-local user data is only temporary. However these applications lack a standard file system where they can store their data files. For instance, when these applications are run as Java applets in a web browser, they are prevented from accessing the local storage system by the Java environment. This paper proposes the Java File System (JFS), a distributed file system for use by these applications, which provides strong security and client/server heterogeneity.

Resumo-O esquema de arbitragem tipicamente usado nos barramentos de campo "tempo-real" baseia-se num scheduler do tipo off-line que gera uma tabela estática e cíclica contendo a atribuição de tempo de barramento às transacções associadas a variáveis do processo. Esta abordagem, por exemplo usada no barramento de campo FIP (Factory Instrumentation Protocol) é altamente inflexível no sentido de que qualquer alteração ao sistema, tal como adicionar um sensor, obriga à interrupção do funcionamento do barramento. Neste artigo propomos a utilização de um scheduler do tipo planeamento para resolver esta inflexibilidade. Este scheduler representa uma situação de compromisso entre entre as vantagens e desvantagens do escalonamento (scheduling) dinâmico e estático típicos. É, também, apresentada uma condição suficiente de escalonabilidade (schedulability) que implica um custo de desempenho mínimo (overhead) e que, por essa razão, é adequada para a análise on-line. A possibilidade de utilização do scheduler de planeamento no contexto do barramento FIP também é descrita obtendo-se, desse modo, um barramento compatível tipo-FIP.

This report presents results of the San Gabriel Valley Smart Shuttle (SGVSS) Field Operational Test (FOT). Results are drawn from a comprehensive evaluation of the FOT conducted over a three-year period. The SGVSS attempted to integrate services of three local municipal public transit operators and a regional fixed route operator via networked computer-assisted dispatching, automated vehicle location, and mobile data

Classical client/server interaction primitives such as remote procedure call or rendezvous are not adequate when we need to describe the behaviour of three or more processes that need to collaborate simultaneously in order to solve a problem. Multiparty interactions are the key to describe these problems, and there are several languages that use them for the description of reactive systems. In this paper, we show and compare two different fair implementations of this mechanism and also outline the research we are carrying out in an effort to improve them.

The Mneme project is an investigation of techniques for integrating programming language and database features to provide better support for cooperative, information-intensive tasks such as computer-aided software engineering. The project strategy is to implement efficient, distributed, persistent programming languages. We report here on the Mneme persistent object store, a fundamental component of the project, discussing its design and initial prototype. Mneme storesobjectsin a simple and general format, preserving object identity and object interrelationships. Specific goals for the store include portability, extensibility (especially with respect to object management policies), and performance. The model of memory that the store aims at is a single, cooperatively-shared heap, distributed across a collection of networked computers. The initial prototype is intended mainly to explore performance issues and to support object-oriented persistent programming languages. We include perf...

This research focuses on the performance and timing behavior of a two level survivability architecture. The lower level of the architecture involves attack analysis based on kernel attack signatures and survivability handlers. Higher level survivability mechanisms are implemented using migratory autonomous agents. The potential for fast response to, and recovery from, malicious attacks is the main motivation to implement attack detection and survivability mechanisms at the kernel level. A timing analysis is presented that suggests the real-time feasibility of the two level approach. The limits to real-time response are identified from the host and network point of view. The experimental data derived is important for risk management and analysis in the presence of malicious network and computer attacks.

In June 1997 an industry consortium announced that it was working on a mobile extension of the Open Group's Network Computer Reference Profile [NCRP] to address the unique requirements of the many new mobile computing devices. (For participating firms, see the sidebar Consortium Members.)This specification will propose a set of standards for developers deploying a Java-based lightweight network computing solution for mobile use. It will also include a new set of trimmed-down Java APIs to support disconnected operation, secure remote access, power management and device adaptivity to different network environments. The Network Computer Reference Specification defined a Network Computer (or NC) as being lightweight, ubiquitous, extensible, secure and easy to administer and used widely deployed technologies such as HTTP, HTML and Java to ensure universality. The Mobile Network Computer Reference Specification [MNCRS] extends the concept of a network computer to define a Mobile Network Computer (or MNC). The consortium will produce open standards that specify APIs visible to applications, network protocols and server interactions. Naturally, these standards will have implications to software developers, original equipment manufacturers, operating system vendors and service providers. Since the intent is to enable MNCs and servers from a variety of manufacturers to interoperate, the consortium will adopt industry standards whenever possible, modifying existing technologies or inventing new ones only if necessary. Accordingly, ongoing convergence efforts with entities such as the NCRP and the Internet Engineering Task Force (http://www.ietf.org) are intended to avoid duplication of efforts in overlapping areas such as security, communications, tunneling and boot sequence. Mobile Network Computers differ from their stationary counterparts in several ways: Portable devices are resource poor, having less powerful CPUs, smaller screens, less storage, and so on. MNCs operate in a changing computing environment. Network characteristics like latency and bandwidth vary dramatically, including complete disconnection. Power resources and network resources like printers and directory servers also change. Users cannot count on the physical security of a locked office, so data and device security are imperative. Likewise, network security must protect data following unpredictable paths over potentially hostile networks (for example,. the Internet). The MNCRS defines a software platform. Nevertheless, three classes of devices are distinguished on the basis of their hardware characteristics: Class 1, professional assistants, are equivalent to a network computer with the additional flexibility of mobility and disconnected operations. These are expected to be similar to current laptops, but lighter and with a longer battery life. Class 2, information access devices, are limited to screen resolutions less than that of VGA, and offer only e-mail and perhaps some form of Web browsing. This class might include personal digital assistants (PDAs) as well as some high-end telephony devices. Class 3, basic messaging, paging or telephony devices, are limited to screen resolutions of less than 120 x 160 pixels.

We propose DSL, a new Scalable Distributed Data Structure for the dictionary problem, based on a version of Skip Lists, as an alternative to both random trees and deterministic height balanced trees. Our scheme exhibits, with high probability, logarithmic search time, constant reconstruction time, and linear space overhead. Additionally, at the expense of two additional pointers per internal node, the search operation could cost O log d¡ expected messages, where d is the distance between guessed and actual position.

Computer-supported cooperative work environments change some of the underlying beliefs about solutions that have been built for distributed computing. Electronic mail and file transfers have worked efficiently and effectively by breaking the information to be transferred into packets and reassembling the packets at the destination. This is not a viable solution for handling shared real time drawing or writing. Integrity has been maintained by locking out portions of a database until an update is completed. Such lockout is not always suitable in a groupware interface. Other solutions are necessary for resolving conflicts. Client-server architectures have worked well for managing distributed processing but replicated architectures. coupled with their fragility and synchronization problems are needed for groupware products in order to preserve acceptable local response time. The addition of multi-media to the environment complicates the problem more. Control of analog video requires a client-server environment which can build in intolerable delays as distances between communicating parties increase. This paper approaches design criteria for shared software from the human side and points out profound architectural problems that need to be solved if multi-media cooperative work environments are to function effectively. Resume Les environments de taches cooperatives mediatises par ordinateur changent certaines des croyances sur les solutions qui ont ete mise en oeuvre pour l'informatique distribuee. Le courrier electronique et le transfert de fichiers ont fonctionnes avec succes et efficacite en separant l'information en paquets qui sont reassembles a' l'arrivee. Cette solution n'est pas viable pour traiter la composition cooperative de textes ou de dessins en temps reel. L'integrite a ete maintenue en bloquant l'acces de portions de base de donnees jusqu'a la fin de leurs mise a jour. De tel bloc ages ne toujours conviennent pas pour une interface de informatique de groupe. Des solutions nouvelles sont necessaires pour resoudre ces conflits. Les architectures client-serveur fonctionnent bien pour gerer des processus distribues. Mais des architectures repliques. avec leurs fragilite et leurs problemes de synchronisation. sont necessaires aux produits groupware afin de preserver les temps de reponse locaux acceptables. L'addition de multimedia complique encore le probleme. Le controle de la video analogique demande un environement client-serveur qui cree des delais intolerables lorsque les distances entre interlocuteurs augmentent. Cet article presente des criteres de conception de informatique de groupe d'un point de vue utilisateur et fait apparaitre des problemes architecturaux profonds qui doivent etre Tt!solus si ron veut permettre aux environements cooperatifs multi-media de fonctionner de maniere efficace.

This study examined the pattern of use of different forms of contact between grandparents and grandchildren, and especially the use of new technologies (SMS, e-mail) and factors affecting this. Questionnaire data are reported from 408 grandparents in the UK, Spain, Finland and Estonia, regarding contacts with grandchildren mostly in the 10-15-year age range. Face-to-face contact remained the most frequent mean, followed closely by landline telephone; there was moderate use of mobile phones, and many used letters/cards occasionally; and a minority used SMS and e-mails (about one-half to one-third of those with mobile phones, and networked computers, respectively). When contacting grandchildren, most grandparents accumulate different forms of contact, but others compensate some forms of contact. There were no differences by age of grandparent, but grandmothers made more use of e-mail than grandfathers, as did more highly educated grandparents and those with older grandchildren. Implications for use of Information and Communication Technology by older people are discussed.

Large scale High Performance Computing and Communication (HPCC) applications (e.g. Video-on-Demand, and HPDC) would require storage and processing capabilities which are beyond existing single computer systems. The current advances in networking technology (e.g. ATM) have made high performance network computing an attractive computing environment for such applications. However, using only high speed network is not su cient to achieve high performance distributed computing environment unless some hardware and software problems have been resolved. These problems include the limited communication bandwidth available to the application, high overhead associated with context switching, redundant data copying during protocol processing and lack of support to overlap computation and communication at application level. In this paper, we propose a Multithreaded Message Passing system for Parallel/Distributed processing, that we refer to as NYNET Communication System (NCS). NCS, being developed for NYNET (ATM wide area network testbed), is built on top of an ATM Application Programmer Interface (API). The Multithreaded environment allows applications to overlap computations and communications and provide a modular approach to support e ciently HPDC applications with di erent quality of service (QOS) requirements.

results in high latency and low bandwidth to web-enabled clients and prevents the timely delivery of software. We present an algorithm for modifying delta compressed files so that the compressed versions may be reconstructed without scratch space. This allows network clients with limited resources to efficiently update software by retrieving delta compressed versions over a network. Delta compression for binary files, compactly encoding a version of data with only the changed bytes from a previous version, may be used to efficiently distribute software over low bandwidth channels, such as the Internet. Traditional methods for rebuilding these delta files require memory or storage space on the target machine for both the old and new version of the file to be reconstructed. With the advent of network computing and Internet-enabled devices, many of these network attached target machines have limited additional scratch space. We present an algorithm for modifying a delta compressed version file so that it may rebuild the new me version in the space that the current version occupies. Differential or delta compression [5, 11, compactly encoding a new version of a file using only the changed bytes from a previous version, can be used to reduce the size of the file to be transmitted and consequently the time to perform software update. Currently, decompressing delta encoded files requires scratch space, additional disk or memory storage, used to hold a required second copy of the file. Two copies of the compressed file must be concurrently available, as the delta file contains directives to read data from the old file version while the new file version is being materialized in another region of storage. This presents a problem. Network attached devices often have limited memory resources and no disks and therefore are not capable of storing two file versions at the same time. Furthermore, adding storage to network attached devices is not viable, as keeping these devices simple limits their production costs.

In-place reconstruction of delta compressed data allows information on devices with limited storage capability to be updated efficiently over low-bandwidth channels. Delta compression encodes a version of data compactly as a small set of changes from a previous version. Transmitting updates to data as delta versions saves both time and bandwidth. In-place reconstruction rebuilds the new version of the data in the storage or memory space the current version occupies-no additional scratch space is needed. By combining these technologies, we support large-scale, highly-mobile applications on inexpensive hardware. We present an experimental study of in-place reconstruction algorithms. We take a datadriven approach to determine important performance features, classifying files distributed on the Internet based on their in-place properties, and exploring the scaling relationship between files and data structures used by in-place algorithms. We conclude that in-place algorithms are I/O bound and that the performance of algorithms is most sensitive to the size of inputs and outputs, rather than asymptotic bounds.

In Distributed Interactive Applications (DIAs) embedded priority, the sender node forwards data to receivers due to their respective priorities. One application level multicast protocol: the priority-based directed minimum spanning tree (PST) was designed for these DIAs. However, the existing PST has no efficient priority selection mechanism and the system consumes a tremendous amount of resource and becomes unstable and unscalable when its size increases. In this paper we propose a novel application level multicast algorithm: Predict-and-Quantize for Priority with directed minimum Spanning Tree (PQPST), which can efficiently predict priorities for the receivers and quantize the predicted priorities to build a multicast distribution tree. Furthermore, PQPST can easily integrate the quantized priority into game environment and simplify priority deployment. The simulation results show that PQPST is able to efficiently make priority selection and keep system stable with reasonable scalability.

Distributed Embedded Systems (DESs) carrying out critical tasks must be highly reliable and hard in real-time. Moreover, to operate in dynamic operational contexts in an effective and efficient manner, they must also be adaptive. Adaptivity is particularly interesting from a dependability perspective, as it can be used to develop dynamic fault tolerance mechanisms, which, in combination with static ones, make it possible to provide better and more efficient fault tolerance. However, constructing a DES with such complexity presents many challenges. This is because all the mechanisms that support fault tolerance, real-time, and adaptivity must be designed to operate in a coordinated manner. This paper presents the Dynamic Fault Tolerance for Flexible Time-Triggered Ethernet (DFT4FTT), a self-reconfigurable infrastructure for implementing highly reliable adaptive DES. Here, we describe the design of its hardware and software architecture and the main set of mechanisms, with a focus on ...

Recently, unmanned aerial vehicles (UAVs) enhance connectivity and accessibility for civilian and military applications. A group of UAVs with on-board cameras usually monitors or collects information about designated areas. The UAVs can build a distributed network to share/exchange and to process collected sensing data before sending to a data processing center. A huge data transmission among them may cause latency and high-energy consumption. This paper deploys artificial intelligent (AI) techniques to process the video data streaming among the UAVs. Thus, each distributed UAV only needs to send a certain required information to each other. Each UAV processes data utilizing AI and only sends the data that matters to the others. The UAVs, formed as a connected network, communicate within a short communication range and share their own data to each other. Convolution neural network (CNN) technique extracts feature from images automatically that the UAVs only send the moving objects i...

The widely used Message Passing Interface (MPI) is complex and rich. As a result, application developers require automated tools to avoid and to detect MPI programming errors. We present the Marmot Umpire Scalable Tool (MUST) that detects such errors with significantly increased scalability. We present improvements to our graph-based deadlock detection approach for MPI, which cover future MPI extensions. Our enhancements also check complex MPI constructs that no previous graph-based detection approach handled correctly. Finally, we present optimizations for the processing of MPI operations that reduce runtime deadlock detection overheads. Existing approaches often require O(p) analysis time per MPI operation, for p processes. We empirically observe that our improvements lead to sub-linear or better analysis time per operation for a wide range of real world applications.