Showing papers on "Server published in 1992"

Stability properties of constrained queueing systems and scheduling policies for maximum throughput in multihop radio networks

Abstract: The stability of a queueing network with interdependent servers is considered. The dependency among the servers is described by the definition of their subsets that can be activated simultaneously. Multihop radio networks provide a motivation for the consideration of this system. The problem of scheduling the server activation under the constraints imposed by the dependency among servers is studied. The performance criterion of a scheduling policy is its throughput that is characterized by its stability region, that is, the set of vectors of arrival and service rates for which the system is stable. A policy is obtained which is optimal in the sense that its stability region is a superset of the stability region of every other scheduling policy, and this stability region is characterized. The behavior of the network is studied for arrival rates that lie outside the stability region. Implications of the results in certain types of concurrent database and parallel processing systems are discussed. >

Method and apparatus for message handling in computer systems

Abstract: An Integrated Messaging System which integrates mail from a plurality of mail servers handling messages of different media types such as text, voice, facsimile, video and image. The IMS maintains the in-basket for all mail systems, eliminating the need to collect each type of mail separately. Any terminal interface, telephone or computer workstation can be used to collect, generate and act upon a message of any media type. Further, from the preferred system, the user can request another file server and its associated terminals to perform messaging services. The system includes a variety traditional E-mail and voice mail functions which can be used for any message. The IMS also includes synchronization means which checks to see whether the mail count in each in-basket is the same to guarantee that the same mail items are in each file server in-basket. An integrated messaging protocol is used to pass information between the file servers. It is designed in a generic manner to allow wide variety of foreign message formats at each of the respected media file server. For the majority of transactions, the protocol is independent of message body format, referencing only the message pointer or header.

Cooperative processing interface and communication broker for heterogeneous computing environments

Abstract: An object interface is disclosed that supports three modes of inter-object communication--message processing (store and forward), conversational communication, and remote procedure call. A service broker manages service requests from, and responsive services provided by, a plurality of clients and servers, respectively, which may reside on different hardware platforms and operating systems and may be connected to computer networks having different network architectures and associated communications protocols. The broker manages the service offerings from servers and service requests from clients, and clients and servers communicate and exchange information with one another via the broker. The service broker includes different application programming interfaces for allowing participants to access the functionality of the service broker.

Access control subsystem and method for distributed computer system using locally cached authentication credentials

Abstract: A distributed computer system has a number of computers coupled thereto at distinct nodes. The computer at each node of the distributed system has a trusted computing base that includes an authentication agent for authenticating requests received from principals at other nodes in the system. Requests are transmitted to servers as messages that include a first identifier provided by the requester and a second identifier provided by the authentication agent of the requester node. Each server process is provided with a local cache of authentication data that identifies requesters whose previous request messages have been authenticated. When a request is received, the server checks the request's first and second identifiers against the entries in its local cache. If there is a match, then the request is known to be authentic. Otherwise, the server node's authentication agent is called to obtain authentication credentials from the requester's node to authenticate the request message. The principal identifier of the requester and the received credentials are stored in a local cache by the server node's authentication agent. The server process also stores a record in its local cache indicating that request messages from the specified requester are known to be authentic, thereby expediting the process of authenticating received requests.

Network Time Protocol (Version 3) Specification, Implementation

Abstract: This document describes the Network Time Protocol (NTP), specifies itsformal structure and summarizes information useful for itsimplementation. NTP provides the mechanisms to synchronize time andcoordinate time distribution in a large, diverse internet operating atrates from mundane to lightwave. It uses a returnable-time design inwhich a distributed subnet of time servers operating in a self-organizing, hierarchical-master-slave configuration synchronizes localclocks within the subnet and to national time standards via wire orradio. The servers can also redistribute reference time via localrouting algorithms and time daemons.

A reservation principle with applications to the ATM traffic control

Abstract: In a network based upon the Asynchronous Transfer Model (ATM), stepwise variable bit rate traffic sources can be efficiently multiplexed without significant degradation of the transfer equality by introducing an access control at an intermediate burst (step) level. This multiplexing is based upon a peak bit rate reservation in the switching elements of the general network. Two Fast Reservation Protocols (FRPs) are presented based upon mono-cell bandwidth management messages processed “on the fly” in the switching elements. The first protocol is FRP/DT which could be run to multiplex file banking calls and LAN-to-LAN interconnections using leased lines. The second is FRP/IT and could be run to multiplex connectionless servers and some real-time codecs. Evaluation of the hardware complexity and performance are addressed.

The RightPages image-based electronic library for alerting and browsing

Abstract: The RightPages electronic library prototype system, which gives users full online library services, is described. The prototype takes advantage of fast hardware, multimedia workstations, and broadband networks to process scientific and technical journals for users and to offer a service that: alerts them to the arrival of new journal articles matching their interest profiles; lets them immediately examine images of pages in the alerted articles and browse through other articles in the database; and enables them to order paper copies of any articles in the database. The system runs on a local area network that connects one or more scanning stations, a centralized document database server and multiple user stations running X Windows servers. The RightPages interface runs as an X Windows application on Sun workstations or X terminals. The system's image and document processing, including noise reduction, document layout analysis, text processing, and display processing are discussed. >

Understanding the Efficiency of Multi-Server Service Systems

Abstract: In the design and operation of service systems, it is important to determine an appropriate level of server utilization the proportion of time each server should be working. In a multi-server queue with unlimited waiting space, the appropriate server utilization typically increases as the number of servers and the arrival rate increases. We explain this economy of scale and give a rough quantitative characterization. We also show how increased variability in the arrival and service processes tends to reduce server utilization with a given grade of service. As part of this analysis, we develop simple approximations for the mean steady-state waiting time and the full steady-state waiting-time distribution. These approximations exploit an infinite-server approximation for the probability of delay and a single-server approximation for the conditional waiting-time distribution given that waiting occurs. The emphasis is on simple formulas that directly convey understanding.

Power Programming with RPC

Abstract: Preface Scope of This Handbook Some Assumptions About the Reader Software Mentioned in This Book Examples Additional Sources of Information Conventions Acknowledgments Chapter 1: Introduction to Remote Procedure Calling Client /Server Computing Remote versus Local Procedure Calling RPC Application Development Defining the Protocol Developing Server and Client Application Code Compiling and Running the Application RPC Makes Interprocess Communications Less Painful How RPC Systems Work RPC Systems and the OSI Reference Model What is "State" and Why is it Important? Top Twenty Distributed Computing Terms and Acronyms Chapter 2: Network Computing Today Distributed Computing Standards Features of a Distributed Computing Environment NCS ONC Examining ONC and NCS Machine-independent Data Representation RPC Mechanism Protocol Compilers: The Most Important Feature Authentication Services Network Resource Naming Services Network Time Service Distributed File System There May be a Common Application Environment in Your Future Netwise: A Second ONC Alternative Machine-independent Data Representation RPC Mechanism Protocol Compiler Authentication Services Network Resource Naming Services Network Time Service Distributed File System Summary of Distributed Computing Environments Rapid Evolution Which One Should You Choose? Put the ONC Suite on Your Machine for Free Chapter 3: Developing High-level RPC Applications Development Overview From Local to Remote Directory Reading Define the Protocol Data Types Program, Procedure, and Version Numbers High-level ONC RPC Library Calls At the Server At the Client Shared XDR Routines Using the XDR Library Writing the Client and Server Programs At the Server At the Client Compile, Link, and Run Why High-level Calls and Not Protocol Compiling? Some Limitations of the High-level Calls TCP Transport Requires Lower-level Calls Chapter 4: Protocol Compiling and Lower-level RPC Programming Development Overview Filename Conventions and Make Using RPCGEN The Protocol Definition Language Definitions Symbolic Constants Enumerations Structures Unions Typedefs Programs Declarations Special Cases Booleans Strings Opaque Data Voids Preprocessor Symbols and Control Lower-level ONC RPC Library Calls At the Client At the Server An Example: One Client Talks to One Server Debugging Step 1: Debug Without the Network Step 2: Use the Raw Transport Step 3: Debug Over the Network Deploying Servers During Development Real RPC Power Means Using IPC Chapter 5: UNIX Networking and Interprocess ommunication A Network Protocol Primer Internet Addressing Connectionless at the Lowest Level Internet Protocols Applications Protocols Connections and Well-known Ports Types of Servers Network Transport Selection: UDP or TCP? Adding a Server to the System inetd and Other Lurking Network Daemons Installing a Server Yourself A Digression on Remote Execution Daemons Configuring inetd An Overview of UNIX Interprocess Communication Pipes and FIFOs Message Queues, Semaphores, and Shared Memory Berkeley Sockets and System V TLI Data Representation or Byte Ordering Retrieving Host, Network, and Address Information Getting ONC RPC Information Berkeley Sockets Yeah, But How Do I Use Sockets? Socket Examples Advanced Socket Programming Issues Remote Execution, Security, and Authentication Chapter 6: Application Development: Networked Parallel Image Processing Developing Parallel Algorithms for a Multi -server Network A Simple Model For Parallel Processing on a Network System Requirements and Network Constraints Server Access Server and Network Performance Brute-force Scheduling Using Process Control Programming Asynchronous and Concurrent Processing at the Client Making Use of Timers to Watch Child Processes Development Steps Remote Image Processing (RIP) Specifying Filter Coefficients rip Development: It's as Easy as 1 2 3 Step 1: Define the Protocol Step 2: Building the Client Procedure Step 3: Developing the Server Procedure Testing and Running the Program Extending RIP Fast SunView Frame Buffer Access Needed Chapter 7: Distributing Existing Applications A Local Image Manager The Header File Modularity Functions Compile, Link, and Run Moving a Local Application to the Network Answer Fourteen Questions First The Strategy Chapter 8: Managing RPC Servers How to Start the Remote Server Shell Scripts For Starting Servers Starting a Remote Server From within Your Client Application Terminating Your Services System Error: "STF-E-OPENIN, Server Too Fat" Hit Reset Report Server Information with rpcinfo Changes Under TIRPC Data Sharing: NFS versus Sending it Yourself Host-qualified Filenames Chapter 9: Multiple Clients and Servers Remote Asynchronous Calls, Multi-server Processing ONC RPC Support The Follow-up RPC Multi-tasking at the Server Multi-tasking with Child Processes Alternatives to Avoid Run-time Process Creation Combining Asynchronous and Multi-tasking at the Server Lightweight Processing Remote Asynchronous Calls with LWP A Minimal Set of LWP Routines Client Multi-server Example Server Multi-tasking Example Caveats Chapter 10: RPC Under Windowing Systems The X Window System X Toolkit Client Application Flow of Control Low-level RPC and X Protocol Similarities Strategies for Using RPC Under X Placing and Servicing RPCs in an Event-driven Environment RPC and XView Synchronous RPCs with a Timer Remote Asynchronous Calls, FRPC Polled with a Timer The Event Notifier and Associated Complications Remote Asynchronous Call Servers Using notify_enable_rpc_svc() RPC and Xol/Xt Watching IPC with XtAppAddInput(3Xt) Remote Asynchronous Calls, Servers Using XtAppAddInput() Comparing Network Windowing Systems and RPC A Digression: Performance of Typical versus Network Windowing Systems Windowing System Evolution Can Hide a Frame Buffer X11 Pixmaps versus SunView Pixrects Augment Typical Windowing Systems with RPCs Chapter 11: ONC Transport-independent RPC Maintains the ONC RPC Protocol Run-time Transport Independence Network Selection Uniform Addressing TIRPC API Backward Compatibility Levels of the Library An Example Availability Chapter 12: Advanced Programming Issues Authentication and How to Use It ONC RPC Credentials and Verification at the Client Adding Authentication to the dim Client Authentication at the Server Adding Access Control to the dim Server Error Reporting Summarized Fault Tolerance, Connection Errors, and Crash Recovery Connection Errors and Recovery Caching Replies at the Server Broken Connections and Testing Crash Recovery The ONC RPC Programming Reference Section One: ONC XDR Library Routines Section Two: ONC Portmap Library Routines Section Three: ONC RPC Library Routines Section 1: ONC XDR Library Routines Overview XDR Streams and Their Management Conversion Filters Synopsis xdr_array() xdr_bool() xdr_bytes() xdr_char() xdr_destroy() xdr_double() xdr_enum() xdr_ float() xdr_ free() xdr_ getpos() xdr_inline() xdr_int() xdr_long() xdrmem_create() xdr_opaque() xdr_ pointer() xdrrec_create() xdrrec_endofrecord() xdrrec_eof() xdrrec_readbytes() xdrrec_skiprecord() xdr_reference() xdr_setpos() xdr_short() xdrstdio_create() xdr_string() xdr_u_char() xdr_u_int() xdr_u_long() xdr_union() xdr_u_short() xdr_vector() xdr_void() xdr_wrapstring() Section 2: ONC Portmap Library Routines Overview Synopsis pmap_ getmaps() pmap_ getport() pmap_rmtcall() pmap_set() pmap_unset() xdr_ pmap() xdr_ pmaplist() Section 3: ONC RPC Library Routines Overview Functional Summary Building Client Authentication Making the Call from the Client CLIENT Handle Management Server Registration With The Portmap SVCXPRT Service Transport Handle Management Server Side Error Handling And Reporting Server I/O and Utility Direct XDR Access Making Secure RPCs Synopsis authdes_create() authdes_ getucred() auth_destroy() authnone_create() uthunix_create() authunix_create_default() callrpc() clnt_broadcast() clnt_call() clnt_control() clnt_create() clnt_create_vers() clnt_destroy() clnt_ freeres() clnt_ geterr() clnt_ pcreateerror() clnt_ perrno() clnt_ perror() clntraw_create() clnt_spcreateerror() clnt_sperrno() clnt_sperror() clnttcp_create() clntudp_bufcreate() clntudp_create() get_myaddress() getnetname() getrpcport() host2netname() key_decryptsession() key_encryptsession() key_ gendes() key_setsecret() netname2host() netname2user() registerrpc() rpc_createerr svc_destroy() svcfd_create() svc_ fds svc_ fdset svc_ freeargs() svc_ getargs() svc_ getcaller() svc_ getreq() svc_ getreqset() svcerr_auth() svcerr_decode() svcerr_noproc() svcerr_noprog() svcerr_ progvers() svcerr_systemerr() svcerr_weakauth() svcraw_create() svc_register() svc_run() svc_sendreply() svctcp_create() svcudp_create() svcudp_bufcreate() svc_unregister() user2netname() xdr_accepted_reply() xdr_authunix_ parms() xdr_callhdr() xdr_callmsg() xdr_opaque_auth() xdr_rejected_reply() xdr_replymsg() xprt_register() xprt_unregister() Error Codes Appendix A: Obtaining RFCs (Internet Standards, Request for Comment) Appendix B: An RPC Case Study: Networked Ray Tracing Introduction to Ray Tracing Accelerating Ray Tracing Multiple Processor Ray Tracing and Data Distribution Networked Ray Tracing Using RPC Dynamic Scheduling and Load Balancing But Why Scan-line Parallelism and Not Frame Parallelism? Performance Results Conclusions References Documentation and Source Code Comments on Augmenting the Client to Run Under X11 with the XView Toolkit Appendix C: Generalized Server Initialization, Inquiry, and Removal Appendix D: Parallel Processing In A Nutshell Parallelism Interprocess Control Interprocess Communication Buzzwords: A Glossary Index

Optimal service speeds in a competitive environment

Abstract: This is a study of the economic behavior of vendors of service in competition. A simple model with two competing exponential servers and Poisson arrivals is considered. Each server is free to choose his own service rate at a cost per time unit that is strictly convex and increasing. There is a fixed reward to a server for each customer that he serves. The model is designed to study one specific aspect of competition, namely, competition in speed of service as a means for capturing a larger market share in order to maximize long-run expected profit per time unit. A two-person strategic game is formulated and its solutions are characterized. Depending on the revenue per customer served and on the cost of maintaining service rates, the following three situations may arise: i a unique symmetric strategic Nash equilibrium in which expected waiting time is infinite; ii a unique symmetric strategic equilibrium in which expected waiting time is finite; and iii several, nonsymmetric strategic equilibria with infinite expected waiting time. An explicit expression for the market share of each server as a function of the service rates of the two servers is also given.

An analysis of wide-area name server traffic: a study of the Internet Domain Name System

Abstract: Over a million computers implement the Internet's Domain Name System of DNS, making it the world's most distributed database and the Internet's most significant source of wide-area RPC-like traffic. Last year, over eight percent of the packets and four percent of the bytes that traversed the NSFnet were due to DNS. We estimate that a third of this wide-area DNS traffic was destined to seven root name servers. This paper explores the performance of DNS based on two 24-hour traces of traffic destined to one of these root name servers. It considers the effectiveness of name caching and retransmission timeout calculation, shows how algorithms to increase DNS's resiliency lead to disastrous behavior when servers fail or when certain implementation faults are triggered, explains the paradoxically high fraction of wide-area DNS packets, and evaluates the impact of flaws in various implementations of DNS. It shows that negative caching would improve DNS performance only marginally in an internet of correctly implemented name servers. It concludes by calling for a fundamental change in the way we specify and implement future name servers and distributed applications.

On-line load balancing

Abstract: The setup for the authors' problem consists of n servers that must complete a set of tasks. Each task can be handled only by a subset of the servers, requires a different level of service, and once assigned can not be re-assigned. They make the natural assumption that the level of service is known at arrival time, but that the duration of service is not. The on-line load balancing problem is to assign each task to an appropriate server in such a way that the maximum load on the servers is minimized. The authors derive matching upper and lower bounds for the competitive ratio of the on-line greedy algorithm for this problem, namely /sup (3n)2/3///sub 2/(1+o(1)), and derive a lower bound, Omega ( square root n), for any other deterministic or randomized on-line algorithm. >

TCP/IP Network Administration

Abstract: TCP/IP Network Administration, Third Edition is a complete guide to setting up and running a TCP/IP network for administrators of networks of systems or users of home systems that access the Internet. It starts with the fundamentals: what the protocols do and how they work, how addresses and routing are used to move data through the network, and how to set up your network connection. Beyond basic setup, this new second edition discusses advanced routing protocols (RIPv2, OSPF, and BGP) and the gated software package that implements them. It also provides a tutorial on how to configure important network services, including PPP, SLIP, sendmail, Domain Name Service (DNS), BOOTP and DHCP configuration servers, and some simple setups for NIS and NFS. There are also chapters on troubleshooting and security. In addition, this book is a command and syntax reference for several important packages, including gated, pppd, named, dhcpd, and sendmail.

System and method for remote program load

Abstract: An apparatus and method for retrieving and sending the bootstrap loader and the DOS code from disk storage on a file server in a network to a workstation in the network during the boot process of the workstation where there are one or more file servers in the network and where the file servers may be different types of computer systems including mainframe computers.

Method and apparatus for resource arbitration

Abstract: The present invention provides a rapid one-to-one match between requesters that must arbitrate for service from one of a number of servers. Each requester presents a set of requests, and the requesters are indifferent to which server is chosen, no priority existing among the requests seen by a particular server. Requests are presented synchronously to all servers to which access is desired. Each server selects precisely one such request, preferably randomly, and asserts a response signal so stating to all requesters. Each requester then selects precisely one incoming grant responses (if any there are), and de-asserts requests to all other servers. This iteration is repeated for a predetermined number of cycles, at which time substantially most of the requested matches will have been made. The iteration algorithm is preferably implemented with choice units, multiplexers, registers and logic units, all of which may be obtained commercially.

Extensions to ANSA for multimedia computing

Abstract: Distributed systems has been an area of intense activity over the past ten years. The subject is entering a stage of maturity with work now focussing on standards for Open Distributed Processing (ODP). However, it is important that standardisation is responsive to new demands on the technology. This is especially true with the emergence of high speed networks in general and multimedia computing in particular. This paper describes research into handling multimedia in the Advanced Networked Systems Architecture (ANSA), a state of the art distributed platform. It is shown that ANSA, although comprehensive in scope, fails to explicitly address multimedia requirements. A number of specific extensions are proposed to remedy these omissions. These include streams, as an abstraction over multimedia protocols, and chains, as a generic control interface to multimedia devices (including storage servers). Importantly, the proposed extensions do not affect the core ANSA architecture, but instead add the necessary functionality in terms of additional services.

The influence of scale on distributed file system design

Abstract: The proposition that scale should be recognized as a primary factor influencing the architecture and implementation of distributed systems is validated using Andrew and Coda, two distributed file systems. Performance, operability, and security are dominant considerations in the design of these systems. Availability is a further consideration the design of Coda. Client caching, bulk data transfer, token-based mutual authentication and hierarchical organization of the protection domain have emerged as mechanisms that enhance scalability. The separation of concerns made possible by functional specialization has also proved valuable in scaling. Heterogeneity is an important by-product of growth, but the mechanisms available to cope with it are rudimentary. Physical separation of clients and servers turns out to be a critical requirement for scalability. >

A weak-consistency architecture for distributed information services

Abstract: Services provided on wide-area networks like the Internet present several challenges. The reliability, performance, and scalability expected of such services often requires they be implemented using multiple, replicated servers. One possible architecture implements the replicas as a {\em weak-consistency process group.} This architecture provides good scalability and availability, handles portable computer systems, and minimizes the effect of users on each other. The key principles in this architecture are component independence, a process group protocol that provides small summaries of database contents, caching database {\em slices}, and the {\em quorum multicast} client-to-server communication protocol. A distributed bibliographic database system serves as an example.

The competitiveness of on-line assignments

Abstract: Consider the on-line problem where a number of servers are ready to provide service to a set of customers. Each customer's job can be handled by any of a subset of the servers. Customers arrive one-by-one and the problem is to assign each customer to an appropriate server in a manner that will balance the load on the servers. This problem can be modeled in a natural way by a bipartite graph where the vertices of one side (customers) appear one at a time and the vertices of the other side (servers) are known in advance. We derive tight bounds on the competitive ratio in both deterministic and randomized cases. Let n denote the number of servers. In the deterministic case we provide an on-line algorithm that achieves a competitive ratio of k = [log2 n] (up to an additive 1) and prove that this is the best competitive ratio that can be achieved by any deterministic on-line algorithm. In a similar way we prove that the competitive ratio for the randomized case is k=ln(n) (up to an additive 1). We conclude that for this problem, randomized algorithms differ from deterministic ones by precisely a constant factor.

Broadband multimedia applications using ATM networks: high-performance computing, high-capacity storage, and high-speed communication

Abstract: The authors discuss the classification of a large variety of multimedia communication applications (with the emphasis on broadband and hypermedia applications), the requirements at network levels, multimedia communications concepts with suitable platforms, and the modular standardization of multimedia teleservices and the corresponding technical systems. Appropriate technologies for meeting multimedia requirements comprise public and private networks, servers for information storage and processing, intelligent user and system agents, and multimedia terminals, e.g., workstations. The authors deal with the question of service-specific or service-integrated networks, the architecture, components, and functions of asynchronous transfer model (ATM) switching systems for broadband ISDN and ATM-based signaling, and management for controlling connectivity and interactivity. Network evolution, pilot applications, and coordinated standardization are briefly considered. >

Network redirection for the enhancement of server capacities

Abstract: A network architecture and methodology provides two stages or levels of network redirection to allow effective importation of storage into either or both of a client/requester and a server. Each level of redirection is achieved through the establishment of a bi-directional communication link, preferably by the creation of an access list or export file at each host containing a portion of a distributed resource. By using two levels of network redirection, a terminal communicates only with a single, predetermined server, thus permitting the use of user-friendly Lan server systems. The predetermined server communicates with the remainder of the network as a client through the network communication protocols, thus removing constraints on operating systems which may be running at hosts. Full access to the distributed resource is thereby made transparent to the user and operational limitations engendered by limitations of hard disk capacity at terminals and servers are overcome.

Optimal workload allocation in open networks of multiserver queues

Abstract: In this paper, we examine the general problem of workload allocation in an open Jackson network of multiserver queues. We show that use of the open network model leads to a separable, convex formulation of the problem with relatively simple optimality conditions. Using these conditions, we prove in general that server groups with the same number of servers should be loaded equally and larger groups should be loaded more heavily than smaller groups. It is also shown that server pooling, combining servers into larger groups, will always reduce congestion and/or increase throughput. We discuss the significance of our results for job shop applications and also for the concept of a production bottleneck. In systems with an unbalanced configuration of servers, traditional, deterministic bottleneck analysis is distinctly nonoptimal; our results provide a simple way to locate production bottlenecks in shops with significant queueing effects.

Method and system for error recovery for cascaded servers

Abstract: An improved method for error recovery in client-server distributed processing systems using cascaded servers. The method adds an additional form of acknowledgement message from a server to its predecessor. This message is sent when the server has completed all its processing and has received an ACK message from its successor, indicating that its result has been successfully received by the successor. The predecessor server retains a copy of its results until it receives the done message, at which time the copy of the results is discarded. If a done message is not received by the time a timer has expired, indicating that a problem has occurred with a server or its communications, the predecessor server resends its stored results.

Inter-network transport element of sonet overhead

Abstract: In a SONET cross connect, the same physical link is used between the interfaces and the matrix to carry the overhead and the payload The cross connection function within the matrix may be used to group, concentrate and route the overhead signals between a server and the matrix The matrix may also be used to transport signals between servers Overhead may be grouped and transported as payload

Wide Area Information Servers: An Executive Information System for Unstructured Files

Abstract: In this paper we present a corporate information system for untrained users to search gigabytes of unformatted data using quasi‐natural language and relevance feedback queries. The data can reside on distributed servers anywhere on a wide area network, giving the users access to personal, corporate, and published information from a single interface. Effective queries can be turned into profiles, allowing the system to automatically alert the user when new data are available. The system was tested by twenty executive users located in six cities. Our primary goal in building the system was to determine if the technology and infrastructure existed to make end‐user searching of unstructured information profitable. We found that effective search and user interface technologies for end‐users are available, but network technologies are still a limiting cost factor. As a result of the experiment, we are continuing the development of the system. This article will describe the overall system architecture, the implemented subset, and the lessons learned.

Method and apparatus for multiprotocol operation of a networked peripheral

Abstract: Method and apparatus for multiprotocol operation of an interactive network board which ties a peripheral to a local area network and which executes first and second peripheral servers for servicing jobs from respectively different operating systems. The first server checks its operating system job queue and receives job information from the operating system according to a first protocol. The first server seizes exclusive control over the peripheral and signals to other servers that the first server has control. In coordination with the first server, the second server checks its operating system job queue and receives job information from the operating system. When the first server relinquishes control over the peripheral, the second server can seize exclusive control over the peripheral so as to send its job information to the peripheral. Preferably, while each server has exclusive control over the peripheral, it checks with other servers to determine whether those other servers have pending requests for use of the peripheral. If there are pending requests, then control over the server can be relinquished at the end of a job, even though there are remaining jobs in the job queues, so as to allow alternate usage of the peripheral by each server. The interactive network board automatically and adaptively determines the frame packet types being used for LAN communication by the respective operating systems by prescanning broadcast requests on the LAN bus for predetermined header information corresponding to each of the operating protocol stacks (or towers) for the servers.

Admission Control Algorithm for Multimedia On-Demand Servers

Abstract: Future advances in networking coupled with the rapid advances in storage technologies will make it feasible to build a multimedia on-demand server on a metropolitan-area network. However, the economic viability of such a multimedia on-demand server is dependent on the ability to amortize its operational costs over a large number of clients. In this paper, we present a taxonomy of policies for servicing multiple clients, and propose round robin and quality proportional admission control algorithms for determining whether a retrieval request from a new client can be admitted without violating the real-time requirements of any of the clients already being serviced. We evaluate the performance of various servicing policies, and show that they are an order of magnitude scalable compared to straightforward multiplexing techniques such as servicing one subscriber per disk head.

Cluster-based file replication in large-scale distributed systems

Abstract: The increasing need for data sharing in large-scale distributed systems may place a heavy burden on critical resources such as file servers and networks. Our examination of the workload in one large commercial engineering environment shows that wide-spread sharing of unstable files among tens to hundreds of users is common. Traditional client-based file cacheing techniques are not scalable in such environments.We propose Frolic, a scheme for cluster-based file replication in large-scale distributed file systems. A cluster is a group of workstations and one or more file servers on a local area network. Large distributed systems may have tens or hundreds of clusters connected by a backbone network. By dynamically creating and maintaining replicas of shared files on the file servers in the clusters using those files, we effectively reduce reliance on central servers supporting such files, as well as reduce the distances between the accessing sites and data. We propose and study algorithms for the two main issues in Frolic, 1) locating a valid file replica, and 2) maintaining consistency among replicas. Our simulation experiments using a statistical workload model based upon measurement data and real workload characteristics show that cluster-based file replication can significantly reduce file access delays and server and backbone network utilizations in large-scale distributed systems over a wide range of workload conditions. The workload characteristics most critical to replication performance are: the size of shared files, the number of clusters that modify a file, and the number of consecutive accesses to files from a particular cluster.

Method and apparatus for automatic telephone scheduling system

Abstract: An automatic telephone scheduling system (20) includes a data server (22), a dial server (24), and a plurality of agent workstations (28) which communicate with each other through messages transmitted over a local area network (30). The data server (22) has stored in its memory a plurality of account information records to be utilized for one or more calling cam-paigns. The dial server (24) includes an independent intelligent processor (80) which functions as a master to the data server (22) but as a slave to the agent workstations (28). The dial server (24) further icludes a plurality of audio MUXes (88) and dialers (90) for automatically dialing and connecting calls to the agent workstations (28). In various embodiments, a plurality of da-ta servers and/or a plurality of dial servers are employed.