Showing papers in "IEEE Network in 1993"

RSVP: a new resource ReSerVation Protocol

Abstract: A resource reservation protocol (RSVP), a flexible and scalable receiver-oriented simplex protocol, is described. RSVP provides receiver-initiated reservations to accommodate heterogeneity among receivers as well as dynamic membership changes; separates the filters from the reservation, thus allowing channel changing behavior; supports a dynamic and robust multipoint-to-multipoint communication model by taking a soft-state approach in maintaining resource reservations; and decouples the reservation and routing functions. A simple network configuration with five hosts connected by seven point-to-point links and three switches is presented to illustrate how RSVP works. Related work and unresolved issues are discussed. >

Alarm correlation

Abstract: The authors discuss the development of an alarm correlation model and a corresponding software support system that allow efficient specification of alarm correlation by the domain experts themselves. Emphasis is placed on the end-user orientation of IMPACT, the intelligent management platform for alarm correlation tasks which implements the proposed model. The desire was to lower the barrier between the network management application development process and the end user of the application, the network management personnel. IMPACT is a step towards this goal. The proposed alarm correlation model was used for three purposes: intelligent alarm filtering, alarm generalization and fault diagnosis. >

Giving applications access to Gb/s networking

Abstract: Operating systems (OSs) used in high-speed networks must reduce copying to deliver maximal throughput to applications and must deliver this throughput while preserving the capability of the host to perform applications processing. One way to reduce copying by enabling data transfers directly to and from buffers located in application-process address spaces is presented. The method has been demonstrated experimentally and is shown to deliver high throughputs. OS support must also include scheduling, which allows bandwidth-allocated traffic streams to be delivered. The implementation described provides resource scheduling for network users, and considerably reduces interrupt overhead. >

Network subsystem design

Abstract: It is argued that the bandwidth of the CPU/memory data path on workstations will remain within the same order of magnitude as the network bandwidth delivered to the workstation. This makes it essential that the number of times network data traverses the CPU/memory data path be minimized. Evidence which suggests that the cache cannot be expected to significantly reduce the number of data movements over this path is reviewed. Hardware and software techniques for avoiding the CPU/memory bottleneck are discussed. It is concluded that naively applying these techniques is not sufficient for achieving good application-to-application throughput; they must also be carefully integrated. Various techniques that can be integrated to provide a high bandwidth data path between I/O devices and application programs are outlined. >

Monitoring distributed systems

Abstract: It is suggested that a functional model of monitoring in terms of the generation, processing, dissemination, and presentation of information can help determine the facilities needed to design and construct distributed systems. Implementation issues are also discussed, with attention given to the intrusiveness of monitoring systems and object-based implementation. It is concluded that generic monitoring services are important tools for managing distributed systems and for debugging during system development. Monitoring services may also be needed as part of the application itself, such as in process control and factory automation. >

Designing a practical ATM LAN

Abstract: The adaptor cards and driver software for workstations and local asynchronous transfer mode (ATM) switches and switch control software used in an ATM local area network (LAN) system are discussed. It is shown that the ATM hardware and software components together provide services that are essential for ATM to be considered a realistic alternative to current shared-media LANs. These services include: completely transparent support for the TCP/IP protocol suite, an application programming interface for full access to the underlying ATM capabilities, support for AAL5, AAL3/4, and the null AAL, both connection-oriented and connectionless service, dynamic connection establishment or switched virtual circuits, resource reservation of guaranteed bandwidth and quality of service, full-bandwidth multicast and broadcast, virtual path and channel routing among multiple switches, automatic configuration and failure recovery, dynamic address assignment and internetwork address resolution, and network management via the simple network management protocol (SNMP). >

Packet reassembly during cell loss

Abstract: Asynchronous transfer mode (ATM) is a packet switched data transport system based on short, fixed length cells. Each cell carries a virtual channel indicator (VCI) and virtual path indicator (VPI) in its header. Essential to the services offered by the ATM networks is the ATM adaptation layer (AAL), an ITU-TSS defined layer that adapts the cell-based ATM physical layer to packet, datagram, or bit-stream-oriented higher layers. Failure modes causing cell loss along a virtual connection are examined, and the ways AALs cope are analyzed. The sources of cell loss and their effects on AAL3/4 or AAL5 type of service are described. The usefulness of the ability of AAL3/4 to pass fragments of corrupted data up to higher layer protocols is discussed, and the implementation of selective cell discarding within switching nodes is considered, and the limitations imposed by each AAL are examined. >

Afterburner (network-independent card for protocols)

Abstract: Many current implementations of protocols such as the Transmission Control Protocol/Internet Protocol (TCP/IP) are inefficient because data are often accessed more frequently than necessary. Three techniques that reduce the need for memory bandwidth are proposed. The techniques are copy-on-write, page remapping, and single-copy. Afterburner, a network-independent card that provides the services that are necessary for a single-copy protocol stack, is described. The card has 1 MByte of local buffers and provides a simple interface to a variety of network link adapters, including HIPPI and asynchronous transfer mode (ATM). Afterburner can support transfers to and from the link adapter card at rates up to 1 Gbit/s. An implementation of TCP/IP that uses the features provided by Afterburner to reduce the movement of data to a single copy is discussed. Measurements of the end-to-end performance of Afterburner and the single-copy implementation of TCP/IP are presented. >

Extending trouble ticket systems to fault diagnostics

Abstract: Network management in large heterogeneous communication environments requires the incorporation of more advanced tools than are currently available, especially in the area of fault management. By providing an informational infrastructure for documenting network behavior and maintenance activities, TTSs (trouble ticket systems) have provided some relief for the fault management problem. Researchers have recognized TTSs as a means for realizing more advanced functions in network management platforms to effect automatic trouble ticket generation, fault diagnostics, and the correlation of multiple views network problems and behavior. The authors describe current research trends towards understanding and implementing these advanced functions in a TTS framework. >

ATM everywhere

Abstract: The use and deployment of asynchronous transfer mode (ATM) in a target of operating environments and administrative domains are discussed. The operating environments include wide area ATM, regional area ATM, local area ATM, desk area ATM, and local wireless ATM. The challenges of interdomain ATM are reviewed. >

Voice synchronization in packet switching networks

Abstract: An algorithm for voice synchronization for packet switching networks is presented. The algorithm has been tested both in simulation and on a real network. The algorithm runs on the TRAME packet switching network for both the Vocoder and CELP DoD voice coding standards. Some results of these tests are presented. Some details of the algorithm development and implementation are given as well. >

B-ISDN and the OSI protocol reference model

Abstract: The evolution of public telecommunication networks toward broadband integrated services digital networks (B-ISDNs) is presented. The asynchronous transfer mode (ATM), which supports B-ISDNs, and the B-ISDN protocol reference model are discussed. It is shown that the service offered by the ATM layer in the B-ISDN protocol reference model is equivalent to the service offered by the OSI physical layer. It is also shown that the service offered by the ATM adaptation layer (AAL) type 3/4 is similar to an OSI data link service. The emergence of the metropolitan area network (MAN) standard as an intermediate support for broadband services and the similarities between the DQDB MAC and the AAL type 3/4 in its connectionless mode are discussed. >

HIPPI: simplicity yields success

Abstract: The high-performance parallel interface (HIPPI), a simplex point-to-point interface for transferring data at peak data rates of 800 or 1600 Mb/s over distances of up to 25 m, is reviewed. The HIPPI physical layer, framing protocol, link encapsulation, its mapping to fiber channels, mapping to IPI-3, switch control, fiber extender, and tester are described. The role of HIPPI in networks is discussed. >

Project Zeus

Abstract: The concept behind Zeus, an asynchronous transfer mode (ATM) network implemented on the campus of Washington University, is discussed. The first phase of the project, which demonstrated the feasibility of the core technology, provided a basis for a more complete design, and served as a testbed for application development, is reviewed. The switch architecture, network control software, application interfaces, and internetworking developed in the first phase are described. >

SIP: Simple Internet Protocol

Abstract: Several features of the Simple Internet Protocol (SIP) are described and compared to those offered by the Internet Protocol (IP). The changes required for other protocols in the TCP/IP suite to accommodate SIP are discussed, as are the mechanisms available to allow gradual transition of the Internet from IP to SIP. Future directions for SIP development, a report on current implementation status, and a summary of the specific improvements offered by SIP over IP are presented. >

Improving the routing and addressing of IP

Abstract: Simple mechanisms for introducing hierarchy into the interdomain routing system, making it practical to route a truly large Internet, are described. It is shown that addresses need to be assigned along network topological lines to maximise the reduction in routing overhead. Simplifying the mechanisms for changing hot addresses makes it possible to renumber a network so that it has a topologically significant address when the site changes position within the routing system. These proposals make it possible to significantly increase the utilization of addresses and extend the time the Internet can use the Internet Protocol (IP) Version 4. >

A near-term architecture for deploying Pip

Abstract: Pip, a new Internet Protocol (IP) that is a candidate to replace IP Version 4, is discussed. Pip solves the dual problems of scaling and address depletion that currently plague IP and provides new features such as provider selection. Pip also has a means of evolving to new features, such as real-time flow handling. Pip's architecture, host input processing, evolution to real-time flows, fast options handling, and address management are described. The IP to Pip transition scheme is outlined. >

Early experiments with asynchronous time division networks

Abstract: Asynchronous time division multiplexing (ATDM) is a technique for sharing a transmission line by carrying user data in small fixed-size cells, usually no larger than 64 bytes. The technique is attractive because it can simultaneously support voice, video, and data communications at a variety of transmission speeds and using many protocols. A retrospective account of three experimental ATDM networks that were developed during the period 1969 through 1984 is presented. Spider was an initial exploration of asynchronous time division multiplexing as a foundation for data communications. Lessons drawn from that experience were built into Datakit, an experiment that eventually became an AT&T product and Incon, an experiment which put wall sockets in offices and in people's homes. >

Parallel host interface for an ATM network

Abstract: The design of a parallel protocol processing system for a host system connected to an asynchronous transfer mode (ATM) network is described. Interesting aspects of the design are that the AAL5 adaptation layer is realized in hardware directly on the network adapter, and that the network adapter semetric interfaces to a multiprocessor using the processor's memory bus. Support also exists for implementing parallel protocols up to and including the application layer. Parallel implementations of TCP/IP, ISO Session, and presentation protocols are outlined. >

Management and control for giant gigabit networks

Abstract: The authors describe the new challenges presented for the management of broadband networks and outline the framework of the COMET research group at Columbia University for finding solutions to these challenges They review a prototype that implements some of the requirements for managing the broadband networks of the future In the COMET framework, management operations are defined in terms of primitives with a much higher semantic level than that typified by get and set in the Open Systems Interconnection (OSI) management model, and the user interface to the managed system is based on virtual worlds that allow the manager to directly visualize and interact with the network The impact that this new environment may have on existing management architectures is discussed >

Challenges in network management

Abstract: The author notes that standard network management has certainly had a great impact on the industry. However, he further points out that it is important to distinguish between two kinds of standards, horizontal and vertical. Horizontal standards, such as management protocols and management objects, can at most provide a technology base for enabling products. However, in order to produce a robust market, vertical standards, such as APIs and a framework for management applications are needed. The author notes that the industry continues its struggle to distinguish between dreams and reality; 'architectures' and marketeers are very good at seducing the industry's attention for long periods of time. Fortunately, it is the engineers who must ultimately deliver products that provide cost-effective solutions. As such, the outcome is uncertain, the author concludes. >

Youth often risk unsafe abortions.

Abstract: The topic of this article is the use of unsafe abortion for unwanted pregnancies among adolescents. The significance of unsafe abortion is identified as a high risk of serious health problems, such as infection, hemorrhage, infertility, and mortality, and as a strain on emergency room services. The World Health Organization estimates that at least 33% of all women seeking hospital care for abortion complications are aged under 20 years. 50 million abortions are estimated to be induced annually, of which 33% are illegal and almost 50% are performed outside the health care system. Complications are identified as occurring due to the procedure itself (perforation of the uterus, cervical lacerations, or hemorrhage) and due to incomplete abortion or introduction of bacteria into the uterus. Long-term complications include an increased risk of ectopic pregnancy, chronic pelvic infection, and infertility. Mortality from unsafe abortion is estimated at 1000/100,000 procedures. Safe abortion mortality is estimated at 0.6/100,000. When infertility results, some cultures ascribe an outcast status or marriages are prevented or prostitution is assured. The risk of complications is considered higher for adolescents. Adolescents tend to delay seeking an abortion, lack knowledge on where to go for a safe procedure, and delay seeking help for complications. Peer advice may be limited or inadequate knowledge. Five studies are cited that illustrate the impact of unsafe abortion on individuals and health care systems. Abortions may be desired due to fear of parental disapproval of the pregnancy, abandonment by the father, financial and emotional responsibilities of child rearing, expulsion from school, or inability to marry if the child is out of wedlock. Medical, legal, and social barriers may prevent women and girls from obtaining safe abortion. Parental permission is sometimes a requirement for safe abortion. Fears of judgmental or callous health personnel may be barriers to seeking safe abortion. Some countries lack adequately trained medical personnel and supplies. Mortality and morbidity declines are considered possible with legalization, more trained health personnel, and family planning programs for youth and education for parents. Language: en

Dynamically selecting protocols for socket applications

Abstract: An application enablement method that enables communications applications to run on protocols for which they were not written is discussed. The focus is on a subset of networks and communications application programming interfaces. The communications interface chosen is Berkeley sockets. The network protocols used are SNA (Systems Network Architecture) and TCP/IP (Transmission Control Protocol/Internet Protocol). The running of existing socket applications over SNA networks, which requires support for transparently masking the differences between TCP/IP and SNA from the applications, is described. Protocol selection, address mapping, and connection management are also described. >

Automatic failure isolation and reconfiguration (ring networks)

Abstract: An automatic reconfiguration of counter-rotating rings that significantly increases the reliability of a network is described. The automatic failure isolation and reconfiguration methodology (FIRM) uses flags and management frames to bypass both single and multiple link failures as well as cable failures. FIRM uses WRAP-D switching along with the beaconing process to locate some failures which cannot be located by beaconing process alone. Examples illustrating the reconfiguration procedures are presented. >

TUBA: replacing IP with CLNP

Abstract: The Connectionless Network Protocol (CLNP), which is supported by the associated OSI routing protocols, is proposed as a replacement for the Internet Protocol (IP). The basis of the proposal is to run the Internet transport protocols, the Transmission Control Protocol (TCP), and the User Datagram Protocol (UDP) on top of CLNP in an approach known as TCP and UDP with bigger addresses (TUBA). The fundamentals of CLNP and the OSI connectionless routing architecture, the operation of the IP suite with CLNP replacing IP, the support of Internet applications operating on top of TUBA, and a transition plan to a TUBA Internet are discussed. >

Dealing with history and time in a distributed enterprise manager

Abstract: A family of integrated systems has been developed within Digital's enterprise management architecture (EMA). The product called DECmcc is part of a larger EMA implementation, which consists of both an enterprise management director and a framework for developing consistent and integrated management solutions with EMA. An object-based framework provides time-oriented dispatching of operations and serves as a foundation for modules supporting past, present, and future time views of managed objects. Fully integrated modules implement a conceptual diachronic object-base that allows time-oriented, object-based modeling of resources. To maintain the historical part of this diachronic object base, DECmcc provides comprehensive facilities for collecting, processing, purging, archiving, and exporting historical attribute data. The managed object instances, for the existing objects that need to be globally and directly accessible, are stored using DECdns, i.e., essentially DCE/CDS. The historical data are stored on the domain basis locally. Recorded historical data and 'live' enterprise data can be exported to a relational database in a uniform fashion for any combination of past time, current, and future data. >

Interconnecting high-speed LANs and backbones

Abstract: Transparent source routing, an algorithm that exploits the best features of the transparent spanning tree and source routing algorithms for medium-access-control (MAC) bridges used in local area network (LAN) interconnections, is presented. The scheme uses optimal routes, is transparent to end stations, and reduces bridge processing overheads. Aspects of the existing transparent spanning tree and source routing algorithms are described. A comparison of the new scheme with the two existing schemes is presented. >

Managing telecommunications networks

Abstract: The authors suggest that one must view the network itself as the customer's database, and make sure it is self-diagnosing, self-healing, and intelligent. The perspective that the network is the database means moving away from the feeling that no trouble means total health, towards more preemptive maintenance. With fiber, capacity as a limiting factor is no longer an issue; the creative work will be in developing network capabilities. Universities need to accept the challenge of developing the theory of software stability. A crazy quilt of multiple equipment vendors cannot produce an integrated, problem-solving network. But if a network is self-diagnosing, self-healing, and intelligent, the available staff can focus on the customer and seize opportunities to introduce new services. >

Manila street children face many sexual risks.

Abstract: Estimates of as many of 100 million 5-18 year old children have been generated on the number of street children in urban areas worldwide. About 70% are on the streets during the day and return home at night and 20-25% know where their families are but prefer to live on the streets. 5-10% are without family and live on the streets. About 25-33% are young girls. Poverty is one of the driving forces for street children from rural areas. These children face illnesses such as respiratory infections skin diseases gastrointestinal problems trauma sexual abuse exploitation unwanted pregnancy and sexually transmitted diseases such as syphilis and gonorrhea. In Kenya 4 out of 5 street children know that AIDS while in India few are aware and those who are have a fatalist attitude since violence and death are faced by many of these children daily. As the numbers of street children have grown so have health services to meet their needs. In the Philippines Metro Manila has established a model program in providing health nutrition counseling referral.

Making broadband-ISDN successful

Abstract: It is argued that the technology needed to support the emerging broadband integrated services digital network technology (B-ISDN) is on the horizon, but the deployment of this technology in the telecommunication plant will not be sufficient to make B-ISDN a commercial success. Two corequisites for ensuring the long-term success of the emerging B-ISDN include stimulating the demand for bandwidth by inventing and promoting new telecommunication applications, and enabling the interconnection or convergence of the emerging high-speed network and the existing cable television infrastructure. The author believes that this is necessary because no alternative seems to offer a workable base for financing the transition to B-ISDN and for reaching the needed economy of scale, and because the intrinsic value of the public network grows combinatorially with the number of subscribers. >