Applied Research Laboratory
Publications by Year
1997

Buddhikot, M. and Chen, X. "Project MARS: WWW based Scalable, Interactive Multimedia Recording and Playback Services,'' Winner of the first prize at the Research Poster Competition, ACM SIGCSE/SAC'97, San Jose, CA, Feb 27-Mar 1, 1997

In recent years, the World Wide Web (WWW) infrastructure has gained immense popularity and experienced explosive growth as a means to disseminate information across the Internet. However, the current web infrastructure does not support streaming access to bandwidth intensive streams such as video, graphics, high quality animation and audio. Clearly, services that offer interactive recording/playback of such data and scale to support hundreds or thousands of independent concurrent clients will be critical to realize full potential of future information superhighway. Such services must provide periodic data retrieval and transmission for each of its clients to support Quality-of-Service (QoS) guarantees. The two main components of these services are the storage server and the client devices. In an ambitious and large project, named ProjectMARS, currently underway at the Applied Research Labs in Washington University in St. Louis, we are building high performance servers and clients to realize such services.

(full description)

Buddhikot, M., and Parulkar, G. ``Efficient Data Layout, Scheduling and Playout Control in MARS,'' ACM/Springer Multimedia Systems Journal, pp. 199-211, Volume 5, Number 3, 1997.

Many application domains (such as avionics, telecommunications, and multimedia) require real- Large scale on-demand multimedia servers, that can provide independent and interactive access to a vast amount of multimedia information to a large number of concurrent clients, will be required for a wide spread deployment of exciting multimedia applications. Our project, called Massively-parallel And Real-time Storage (MARS) is aimed at prototype development of such a large scale server. This paper primarily focuses on the distributed data layout and scheduling techniques developed as a part of this project. These techniques support a high degree of parallelism and concurrency, and efficiently implement various playout control operations, such as fast forward, rewind, pause, resume, frame advance and random access.

Decasper, D., Waldvogel M., Dittia, Z., Adiseshu, H., Parulkar, G., and Plattner B., "Crossbow - A Toolkit for Integrated Services over Cell Switched IPv6," Proceedings of the IEEE ATM'97 workshop May 25-28, 1997, Lisbo, Portugal.

The project Crossbow provides a framework to investigate services and mechanisms including resource management and packet scheduling for multimedia/multicast applications. In particular the Internet Protocol version 6 (IPv6, IP next generation, IPng) protocol suite on top of ATM is considered to demonstrate possible synergy between ATM and IPv6. The presented architecture includes IPv6 and RSVP, running on BSD Unix using the 1.2 Gbps APIC (ATM Port Interconnect Controller) chip, as well as support for Ethernet networks.

Dittia, Z., Parulkar, G., Cox, J. "The APIC Approach to High Performance Network Interface Design: Protected DMA and Other Techniques" Proceedings of INFOCOM'97, April 7-11, 1997, Kobe, Japan.

We are building a high performance 1.2 Gb/s ATM network interface chip called the APIC (ATM Port Interconnect Controller). In addition to borrowing useful ideas from a number of research and commercial prototypes, the APIC design embraces several innovative features, and integrates all of these pieces into a coherent whole. Some of the novel ideas incorporated in the APIC design include: Protected DMA and Protected I/O, which allow applications to queue data for transmission or reception directly from user-space, effectively bypassing the kernel. This argues for moving the entire protocol stack including the interface device driver into user-space, thereby yielding better latency and throughput performance than kernel-resident implementations. Pool DMA when used with Packet Splitting, is a technique that can be used to build true zero-copy kernel-resident protocol stack implementations, using a page-remapping technique. Finally, Orchestrated Interrupts and Interrupt Demultiplexing are mechanisms used to reduce the frequency of interrupts issued by the APIC.

Although many of these ideas have been developed in the context of an ATM network interface, we believe they are also applicable in other contexts. In particular, protected DMA and I/O are promising techniques for improving the performance of several different types of I/O devices.

Engel, Robert; and Turner, Jonathan S. "Real Time Connection Admission Control with Multistate Traffic Sources," submitted to ICC 1997.

This paper develops a practical method for connection admission control in ATM networks. The method is based on a virtual cell loss probability criterion, is designed to handle heterogeneous traffic types and allows each traffic source to be described by an individual finite-state model with as many states as are needed to describe the source traffic. To make connection admission decisions with respect to individual links, an aggregate finite state model is computed from the individual models and used to estimate the virtual cell loss probabilities. To reduce the computational requirements for maintaining the aggregate traffic model, the aggregate model uses quantized data rates and is maintained incrementally using direct numerical convolution. The approximations required by the quantization process can be done in a strictly conservative way.

Gopalakrishnan, R., Parulkar, G. "Efficient User space Protocol Implementations with QoS Guarantees using Real-time Upcalls." Submitted to SIGCOMM'97.

Two important requirements for protocol implementations to be able to provide quality of service (QoS) guarantees within the endsystem are (1) Efficient CPU scheduling for application and protocol processing, and (2) efficient mechanisms for data movement. We first consider CPU scheduling. Scheduling is needed to guarantee that the application and protocol tasks involved in processing each stream execute in a timely manner and obtain their required share of the CPU. We have designed and implemented an operating system mechanism called the real-time upcall (RTU) to provide such guarantees. The RTU is a user level handler function that implements protocol or application code, and the operating system ensures that RTU handlers obtain their required share of CPU over periodic intervals.

We next consider data movement. For efficient data movement, we have implemented a number of techniques such as (1) direct movement of data between the application and the network adaptor to avoid data copying, (2) batching of network I/O requests to reduce context switches, (3) lockfree receive and transmit queues in kernel-user shared memory, and (4) header-data splitting at the receiver to keep bulk data page aligned. These techniques work well in conjunction with the RTU mechanism and can also be used in isolation.

Using the above scheduling and state movement techniques, we have implemented the TCP/IP protocol suite in user space. For each TCP connection, RTUs are set up for performing TCP output, input, and timer processing functions. RTUs associated with bulk data connections are scheduled periodically and process a batch of packets each period. THus the application layer throughput for these connections is guaranteed, even with competing real-time and "best effort" load on the endsystem. In addition, the maximum throughput that can be obtained using RTU based TCP/IP is higher than the NetBSD kernel implementation by virtue of our efficient data movement techniques. RTUs associated with low delay connections (called reactive RTUs) are invoked directly in response to packet arrivals thus circumventing delays due to time shared scheduling in the endsystem. We believe that ours is the first solution that provides good data path performance in addition to application level bandwidth and delay guarantees for standard protocols and operating systems.

Kantawala, A., Parulkar, G., DeHart, J., and Marz, T. "Supporting DIS Applications using ATM Multipoint Connection Caching," Proceedings of INFOCOM'97, April 7-11, 1997, Kobe, Japan.

This report describes an ATM Multipoint Connection Cashing strategy (AMCC) to control the explosive growth of traffic within the network and at an endpoint in a large Distributed Interactive Simulation (DIS) application such as a battlefield simulation. For very large DIS applications with 100,000 entities, the current method of broadcasting information among entities will no longer be feasible due to computational and network bandwidth limitations. Our scheme divides the simulation space into grids and each grid square or a set of grid squares forms a multicast group. Entities join the groups within their perception range and thus, they receive state updates only from entities within their perception range. AMCC utilizes the unique capabilities of our dynamic multipoint ATM signaling protocol, CMAP. We have implemented AMCC and have successfully demonstrated a small DIS battlefield simulation application running over our ATM testbed.

Kempner, K., Chow, D., Choyke, P., Cox, J., Elson, J., Johnson, C., Okunieff, P., Ostrow, H., Pfeifer, J., and Martino, R. "The development of an ATM-based Radiology Consultation WorkStation for radiotherapy treatment planning." Medical Imaging 1997 Conference Proceedings, Vol 3031. Society of Photo-Optical Instrumentation Engineers, Inc., Bellingham, WA, 1997.

The Radiology Consultation WorkStation (RCWS) is a multimedia, medical imaging workstation being developed for use in an electronic radiology environment, utilizing a prototype Asynchronous Transfer Mode (ATM) telemedicine network, in support of radiotherapy treatment planning. A radiation oncologist in the Radiation Oncology Department, and a radiologist in the Diagnostic Radiology Department, will be able to consult, utilizing high-quality audio/video channels and high-resolution medical image displays, prior to the design of a treatment plan. Organ and lesion contouring is performed via a shared-cursor feature, in a consultation mode, allowing medical specialists to fully interact during the identification and delineation of lesions and other features.

Keywords: Telemedicine, Multimedia System, ATM Network, Medical Consultation, Medical Imaging

Papadopoulos, C., Parulkar, G., Varghese, G. "An Error Control Scheme for Large-Scale Multicast Applications." WUCS-97-23.

Retransmission based error control for large scale multicast applications is difficult because of two main problems: request implosion and lack of local recovery. Existing schemes (SRM, RMTP, TMTP, LBRRM) have good solutions to request implosion, but only approximate solutions (e.g., based on scoped multicast) for the local recovery problem. Our scheme achieved finer grain fault recovery by exploiting new forwarding services that allow us to create a dynamic hierarchy of receivers. We use a new paradigm, where routers provide a more refined form of multicasting (that may be useful to other applications), that enables local recovery. The new services, however, are simple to implement and do not require routers to examine or store application packets; hence, they do not violate layering. Besides providing good local recovery, our scheme integrates well with the current IP model, has small recovery latencies (it requires no back-off delays), produces fewer duplicates than other schemes, and isolates group members from details of group topology.

Parulkar, G., Schmidt, D., Kraemer, E., Turner, J., Kantawala, A. "An Architecture for Monitoring, Visualization and Control of Gigabit Networks," WUCS-97-28.

We propose a network monitoring, visualization and control system (NMVC) that ensures adequate quality of service to network users while maintaining high network resource utilization. The main components of our system are a network probe, and endsystem probe, software network management agents that provide extensible multiattribute event filtering for highly scalable data/event collection, network operation centers (NOCs) which can remotely install and (re)configure these agents, efficient on-line event ordering algorithms that can help synthesize and display a consistent view of network health, status and performance and a View Choreographer that allows management applications and administrators to specify the mapping of network events to higher-level events and to visualization objects and updates.

The network probe will be built using the ATM Port Interconnect Controller chip and a CPU-memory module. The APIC-based probe meets all the requirements for a gigabit link probe. With two full duplex 1.2 Gb/s ATM ports, the APIC can be easily inserted in a link as a probe for packet/cell "snooping" to log traffic measurements without interfering with network traffic. Software network management agents in our proposed system are built atop network probes and used to track event flows, as well as classify and report events of interest to NOCs. We propose using a highly flexible, scalable and high performance event-filtering mechanism for the software network management agents that will effectively eliminate redundant management traffic using a dynamic tri-based filter fusion technique. The View Choreographer uses semantically-guided filtering algorithms to enable low latency updates of displays that present an accurate real-time snapshot of current network state.

Furthermore, the system's feedback control mechanisms will provide support for network configuration management, ATM virtual circuit management, router-to-router link management and application-level congestion management. The system will be experimentally evaluated by building a 3-5 node testbed and using a suite of multimedia traffic generator tools.

Schmidt, Douglas C.;Gokhale, Aniruddha; Harrison, Tim; and Parulkar, Gurudatta M. "A High-performance Endsystem Architecture for Real-time CORBA," IEEE Communications Magazine, February 1997, Vol. 14, No. 2.

Many application domains (such as avionics, telecommunications, and multimedia) require real-time guarantees for the underlying networks, operating systems, and middleware components to achieve their quality of service (QoS) requirements. In addition to providing end-to-end QoS guarantees, applications in these domains must be flexible and reusable. Requirements for flexibility and reusability motivate the use of object-oriented middleware like the Common Object Request Broker Architecture (CORBA). However, the performance of current CORBA implementations is not yet suited for hard real-time systems (e.g., avionics) and constrained latency systems (e.g., teleconferencing).

This paper describes the architectural features and optimizations required to develop real-time ORB endsystems that can deliver end-to-end QoS guarantees to applications. While some operating systems, networks, and protocols now support real-time scheduling, they do not provide integrated solutions. The main thrust of this paper is that advances in real-time distributed object computing can be achieved only by systematically pinpointing performance bottlenecks; optimizing the performance of networks, ORB endsystems, common services, and applications; and simultaneously integrating techniques and tools that simplify application development.

Return to the top

How to request a copy of one of our publications...

Email:

diana@arl.wustl.edu

Snail mail:

Applied Research Laboratory

Washington University, Campus Box 1045

One Brookings Drive

St. Louis, MO 63130-4899

Return to the top

Return to:

ARL's Home Page

Recent Works Index

Publications Index

Author Index

Year Index

Topic Index

Prepared by Diana Ehrlich

Last modified July 29, 1997