An overview of the AURORA gigabit testbed

Clark, David D.; Davie, Bruce S.; Farber, David; Gopal, I.; Kadaba, Bharath; Sincoskie, W.D.; Smith, Jonathan M.; Tennenhouse, David L.

doi:10.1109/infcom.1992.263450

Cited by 25 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A good example is the AURORA Gigabit Testbed [6], one of several such testbeds [2]. In AURORA, much of the focus has been on the development of technologies needed to deliver this performance to workstation-class machines, rather than supercomputers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Giving applications access to Gb/s networking

Smith

Traw

1993

IEEE Network

Self Cite

View full text Add to dashboard Cite

Network fabrics with Gigabit per second (Gbps) bandwidths are available today, but these bandwidths are not yet available to applications. The difficulties lie in the hardware and software architecture through which application data travels between the network and host memory. The hardware portion of the architecture is often called a host interface and the remainder of the protocol stack is implemented in host software.In this paper, we outline a variety of approaches to the architecture of such systems, examine several design points, and study one example in detail. The detailed example, an ATM Host Interface and Operating System support built at the University of Pennsylvania, illustrates design tradeoffs for hardware and software, and some of the implications of these tradeoffs on applications performance.

show abstract

Section: Introductionmentioning

confidence: 99%

“…The Host Interface can be connected directly to the CPU [6], as in augmenting the processing unit with a coprocessor. As in Scheme 4, there is no memory bus traversal, and further, the connection is to a system component which typically operates at speeds higher than memory bandwidths.…”

mentioning

confidence: 99%

Giving applications access to Gb/s networking

Smith

Traw

1993

IEEE Network

Self Cite

View full text Add to dashboard Cite

show abstract

“…A collection of researchers were exploring methods for increasing network throughputs by a factor of 100, using a variety of technologies, such as synchronous optical network (SONET) [19], [20], high-performance parallel interface (HIPPI) [21]- [23] and asynchronous transfer mode (ATM) [24], [25]. For example, research in the AURORA gigabit testbed [26], [27] was centered around ATM technology. While ATM signaling never quite matured, ATM link layers led to the broadband Internet [28], both by providing an infrastructure for high-speed Internet protocol (IP) overlays, and then later evolving into the methodology for building high performance IP switches [29].…”

Section: Broadband Internetmentioning

confidence: 99%

Active Networking: One View of the Past, Present, and Future

Smith

Nettles

2004

IEEE Trans. Syst., Man, Cybern. C

Self Cite

View full text Add to dashboard Cite

All distributed computing systems face the architectural question of the location (and nature) of programmability in the telecommunications networks, computers, and other peripheral devices comprising them. The perspective of this paper is that network elements should be as programmable as possible, to enable the most flexible distributed computing systems.There has been a persistent confluence among operating systems, programming languages, networking and distributed systems. We demonstrate how these interactions led to what is called "active networking", and in the spirit of "vox audita perit, littera scripta manet" (the spoken word perishes, but the written word remains), include an account of how it was made to happen. Lessons are drawn both from the broader research agenda, and the specific goals pursued in the SwitchWare project. We speculate on likely futures for active networking. This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the University of Pennsylvania's products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org. By choosing to view this document, you agree to all provisions of the copyright laws protecting it.This journal article is available at ScholarlyCommons: http://repository.upenn.edu/cis_papers/46 4 IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS-PART C: APPLICATIONS AND REVIEWS, VOL. 34, NO. 1, FEBRUARY 2004 Active Networking: One View of the Past, Present, and Future Jonathan M. Smith, Fellow, IEEE, and Scott M. Nettles, Member, IEEE Abstract-All distributed computing systems face the architectural question of the location (and nature) of programmability in the telecommunications networks, computers, and other peripheral devices comprising them. The perspective of this paper is that network elements should be as programmable as possible, to enable the most flexible distributed computing systems.There has been a persistent confluence among operating systems, programming languages, networking and distributed systems. We demonstrate how these interactions led to what is called "active networking," and in the spirit of "vox audita perit, littera scripta manet" (the spoken word perishes, but the written word remains), include an account of how it was made to happen. Lessons are drawn both from the broader research agenda, and the specific goals pursued in the SwitchWare project. We speculate on likely futures for active networking.

show abstract

“…2 Sequencing in ATM is done using an SAR protocol which requires only 3 bytes overhead per cell and can be implemented in hardware [8]. 3 However in the simulation experiments, we do consider the effect of this overhead in congestion and throughput measures. 4 We assume a single level FEC to be working both at block level and cell level.…”

Section: The Schemementioning

confidence: 99%

“…The two major contributions of this forum has been the LAN emulation service and rate-based congestion control for best-effort traffic also called available-bit-rate (ABR) in the ATM forum [13]. The near future will very likely have a communication infrastructure in which all existing LANs (also called legacy LANs) will interoperate over ATM networks at the backbone which provides high-speed wide area services [3]. Since businesses and users have heavily invested in LANs (Ethernet, Token-ring, FDDI), there is no doubt that we will witness a long period of co-existence of legacy LANs with ATM.…”

Section: Introductionmentioning

confidence: 99%

Increasing throughput in ATM/B-ISDN using simple buffer management and selective discarding

Chatterjee

Xiao²

1997

SIGCOMM Comput. Commun. Rev.

View full text Add to dashboard Cite

If the reliability provided by the network is lower than that requested by the application, the error and flow control systems in the end combination of a gatekeeping operation along with janitor routine that cleans up the buffer to preserve frames on an end-to-end basis is proposed. A simulation study demonstrates positive results. The proposed schemes even point towards efficient sizing of ATM switch buffers which guarantee no-loss transport of real-time information.Keywords: ATM networks, internetworking, forward error correction, congestion control, buffer management. IntroductionTwo significant changes are taking place in the computer communication arena. The first is the rapid standardization and subsequent deployment of Asynchronous Transfer Mode (ATM) networks which has been internationally chosen as the switching architecture for Broadband-ISDN [1]. The other is the changing role of the Internet which is gearing up to provide multimedia services in which real-time conferencing (voice, video, imaging) seems to be viable over IP along with a suitable light-weight adaptive transport layer protocol [12]. Our focus in this paper is on the intricate interworking of these two important technologies and more specifically, we consider mechanisms to control end-to-end packet-loss rates of real-time applications over this mix of IP, legacy LANs and ATM at the wide area.ATM is a switching and multiplexing mechanism operating over a fiber based physical network such as SONET. It uses a cell (53 byte packet with 5 byte header) as its basic switching element and all information types (voice, data and video) are transported inside the cell. The biggest advantage of ATM is in its ability to do statistical multiplexing (rather than synchronous TDM) and thus can effectively handle bursty, VBR and CBR traffic types. It is primarily a connection-oriented technology using a combination of virtual circuits and virtual paths to establish an end-to-end connection. Wide area ATM standards are coming from CCITT (now called ITU). The ATM forum has been established to speed up the deployment of ATM as a high-speed LAN technology. The two major contributions of this forum has been the LAN emulation service and rate-based congestion control for best-effort traffic also called available-bit-rate (ABR) in the ATM forum [13].

show abstract

An overview of the AURORA gigabit testbed

Cited by 25 publications

References 14 publications

Giving applications access to Gb/s networking

Giving applications access to Gb/s networking

Active Networking: One View of the Past, Present, and Future

Increasing throughput in ATM/B-ISDN using simple buffer management and selective discarding

Contact Info

Product

Resources

About