The Credit Net ATM Project

This is the World Wide Web home page of the ARPA-funded Credit Net (VC Nectar) project in the School of Computer Science at Carnegie Mellon University.

Project Overview

The Credit Net project has developed a 622 Mbit/second flow-controlled ATM network. The network consists of switches built by Bell Northern Research and PCI host adapters built by Intel Architecture Labs, designed in cooperation with Harvard and CMU, respectively. Its features include As in its predecessor project, Gigabit Nectar, the research is focused on building a system that supports applications effectively.

Credit Net networks are currently operational at both CMU and Harvard. The CMU testbed currently has links running at OC3 and OC12 rates and includes about 10 nodes, which will grow to about 25 high-end personal computers in the next few months. Research topics on our agenda include: This page has more information on the Credit Net adapter and some results of early credit-based flow control experiments. You can find more detailed information in our Credit Net papers .

Credit Net Adapter

Intel and CMU have developed 622 megabits per second host interface for the PCI bus. The main component on the board is an ASIC that supports standard ATM cell processing and management of host buffers. The ASIC, combined with memory to store per-VC state and a physical layer interface, can form a highly intergrated OC3 or OC12 ATM adapter (below on the left). Alternatively, an optional 960 microcontroller can be added to support more experimental features of the network such as flow control (below on the right).

The performance of the adapters using a 90 MHz Pentium platform running NetBSD is shown below. The sustained bandwidth is close to the maximum OC3 rate for packets as small as 5 KByte, while for TCP and UDP (measured using ttcp) the performance is limited by the performance of the platform.

On a 133 MHz platform, using a newer PCI bridge chip (Triton instead of Neptune), we can sustain full duplex OC3 throughput (aggregate throughput of 268 Mbs), achieving over 90% of the bandwidth for packets as small as 1000 Bytes. We are in the process of collecting TCP and UDP results.

The adapter is described in more detail in: Host and Adapter Buffer Management in the Credit Net ATM Host Interface, Corey Kosak, David Eckhardt, Todd Mummert, Peter Steenkiste, and Allan Fisher, Proceedings of the 20th Annual Conference on Local Computer Networks, IEEE Computer Society, Minneapolis, September 1995. A method of using the adaptor to support remote memory deposits is described in Fine Grain Parallel Communication on General Purpose LANs, Todd Mummert, Corey Kosak, Peter Steenkiste, and Allan Fisher, International Conference on Supercomputing, ACM, Philadelphia, May 1996.

Early credit results

Credit-based flow control has been implemented in both the Credit Net switches and hosts. In February 1995 we demonstrated that credit-based flow control can eliminate the cell loss, and resulting drop in performance, on congested links inside an ATM network. The results are summarized below, and more details can be found here .

The basic result is that without flow control, cells get lost resulting in very poor performance, as measured using ttcp. This is illustrated by the traces shown below (traces on left): packets are lost, resulting in TCP timeouts and loss of throughput. As is shown in the traces on the right, with credit-based flow control we achieve good throughput (the sum of the throughput of the links equals the OC3 link bandwidth) and fair sharing of the bandwidth. If TCP is competing with a traffic stream without backoff, e.g. a video stream, throughput drops to close to 0.

Preliminary results on a more systematic performance comparison between rate- and credit-based flow control schemes can be found in Experimental Evaluation of ATM Congestion Control Mechanisms, Prashant Chandra, Allan Fisher, Corey Kosak, and Peter Steenkiste, IEEE ATM Workshop 1996, San Fransisco, August 1996.

More recently we implemented an all-software credit implementation on the host. The host interprets incoming credit cells and schedules packets based on the availability of credit; no hardware support on the adapter is needed. No flow control is used for the incoming data stream under the assumption that the host should should have enough buffer space to store incoming data. The results are summarized below. The nodes used in the test are 90 MHz Pentium PCs.

A more complete description and evaluation of both rate and credit-based flow control implementations on the end-points can be found in Implementation of ATM Endpoint Congestion Control Protocols, Prashant R. Chandra, Allan L. Fisher, Corey Kosak and Peter A. Steenkiste, Hot Interconnects, Stanford, IEEE, August 1996.

Related Projects

The Credit Net group works closely with several other research projects at CMU, including the iWarp project, the Fx parallel FORTRAN compiler project, Dome, Scotch parallel storage, the Environmental modeling NSF grand challenge application , and the multicomputer project.

People

Allan Fisher and Peter Steenkiste lead the project at CMU. David Eckhardt, Corey Kosak, Todd Mummert, and Prashant Chandra form the rest of the inner circle. H. T. Kung leads the Harvard team's switch building project.
prs@cs.cmu.edu