This paper presents a solution for efficient many-to-many communication over switched Ethernet. The performance of using an ordinary switch and an intelligent switch is compared, and the results of the analysis based on 100 Mbit/s fast Ethernet show that an intelligent switch used to handle many-to-many communication can give better performance (shorter latency and higher utilization) than an ordinary switch. We also extend the network to serve many-to-many traffic with real-time demands by adding a thin software layer to the intelligent Ethernet switch and the end-nodes. Earliest deadline first (EDF) scheduling is used.

There are several different types of communication traffic with real- time demands apparent in distributed industrial and embedded systems, for example, group communication and process synchronization. The length of these messages is often very short but the traffic volume might be really high. Standard network protocols do not normally reach very high utilization for such small messages. This paper presents a solution to efficiently support real-time short message communication over switched Ethernet. In our proposal, the Ethernet switch and the end-nodes are enhanced to combine several short messages into an Ethernet frame to improve the performance, and to give the short-message traffic real-time support on two levels, short-frame level and Ethernet frame level. Earliest Deadline First (EDF) scheduling is used in the switch and in the source nodes on both these two levels. We have characterized the performance of the network in terms of channel utilization and the number of accepted real-time channels, by simulations of the network assuming Fast Ethernet. We also show, by example, that we can reach an improvement of the possible short-message rate of 66%.

This paper proposes a switched Ethernet network that supports real-time communication with guaranteed bit rate and end-to-end delay bound. In our proposal, the source nodes use Earliest Deadline First (EDF) scheduling, while the switch uses First Come First Served (FCFS) to control periodic real-time traffic belonging to logical realtime connections. The schedulability condition is drawn and the end-to-end worst-case delay bound is derived for guaranteed real-time traffic, which also gives, as a sub-result, the needed buffer space in the switch. Moreover, different traffic classes are prioritized and put into different output queues in the end nodes and the switch, which minimizes the interference with other traffic when transmitting periodic time-critical messages. The solution requires no additional hardware or software modifications of the switch and the underlying standard. It is based purely on software implementation in the end nodes. Furthermore, the EDF scheduling strategy used in the source nodes allows good resource utilization, while the FCFS scheduling used in the switch limits the switch processing overhead. The paper presents simulation analysis for Fast Ethernet, which verifies our approach of guaranteeing real-time transmission at the same time as achieving high utilization.