Abstract:This paper presents a protocol to support hard real-time traffic of end-to-end communication over non real-time LAN technology. The network is set up with nodes and switches, and real-time communication is handled by software (protocol) added between the Ethernet protocols and the TCP/IP suite. The proposed protocol establishes a virtual circuit based on admission control and manages hard real-time traffic to bypass the TCP/IP stack. This makes considerably reduce the dwell time in the nodes, and increase the … Show more
“…The interested reader is referred to [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Here, we focus on RT-NoCs, which have often been called guaranteed-service or QoS-enabled NoCs.…”
Abstract-Real-time (RT) communication support is a critical requirement for many complex embedded applications which are currently targeted to Network-on-chip (NoC) platforms. In this paper, we present novel methods to efficiently calculate worst case bandwidth and latency bounds for RT traffic streams on wormhole-switched NoCs with arbitrary topology. The proposed methods apply to best-effort NoC architectures, with no extra hardware dedicated to RT traffic support. By applying our methods to several realistic NoC designs, we show substantial improvements (more than 30 percent in bandwidth and 50 percent in latency, on average) in bound tightness with respect to existing approaches.
“…The interested reader is referred to [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Here, we focus on RT-NoCs, which have often been called guaranteed-service or QoS-enabled NoCs.…”
Abstract-Real-time (RT) communication support is a critical requirement for many complex embedded applications which are currently targeted to Network-on-chip (NoC) platforms. In this paper, we present novel methods to efficiently calculate worst case bandwidth and latency bounds for RT traffic streams on wormhole-switched NoCs with arbitrary topology. The proposed methods apply to best-effort NoC architectures, with no extra hardware dedicated to RT traffic support. By applying our methods to several realistic NoC designs, we show substantial improvements (more than 30 percent in bandwidth and 50 percent in latency, on average) in bound tightness with respect to existing approaches.
“…The HASIS (Hardware Software Interface Synthesis) communication channel supporting message scheduling in reference [13] is based on weighted fair queues for avoiding starvation. Reference [14] proposes a hybrid scheduling scheme based on non-preemptive EDF (Earliest Deadline First) algorithm in order to meet message deadlines and maximize network utilization. We implement and evaluate the scheduling schemes presented by reference [13][14] to compare the performances of providing QoS functionality with our proposed technique.…”
Section: Motivationmentioning
confidence: 99%
“…Reference [14] proposes a hybrid scheduling scheme based on non-preemptive EDF (Earliest Deadline First) algorithm in order to meet message deadlines and maximize network utilization. We implement and evaluate the scheduling schemes presented by reference [13][14] to compare the performances of providing QoS functionality with our proposed technique. In the research work presented in this paper, we attempt to reconsider these features, which have been reviewed so far in this section at different concepts, and then we succeed in building a SoC platform for real-time embedded network/multimedia systems that can resolve a decomposing problem of implementing any network protocol suite into real-time application-specific and reconfigurable hardware/software tasks.…”
Section: Motivationmentioning
confidence: 99%
“…4) Decomposed-TCP/IP with RTHC-HASIS is identical with the Decomposed-TCP/IP with RTHC-FIFO except for using the scheduling scheme proposed by [13]. 5) Decomposed-TCP/IP with RTHC-EDF is identical with the Decomposed-TCP/IP with RTHC-FIFO except for using the scheduling scheme proposed by [14]. 6) Decomposed-TCP/IP with RTHC-PIT is identical with the Decomposed-TCP/IP with RTHC-FIFO except for using the scheduling scheme with the PIT module except the DST module.…”
Section: Analysis Of Response Time Of Aperiodic Hcmmentioning
In this paper, we propose a novel hardware-software codesign technique of network protocol stacks on a SoC (System on Chip) platform for providing the QoS (Quality of Service) functionality. We address that the codesign should guarantee the real-time performance of periodic packets and minimize the average response time of aperiodic packets in order to provide the QoS functionality. We present the following three mechanisms to resolve this issue: 1) task decomposition technique of a network protocol suite into hardware and software tasks, 2) a communication channel between heterogeneous tasks, 3) finegrained and coarse-grained real-time processing simultaneously for application-specific real-time requirements. A case study based on experiments is conducted to illustrate the efficiency of the proposed technique by implementing it on the commercial SoC platform embedded with the Altera's Excalibur EPXA4 including ARM922T core and 400K gates of programmable logic.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.