Milutin Radonjic

Impact of scheduling algorithms on performance of crosspoint-queued switch

The performance analysis of the 32 × 32 crosspoint-queued switch is presented in this paper. Switches with small buffers in crosspoints have been evaluated in the late 1980s but mostly for uniform traffic. However, due to technological limitations of that time, it was impractical to implement large buffers together with switching fabric. The crosspoint-queued switch architecture has been recently brought back into focus since modern technology enables an easy implementation of large buffers in crosspoints. An advantage of this solution is the absence of control communication between linecards and schedulers. In this paper, the performances of four algorithms (longest queue first, round robin, exhaustive round robin, and frame-based round robin matching) are analyzed and compared. The results obtained for the crosspoint-queued switch are compared with the output queued switch. Throughput, average cell latency and instantaneous packet delay variance are evaluated under uniform and nonuniform traffic patterns. The results will show that the longest queue first algorithm has the highest throughput in many simulated cases but the highest average cell latency and delay variance among observed algorithms. It will also be shown that the choice of the scheduling algorithm does not play a role in the switch performance if the buffers are long enough. This will prove that some form of round-robin-based algorithms become a better choice for implementation due to their simplicity, small hardware requirements, and avoidance of the starvation problem, which is a major drawback of the longest queue first algorithm.

The IoT Architectural Framework, Design Issues and Application Domains

The challenge raised by the introduction of Internet of Things (IoT) concept will permanently shape the networking and communications landscape and will therefore have a significant social impact. The ongoing IoT research activities are directed towards the definition and design of open architectures and standards, but there are still many issues requiring a global consensus before the final deployment. The paper presents and discusses the IoT architectural frameworks proposed under the ongoing standardization efforts, design issues in terms of IoT hardware and software components, as well as the IoT application domain representatives, such as smart cities, healthcare, agriculture, and nano-scale applications (addressed within the concept of Internet of Nano-Things). In order to obtain the performances related to recently proposed protocols for emerging Industrial Internet of Things applications, the preliminary results for Message Queuing Telemetry Transport and Time-Slotted Channel Hopping protocols are provided. The testing was performed on OpenMote hardware platform and two IoT operating systems: Contiki and OpenWSN.

Buffer length impact to crosspoint queued crossbar switch performance

Performance analysis of two crosspoint queued switches, with four and sixteen ports, is presented in this paper. Crosspoint queued switch architecture has been recently actualized, since it is now not a problem to implement large buffers in crosspoints using modern technology any more. An advantage of this type of solution is the absence of control communication between linecards and schedulers. Four algorithms are used to implement a scheduler: longest queue first, round-robin, frame based round robin matching and random. Throughput, average cell latency and loss probability, are evaluated under uniform and various nonuniform traffic patterns. Results will show that the longest queue first algorithm has the best performance in most simulated cases. It will also be shown that an implemented algorithm does not have any influence on switch performance if the buffers are long enough.

PErformance analysis of LPF based VOQ crossbar switches

In this paper, LPF based schedulers for VOQ crossbar switch implementation through FPGA circuits are analysed. Switch performance (average cell latency, queue lengths and loss probability) are evaluated under different traffic conditions. Results of behavioral simulations on implemented circuits are compared with simulations on modeled circuits that are not limited as implemented ones. Particularly consideration is given to required queue lengths and impact of its limitation to scheduler behavior. It is shown that queue length of 127 cells is enough to manage almost any offered load under various traffic conditions that we considered.

Buffer length impact to 32x32 crosspoint queued crossbar switch performance

Crosspoint queued switch architecture is recently actualized, since modern technology enables an easy implementation of large buffers in crosspoints. An advantage of this solution is the absence of control communication between linecards and scheduler. Performance analysis of crosspoint queued switch with 32 ports is presented in this paper. Four algorithms are used for scheduler implementation: longest queue first, round-robin, frame based round robin matching and random. Throughput, average cell latency and loss probability are evaluated under uniform and various nonuniform traffic patterns. Results show that longest queue first algorithm has best performance in most simulated cases. It is also shown that implemented algorithm has not any influence to switch performance if buffers are long enough.

Automated algorithm for classification of water-flow signals to support Ambient Assisted Living applications

Characterization of a home water-flow signal, as an outcome of an analysis of the most appropriate representation in time, frequency and time-frequency domain, has not been adequately studied; mainly due to lack of such specific need. Nevertheless, being able to characterize a water-flow signal in the home-environment can support Ambient Assisted Living systems to determine the activity of a senior citizen, thus contributing to the daily pattern recognition, as well as to identify critical situation. This has triggered our research activities to develop an automated decision-making algorithm for detection of different water-flows that may be found in an ordinary bathroom. The preliminary results will be presented in this paper. The time-frequency representations are employed to analyze and characterize different water-flow signals, as well as to provide input parameters for the algorithm. These approach and algorithm can be used in numerous applications where automated decision-making scenario is required.

Analysis of WRR scheduling algorithm frame size impact on CQ switch performance

In this paper we study the influence of Weighted Round Robin (WRR) scheduling algorithm frame size to the performance of Crosspoint Queued (CQ) crossbar switch. In order to show that throughput and cell delay can be adjusted with appropriate WRR frame size, we analyzed switch for different values of frame size and under the unbalanced bursty traffic. We show that WRR scheduling algorithm achieve throughput similar to Output Queued switch when frame size is low. Also, with an increase of the frame size, the delay decreases.

Dynamic weighted round robin in crosspoint queued switch

Achievement of desired performance levels (quality of service - QoS) in switches and routers is one of the most important tasks in switching systems. This implies providing guaranteed throughput, cell loss probability, average and maximal latency within the required bounds. To achieve this task in crosspoint queued switch, we implemented the weighted round robin (WRR) algorithm. We showed that WRR algorithm can achieve very good performance levels regarding the throughput and latency, but had major drawback because it require knowledge of arrival traffic. To overcome this problem, we implemented and presented in this paper the dynamic weighted round robin (DWRR) algorithm that can work with unknown arrival traffic. We showed that DWRR can achieve same performance as WRR, without the need for incoming traffic information, which makes it suitable for practical implementation.

Bandwidth-delay constrained routing algorithms for backbone SDN networks

In this paper we analyzed suitability of different routing algorithms for establishing bandwidth-delay constrained traffic tunnels in SDN backbone networks. Bandwidth rejection ratio was used as indicator of algorithms ability to efficiently utilize network resources. Through set of simulations on two real backbone topologies we showed that network topology can affect algorithms performance, and can lead to biased results. Also, we pointed out problem of high complexity of existing traffic engineering (TE) solutions, which often cannot perform calculations in real-time. Since in SDN networks the controller processing latency can turn out to be a non-negligible factor in the total round-trip latency, we explained heuristic techniques for reducing computation time of bandwidth-delay constrained algorithms and provided illustrative results showing how they impact BRR performance.

Performance evaluation of Dual Crosspoint Queued crossbar packet switch

This paper presents an analysis of new Dual Crosspoint Queued (DCQ) crossbar switch architecture. It is well-known that in order to achieve performance guarantees, the priority separation of incoming traffic is necessary. Therefore, we modified CQ crossbar switch with the separation of different traffic flows, by implementing two buffers in each crosspoint of switch fabric. Different round robin based scheduling algorithms are presented and analyzed. With the particular attention on the cell delay and cell loss probability, results of simulations show that proposed switch architecture can achieve desired QoS only for the high priority traffic flow.