C. Vassilakis, P. Georgiadis, D. Lelis, D. Mouzakis and M. Nikolaidou
Hellenic Informatics Conference, Patra 1993 (in Greek).

Abstract:
As the use of distributed systems is spreading, and applications designed for such systems become more and more demanding, optimal design of distributed systems becomes a critical issue. Designing a distributed system has become more complex, due to the number of alternatives for each decision that must be made and because of the existence of many parameters which influence the overal performance of the distributed system. Thus, it is necessary to use software tools, capable of accepting a description of the user's requirements and suggesting solutions to the problem of designing a distributed system which meets the user's requirements. In this paper, we present a disciplined approach to the construction of such a software tool, which combines methods from the Artificial Intelligence domain, that are used in order to design the distributed system, along with simulation techniques, used to estimate the system's overall performance.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

C. Boukouvalas, P. Georgiadis and C. Vassilakis
Hellenic Informatics Conference, Patra 1993 (in Greek).

Abstract: In this study we present an Enhanced File System for a Distributed Unix Environment. The Enhanced Unix File System implements a flexible protection mechanism for files and directories. It is based on the concept of Access Control Lists (ACL), which allows different permissions for files and directories to be given to specific users. His work was developed under the SunOS/NFS distributed environment, using Remote Procedure Calls (RPC) to implement the communication between the client and the server. The system consists of a daemon process working as a server and a set of client processes that provide different file and directory services. There may be running many different servers and clients over the same network. The client processes transparently locate the appropriate server for each transaction. The system also provides an open programming environment for the development of new applications. Both the user and the programmer interface of the system are kept close to standard UNIX.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis,
To appear in Journal of Computer Networks and Communications

Abstract:
Contemporary networks accommodate handling of multiple priorities, aiming to provide suitable QoS levels to different traffic classes. In the presence of multiple priorities, a scheduling algorithm is employed to select each time the next packet to transmit over the data link. Class-based Weighted Fair Queuing (CBWFQ) scheduling and its variations, is widely used as a scheduling technique, since it is easy to implement and prevents the low-priority queues from being completely neglected during periods of high-priority traffic. By using this scheduling, low-priority queues have the opportunity to transmit packets even though the high-priority queues are not empty. In this paper, the modeling, analysis and performance evaluation of a single-buffered, dual priority multistage interconnection network (MIN) operating under the CBWFQ scheduling policy is presented. Performance evaluation is conducted through simulation, and the performance measures obtained can be valuable assets for MIN designers, in order to minimize the overall deployment costs and delivering efficient systems

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
to appear in International Journal of Parallel, Emergent and Distributed Systems, Taylor & Francis

Abstract:

Contemporary networks support multiple priorities, aiming to differentiate the QoS levels offered to individual traffic classes. Support for multiple priorities necessitates the introduction of a scheduling algorithm, to select each time the next packet to transmit over the data link. Class-based Weighted Fair Queuing (CBWFQ) scheduling and its variations, is widely used as a scheduling technique, since it is easy to implement and prevents the low-priority queues from starvation, i.e. receiving no service during periods of high-priority traffic. CBWFQ effectively thus offers low-priority queues the opportunity to transmit packets even though the high-priority queues are not empty. In this paper, we present the modeling and performance evaluation of a single-buffered, four-priority multistage interconnection network (MIN) operating under the CBWFQ scheduling policy. Performance evaluation is conducted through simulation, and the performance metrics obtained can be used by MIN designers to set the appropriate queue weights according to the expected traffic and the desired QoS levels for each priority class, delivering efficient thus systems.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

P. Kassapidis, C. Vassilakis, M. Nikolaidou, P. Georgiadis, Gr. Votsis and N. Pronios
Hellenic Informatics Conference, Patra 1993 (in Greek).

Abstract:
The forthcoming use of multimedia applications will require powerful computers and high performance networks. While great progress has been made to the physical layer of networks, the upper software layers of the OSI reference model have not kept up pace with them. This paper presents the demands imposed by multimedia applications on the underlying networks and their mapping to transport layer services, that protocols implementing it must provide. Four well-known transport layer protocols are briefly presented (TCP, TP4, VMTP, HSTP/XTP). The mechanisms employed by each one of those are studied and their suitability for demanding multimedia environments are evaluated. We conclude that much more effort needs to be made on the transport layer protocols so that high performance network architectures will be available to fullfill the diverse requirements of tomorrow's multimedia applications.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Dimitrios Theotokis, George-Dimitrios Kapos, Costas Vassilakis, Anya Sotiropoulou and Georgios Gyftodimos
Proceedings of the IEEE Workshop on Future Trends on Distributed Computing, Cape Town, 1999, pp. 95-101.

Abstract:
Distributed software systems need to evolve according to the ever-changing requirements on which they were built. Software systems tailorability can be achieved in terms of component software. Atoms and molecules the basic constructs of the atoma framework, are the building blocks for distributed tailorable component-based software systems. These constructs can be considered as independent agents, that communicate in terms of, unanticipated, connections that are established at run-time, thus forming agent communities. System tailorability can take place at two levels. In high level tailorability whole parts of the functionality of a system, represented as agents, can be altered in order to provide new functionality. At a lower level, the tailorability of an agent itself, that is the tailorability of its functionality, is achieved through a flexible service mapping implementation for rule-based method invocation.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
AIP conference proceedings 1060 (procs. of IECCS 2007), pp. 379-384, 2008

Abstract:
In this paper, a wireless multimedia traffic-oriented network scheme over a fourth generation system (4-G) is presented and analyzed. We conducted an extensive evaluation study for various mobility configurations in order to incorporate the behavior of the IEEE 802.11b standard over a test-bed wireless multimedia network model. In this context, the Quality of Services (QoS) over this network is vital for providing a reliable high-bandwidth platform for data-intensive sources like video streaming. Therefore, the main issues concerned in terms of QoS were the metrics for bandwidth of both dropped and lost packets and their mean packet delay under various traffic conditions. Finally, we used a generic distance-vector routing protocol which was based on an implementation of Distributed Bellman-Ford algorithm. The performance of the test-bed network model has been evaluated by using the simulation environment of NS-2.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis,
Proceedings of the Fourth Advanced International Conference on Telecommunications AICT 2008, IEEE Press

Abstract:
In this paper the performance of asymmetric-sized finite-buffered Delta Networks with 2-class routing traffic is presented and analyzed in the uniform traffic conditions under various loads using simulations. We compared the performance of 2-class priority mechanism against the single priority one, by gathering metrics for the two most important network performance factors, namely packet throughput and delay. We also introduce and calculate a universal performance factor, which includes the importance aspect of each of the above main performance factors. We found that the use of asymmetric-sized buffered systems leads to better exploitation of network capacity, while the increments in delays can be tolerated. The goal of this paper is to help network designers in performance prediction before actual network implementation and in understanding the impact of each parameter factor.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis,
International Journal on Information Technologies & Security, no 3, 2010 Abstract:

Wireless Local Area Networks (WLANs) have developed into a viable technology to support multimedia traffic and are expected to support multimedia services with guaranteed Quality of Service (QoS) for diverse traffic types (video, audio, and data). In this paper, we consider the incorporation of prediction into a generic distance-vector routing protocol for WLANs, evaluate the performance of the resulting routing scheme. Our study considers the enhancement of Distributed Bellman-Ford algorithm, which is a widely used algorithm, and assesses the effectiveness of the enhanced version on top of a fourth generation system (4-G). In order to compare the performance of the standard protocol against that of the prediction-enhanced version, we gather metrics for the two most important network performance factors, namely packet throughput and delay under different mobility and traffic conditions, using the simulation environment of NS-2. Both medium- and high-mobility configurations have been considered in this study.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis and E. Glavas
International Journal of Parallel, Emergent and Distributed Systems, Vol. 26, No. 5, October 2011, 381–397, Taylor & Francis Abstract:

In this paper, the modelling, analysis and performance evaluation of a novel architecture for internal priority finite-buffered Multistage Interconnection Networks (MINs) is presented. We model the proposed architecture giving the details of its operation and describing its states and detailing conditions and effects of state transition; we also provide a formal model for evaluating its performance. The proposed architecture’s performance is subsequently analyzed under the uniform traffic condition, considering various offered loads, buffer-lengths and MIN sizes, using simulations. We compare the internal priority scheme vs. the non priority (or single priority) scheme, by gathering metrics for the two most important network performance factors, namely packet throughput and the mean time a packet needs to traverse the network. We demonstrate and quantify the improvements on MIN performance stemming from the introduction of priorities in terms of throughput and a combined performance indicator which depicts the overall performance of the MIN. These performance measures can be valuable assets for designers of parallel multiprocessor systems and networks in order to minimize the overall deployment costs and delivering efficient systems.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis,
Journal of Networks, June 2011

Abstract
Next-generation network architectures strive to achieve high bandwidth and ultralow latency for the packets traversing the offered end-to-end paths. Multistage Interconnection Networks (MINs) are often employed for implementing NGNs, but while MINs are fairly flexible in handling varieties of traffic loads, they tend to quickly saturate under broadcast and multicast traffic, especially at increasing size networks. As a response to this issue, multilayer MINs have been proposed, however their performance prediction and evaluation has not been studied sufficiently insofar. In this paper, we evaluate and discuss the performance of multilayer MINs under multicast traffic, considering also two levels of packet priorities, since support for multiple QoS levels is an indispensible requirement for NGNs. Different offered loads and buffer size configurations are examined in this context, and performance results are given for the two most important network performance factors, namely packet throughput and delay. We also introduce and calculate a universal performance fac¬tor, which includes the importance aspect of each of the above main performance factors. The findings of this study can be used by NGN system designers in order to predict the performance of each configuration and adjust the design of their communication infrastructure to the traffic requirements at hand.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
Journal of Computer Systems, Networks, and Communications, Hindawi Publishing Corporation

Abstract:
In this paper the performance of multi-layered asymmetric-sized finite-buffered Delta Networks supporting multi-class routing traffic is presented and analyzed in the uniform traffic conditions under various loads using simulations. The rationale behind introducing asymmetric-sized buffered systems is to have a better exploitation of available buffer spaces, while the implementation of multi-layered architecture is applied in order to further improve the overall performance of network. The findings of this performance evaluation can be used by network designers for drawing optimal configurations while setting up the network, so as to best meet the performance and cost requirements under the anticipated traffic load and quality of service specifications.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D.C. Vasiliadis, G.E. Rizos, and C. Vassilakis
Proceedings of the 7th IBIMA Conference.

Abstract:
Multistage Interconnection Networks (MINs) are frequently used for connecting processors in parallel computing systems or constructing high speed networks such as ATM (based on Asynchronous Transfer Mode) and Gigabit Ethernet Switches. New applications require distributed computing implementations, but old networks are too slow to allow efficient use of remote resources. Moreover, multimedia are considered as applications with high bandwidth requirements. Some of them are also sensitive to packet loss and claim reliable data transmission. Specific applications require bulk data transfers for database replication or load balancing and therefore packet loss minimization is necessary in order to increase the performance of them. The demand for high performance multimedia services such as full motion video on demand is becoming an increasingly important driving force in the communication market in the Digital Age. Thus, the performance of MINs is a crucial factor, which we have to take into account in the design of new applications. Their performance is mainly determined by their communication throughput and cell latency, which have to be investigated either by time-consuming simulations or approximated by mathematical models. In this paper we investigate the performance of MINs in order to determine optimal values for hardware parameters under diferent operating conditions.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis , G. E. Rizos , C. Vassilakis
Proceedings of ICNS 07, IEEE press

Abstract:

In this paper a novel architecture of dual priority single-buffered blocking Multistage Interconnection Networks (MINs) is presented. We analyzed their performance in the uniform traffic condition under various loads using simulations. We compared the dual priority architecture against a single priority MIN, by gathering metrics for the two most important network performance factors, namely packet throughput and the mean time a packet needs to traverse the network. We demonstrated the gain of the high priority packets against the low priority packets under different configuration schemas. In this paper we focus on studying the influence of the priority bit in the header field of transmitted packets on the performance of high and low priority traffic of a MIN. Performance prediction before actual network implementation and understanding the impact of parameter settings in a MIN setup are valuable assets for network designers for minimizing overall deployment costs and delivering efficient networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis, E. Glavas
Proceedings of The 18th Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC'07), CAMAD 07 Workshop

Abstract:
In this paper a novel two-priority network schema is presented, and exemplified through its application on single-buffered Delta Networks in packet switching environments. Network operations considered include conflict resolution and communication strategies. The proposed scheme is evaluated and compared against the single-priority scheme. Performance evaluation was conducted through simulation, due to the complexity of the model, and uniform traffic conditions were considered. Metrics were gathered for the two most important network performance factors, namely packet throughput and the mean time a packet needs to traverse the network. The model can also be uniformly applied to several representative networks providing a basis for fair comparison and the necessary data for network designers to select optimal values for network operation parameters.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
Proceedings of CISSE 2007

Abstract:
In this paper, a wireless Circular Model over the Distance Vector routing protocol is presented and analyzed. The performance of this algorithm, which is an implementation of Distributed Bellman-Ford algorithm has been evaluated by using the simulation environment of NS-2. We conducted an extensive evaluation study for various mobility schemes in order to incorporate the behavior of nodes and the routing protocol in a real-life hotspot situation. In the test-bed model, while the number of source nodes was allowed to arbitrarily vary, there was exactly one destination node, closely modeling thus real-life situations where a single hotspot/access point exists. Finally, different constant bit rates (CBR) were used in order to estimate the throughput of receiving, dropping rates, the number of lost packets, as well as the average packet delay under various traffic conditions. This study is aimed to help wireless network designers in choosing the best suited routing protocols for their networks, through making explicit performance figures for common network setups.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis, E. Glavas
Proceedings of the Fifth Advanced International Conference on Telecommunications (AICT 2009)

Abstract:
Multilayer MINs have emerged mainly due to the increased need for routing capacity in the presence of multicast and broadcast traffic, their performance prediction and evaluation however has not been studied sufficiently insofar. In this paper, we use simulation to evaluate the performance of multilayer MINs with switching elements of different buffer sizes and under different offered loads. The findings of this paper can be used by MIN designers to optimally configure their networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
Journal of Computer Systems, Networks, and Communications, (to appear), Hindawi Publishing Corporation

Abstract:
The performance of Multistage Interconnection Networks (MINs) under hotspot traffic, where some percentage of the traffic is targeted at single nodes, which are also called hot-spots, is of crucial interest. The prioritizing of packets has already been proposed at previous works as alleviation to the tree saturation problem, leading to a scheme that natively supports 2-class priority traffic. In order to prevent hotspot traffic from degrading uniform traffic we expand previous studies by introducing multi-layer Switching Elements (SEs) at last stages in an attempt to balance between MIN performance and cost. In this paper the performance evaluation of dual-priority, double-buffered, multi-layer MINs under single hotspot setups is presented and analyzed using simulation experiments. The findings of this paper can be used by MIN designers to optimally configure their networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis,
Journal of Communications Software and Systems, vol. 9, no. 4, December 2013, pp. 222-235
Abstract:

Differentiated Services (DiffServ) and other scheduling strategies are now widespread in the traditional, “best effort” Internet. These Internet Architectures offer Quality of Service (QoS) guarantees for important customers at the same time as supporting less critical applications of lower priority. Strict priority queuing (PQ), weighted round robin (WRR), and class-based weighted fair queuing (CBWFQ) are three common scheduling disciplines for differentiation of services in telecommunication networks. In this paper, a comparative performance study of the above PQ, WRR and CBWFQ queuing scheduling policies applied on a double-buffered, 6-stage Multistage Interconnection Network (MIN) that natively supports a 2-class priority mechanism is presented and analyzed using simulation experiments. We also consider a 10-stage MIN, to validate that the conclusions drawn from the 6-stage MIN apply to MINs of different sizes. The findings of this paper can be used by MIN designers to optimally configure their networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D.C. Vasiliadis, G.E. Rizos, and C. Vassilakis
Proceedings of the IEEE sponsored International Joint Conference on Computer, Information and System Sciences and Engineering - CIS²E 06.

Abstract:
Banyan Networks are a major class of Multistage Interconnection Networks (MINs). They have been widely used as efficient interconnection structures for parallel computer systems, as well as switching nodes for high-speed communication networks. The performance of them is mainly determined by their communication throughput and their mean packet delay. In this paper we use a model that is based on a universal performance factor, which includes the importance aspect of each of the above main performance factors (throughput and delay) in the design process of a MIN. The model can also uniformly be applied to several representative networks. The complexity of the model requires to be investigated by time-consuming simulations. In this paper we study a typical (8X8) Baseline Banyan Switch that consists of (2X2) Switching Elements (SEs). The objective of this simulation is to determine the optimal buffer size for the MIN stages under different conditions.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis, E. Glavas
Proceedings of ICSNC 2008

Abstract:
In this paper the modeling of Omega Networks supporting multi-class routing traffic is presented and their performance is analyzed. We compare the performance of multi-class priority mechanism against the single priority one, by gathering metrics for the two most important network performance factors, namely packet throughput and delay under uniform traffic conditions and various offered loads, using simulations. Moreover, two different test-bed setups were used in order to investigate and analyze the performance of all priority-class traffic, under different Quality of Service (QoS) configurations. In the considered environment, Switching Elements (SEs) that natively support multi-class priority routing traffic are used for constructing the MIN, while we also consider double-buffered SEs, two configuration parameters that have not been addressed insofar. The rationale behind introducing a multiple-priority scheme is to provide different QoS guarantees to traffic from different applications, which is a highly desired feature for many IP network operators, and particularly for enterprise networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
Proceedings of Conference on Advances in Future Internet (AFIN 2009), IEEE Press

Abstract:
Large swings in the demand for content are commonplace within the Internet. Although Multistage Interconnection Networks (MINs) are fairly flexible in handling varieties of traffic loads, their performance considerably degrades by hotspot traffic, especially at increasing size networks. As alleviation to the tree saturation problem, the prioritizing of packets is proposed leading to a scheme that natively supports multi priority traffic. In this paper the performance evaluation of double-buffered Delta Networks under single hotspot setups, with different offered loads, and 2-class routing traffic is presented and analyzed using simulation experiments. Performance comparison of dual vs. single priority scheme is outlined under hotspot environment, by calculating a universal performance factor, which effectively includes the importance aspect of each of the two most important performance metrics, namely packet throughput and delay. The findings of this paper can be used by MIN designers to optimally configure their networks.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

D. C. Vasiliadis, G. E. Rizos, C. Vassilakis
Proceedings of ICCMSE 2007

Abstract:
The main concerns in designing the multistage switching fabrics are speed, throughput, delay and variance of delay for a given bandwidth. The rationale behind using various priority mechanisms is either to offer different quality of service levels to packets or to optimize performance parameters of the network, e.g. minimize internal blocking in the Switching Elements (SEs). We investigated the performance parameters of an enhanced priority (EP) mechanism versus a single priority (SP) one. In the EP scheme, packet priority was computed dynamically and was directly proportional to the transmission queue length of the SE that the packet is currently stored in. Finally, we extended the idea of the priority scheme by proposing a multi-priority (MP) mechanism. In the MP scheme, each SE has two transmission queues per link, with one queue dedicated to high priority packets and the other dedicated to low priority ones. We simulated a multistage network under the uniform traffic condition and concluded that the proposed double-buffered SEs provide higher throughput, and decreased latency.

Note: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.