Khurram Aziz

Position-based and geocast routing protocols in VANETs

We study position-based and geocast routing protocols, which are two categories of Vehicular Ad-hoc Network (VANETs) routing protocols based on the information of position and location respectively. Two scenarios for position-based routing are discussed — the highway scenario and the city scenario. We also study geocast routing protocols that are based on flooding by defining the forwarding zone. Forwarding zone is the region where the destination node is located. The advantages and challenges for these position-based and geocast routing protocols are also discussed followed by a discussion on the suitability of the position-based protocols for the highway and city scenarios.

Dimensioning an optical packet/burst switch — More interconnections or more delay lines?

We present performance evaluation of an optical packet/burst switch with internal fiber rings. Such a switch consists of multiple fabrics linked together by direct interconnections and optical delay lines. Having more number of optical delay lines and direct interconnections per fabric pair reduces contention and results in a reduced blocking probability. However, due to the size and cost of optical delay lines, it is desirable to minimize their number for a specific blocking probability. We propose an analytical model for the switch in order to study its scalability and packet/burst blocking probability. The switch is then dimensioned for given values of blocking probability and a given range of input load. Simulation results concerning switch scalability by taking into account the most relevant physical effects are also shown. We then consider the minimal practical switch size and discuss how the switch can be dimensioned with a combination of delay lines and direct interconnections to achieve our target of a certain blocking probability and range of input load using a minimum number of fiber delay lines.

CloudNetSim++: A GUI Based Framework for Modeling and Simulation of Data Centers in OMNeT++

State-of-the-art cloud simulators in use today are limited in the number of features they provide, lack real network communication models, and do not provide extensive Graphical User Interface (GUI) to support developers and researchers to extend the behavior of the cloud environment. We propose CloudNetSim++, a comprehensive packet level simulator that enables simulation of cloud environments. CloudNetSim++ can be used to evaluate a wide spectrum of cloud components, such as processing elements, storage, networking, Service Level Agreement (SLA), scheduling algorithms, fine grained energy consumption, and VM consolidation algorithms. CloudNetSim++ offers extendibility, which means that the developers and researchers can easily incorporate own algorithms for scheduling, workload consolidation, VM migration, and SLA agreement. The simulation environment of CloudNetSim++ offers a rich GUI that provides a high level view of distributed data centers connected with various network topologies. The package also includes an energy computation module that provides a fine grained analysis of energy consumed by each component. This paper shows the flexibility and effectiveness of CloudNetSim++ through experimental results demonstrated using real-world data center workloads. Moreover, to demonstrate the correctness of CloudNetSim++, we performed formal modeling, analysis, and verification using High-level Petri Nets, Satisfiability Modulo Theories (SMT), and Z3 solver.

Optical Interconnects for Data Center Networks

Traditional data center networks built with copper wires and electronic elements suffer from various problems. These include high energy consumption due to the wired architecture, high latency due to extra hops adding to the routing delay, fixed throughput of links, and very limited configurability. Data center networks built with optical fibers and optical components would solve all of these problems but they suffer from issues of their own including higher cost, immaturity of optical components, lack of optical buffers and complexity of design. It is clear however, that optical interconnects will replace their electronic counterparts in all data center network architectures due to their superior properties.

Performance of Non-exhaustive Cyclic Service Systems with Finite Queues and Non-zero Switchover Times

Queueing systems with cyclic service have a broad range of applications in communication systems, e.g., in switching systems, ring networks, burst assembly in OBS and traffic aggregation in edge nodes. While such systems have been extensively studied, the buffer capacity is often assumed to be unlimited. In real systems, queue sizes are always limited. We present an analytical method using state-space modelling, for obtaining the performance of cyclic service systems with finite queues and non-exhaustive service. The effect of switchover time has also been taken into account. We also show the effect of varying traffic loads and queue sizes on the observed queue.

Performance analysis of optical interconnects' architectures for data center networks

Electrical interconnects in Data Center Networks (DCNs) suffer from various problems which include high energy consumption, high latency, fixed throughput of links and limited reconfigurability. Introducing optical interconnects in DCNs help to reduce these problems to a large extent. Optical interconnects are the technology of the future. To implement optical switching in DCNs various optical components are used which include wavelength selective switch, tunable wavelength converter, arrayed waveguide grating, semiconductor optical amplifier based switch, wavelength division multiplexers and demultiplexers. All these optical components vary the shape, attenuate the optical signal and introduce time delay in bits. A comprehensive study of various architectures for optical interconnects in data center networks (DCN) is carried out. Performance of various architectures is investigated in terms of jitter, bit error rate (BER), receiver sensitivity and eye diagram opening. It is also investigated how different optical components used in optical interconnects in DCNs are effecting the signal degradation in different architectures. The paper concludes with the categorization of the signal degradation types in optical interconnects in DCNs and ways to reduce them. This enables the design of low BER optical interconnects in DCNs.

Performance evaluation of hybrid optical switching with quality of service

Optical Circuit Switching (OCS) is a switching technique which has been used in the core of the optical network for many years but Optical Packet Switching (OPS) and Optical Burst Switching (OBS) are also some of the promising switching techniques fast emerging in the optical network. All of these techniques have some advantages and disadvantages. Flow Transfer Mode (FTM) is a hybrid optical switching technique that aims to leverage strengths of these switching techniques and integrates all of these techniques into a single switching technique. It classifies incoming traffic into different data flows. It is also considered as an extension of the OBS. In this paper, we evaluate the performance of the FTM through simulation and perform comparative analysis with the OBS under same simulation parameters. In future, internet may ultimately be constrained by the capability to provide quality of service (QoS), so we also propose and employ QoS provisioning in the FTM and evaluate its performance with QoS. Our results show noticeable improvement in burst loss ratio, bandwidth utilization and throughput.

Blocking probability in optical interconnects in data center networks

Cloud computing and Web-based applications are creating a need for powerful data centers. Data centers have a great need for high bandwidth, low latency, low blocking probability, and low bit-error rate to sustain the interaction between different applications. Current data center networks (DCNs) suffer from several problems such as high-energy consumption, high latency, fixed throughput of links, and limited reconfigurability. Electronic switches are low radix and have high latency due to a large hop count since each hop employs a store-and-forward mechanism. Optical interconnects, on the other hand, offer several advantages such as low-energy consumption, high bandwidth, reconfigurability, malleability to changing traffic, high-radix switch design, fast switching transition times, and wavelength multiplexing. These benefits provide the incentive to shift from electrical interconnects to optical interconnects in DCNs. Despite several advantages over their electrical counterparts, the performance of optical interconnects can be further improved by considering some performance parameters of optical interconnects. One such important parameter for the performance of any communication network is the blocking probability. This paper makes a comprehensive investigation of the performance of optical interconnects in different DCN architectures on the basis of blocking probability and concludes by suggesting ways to reduce the blocking.

Classification of Optical Interconnects in Data Center Networks

Traditional Data Center Networks (DCNs) are suffering from many problems which include high energy consumption, high latency, fixed throughput of links and limited reconfigurability to the traffic demands. This leads towards the a change in the paradigm from electrical to optical interconnects between the servers. Compared to electrical counterparts optical interconnects give various advantages which include low energy consumption, high bandwidth, reconfigurability, high radix switch design, wavelength multiplexing, and fast switching transition times. Now a days there are a large number of emerging cloud computing applications which include video streaming (YouTube, Google Video etc), Search Engines (Google, Bing etc), Social Networking (Face book, Google Plus etc), Email (Gmail, Yahoo Mail, Hotmail etc), Cloud Computing, Climate and Geographic Information Systems. Each of these application has its own latency, blocking, reconfigurability and radix requirement. Thus classification of architectures is necessary to decide which architecture suits the demands of which application. In this paper we have classified the optical interconnects in DCNs on the basis of their inherent nature. Basic optical components form the basis for the optical interconnects. This paper has following sections, Section I gives a motivation for the optical interconnects in DCNs, Section II highlights basic optical components used in optical interconnects and Section III gives a classification tree of optical Interconnects in DCNs.

Quality of Service in Hybrid Optical Switching

Network traffic is continuously increasing over the internet due to emerging end user applications provided by cloud computing. Optical networks are hence rapidly gaining exclusivity in access networks and data center networks as they provide huge bandwidth support. Optical switching techniques like Optical Circuit Switching (OCS), Optical Packet Switching (OPS) and Optical Burst Switching (OBS) have some advantages and limitations. Flow Transfer Mode (FTM) is hybrid optical switching technique that aims to utilize the benefits of existing switching techniques. It is a universal switching method that classifies traffic flows into different modes. In this paper, we evaluate the performance of FTM and propose QoS provisioning in FTM as QoS provisioning is a desired feature of future next generation network. Our results show improvement in burst loss ratio, bandwidth utilization and throughput.