2. Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/1/7
Browse
14 results
Search Results
Item BBRvl vs BBRv2: Examining Performance Differences through Experimental Evaluation(2020) Nandagiri A.; Tahiliani M.P.; Misra V.; Ramakrishnan K.K.BBR, a congestion control algorithm proposed by Google, regulates the source sending rate by deriving an estimate of the bottleneck's available bandwidth and RTTof the path. The initial version of BBR, called BBRvl, was found to be unfair, getting higher than the fair share of bandwidth when co-existing on bottleneck links with other congestion control algorithms. It also does not perform as well with networks having routers with shallow buffers. To overcome these concerns, a newer version, called BBRv2, has been proposed. Our goal in this paper is to understand the differences between the two versions and examine the primary reasons behind the improvement in performance of BBRv2. We present an experimental evaluation of BBRvl and BBRv2, evaluating their fairness across connections using the same protocol (intra-protocol fairness) and using different protocols (inter-protocol fairness) as well as delay and link utilization. From experiments with shallow and deep buffers, BBRv2 is most effective when it uses Explicit Congestion Notification (ECN), but fairness issues continue to exist in BBRv2 when ECN is disabled. A concern for BBRv2 is that it is somewhat complex to deploy in Wide Area Networks (WAN) because of the dependency with the DCTCP-style reduction of the congestion window, which is primarily usable in low-feedback delay Data Center Networks. © 2020 IEEE.Item Teaching EARS to undergrads in the pandemic - Industry academia experience(2020) Nair G.V.; Jeppu Y.; Tahiliani M.P.The COVID-19 pandemic is rampant in India and this has changed the way the students and teachers interact with each other during a course. An added complexity is the introduction of the Industry Academia participation in terms of Adjunct Faculties. Teaching formal methods to undergraduate students has been difficult and these are well captured in the academic community. The necessity of good requirements writing which can be validated using formal methods is a need of the hour for the industry. Requirements error contribute to 70% of the errors in safety critical projects. A course on Formal Methods is offered at the National Institute of Technology Karnataka, Surathkal as an undergraduate elective. This has 54 students registered and the course is offered online by an adjunct faculty from the industry. The experiences of capturing and writing good requirements using the EARS (Easy Approach to Requirements Syntax) is highlighted in this paper. A survey of before and after the class and an exercise on EARS notations are brought out. The lessons learnt and the efficacy of the teaching is brought out as a three perspective: student, academia and industry. © 2020 IEEE.Item PowerDPDK: Software-Based Real-Time Power Measurement for DPDK Applications(2020) Shah M.; Yunus M.; Vachhani P.; Monis L.; Tahiliani M.P.; Talawar B.Data Plane Development Kit (DPDK) provides a set of libraries for fast packet processing that allow applications in the user space to directly interact with the NIC. Currently, DPDK provides a power management library that enables the applications to save power. However, it lacks features to effectively measure the power consumption of the system. In this paper we propose PowerDPDK, a software-based real-time library to measure the power consumption of DPDK applications. PowerDPDK leverages the Running Average Power Limit (RAPL) feature available on modern Intel processors to provide the power consumed by the CPU package and DRAM. We discuss the architecture of PowerDPDK and describe the process to incorporate it into DPDK applications. Subsequently, we use PowerDPDK to measure the power consumption of a few sample DPDK applications and a chain of Virtual Network Functions (VNFs) in OpenNetVM, a high-performance container-based platform for Network Function Virtualization (NFV). We show that a major share of the power consumed by DPDK is due to the use of Poll Mode Drivers (PMD), and hence, even a simple Layer 2 forwarding application consumes a large amount of power. © 2020 IEEE.Item On the Importance of Traffic Control Subsystem in ICN-based Industrial Networks(2020) Nagaraj A.H.; Kataria B.; Sohoni A.; Tahiliani M.P.; Tandur D.; Satheesh H.The Industrial Automation Control Systems (IACS) are currently dominated by IP-based protocols. Industry 4.0 demands an efficient and scalable networking infrastructure that facilitates data sharing to drive operational improvements and develop business intelligence. The deterministic requirements in industrial networks have led to the emergence of a new IEEE standard in the form of Time-Sensitive Networking (TSN). TSN enables having an upper bound on the latency of data communication, thereby partially fulfilling the requirements of Industry 4.0. However, TSN alone cannot provide the level of determinism required in industrial networks. The efficiency of the layers above TSN can significantly affect the network performance. Information Centric Networking (ICN), which contrasts with IP-based protocols by focusing on the data rather than on the endpoints, is emerging as a promising network layer paradigm. In this paper, we evaluate what it takes for ICN to be integrated with IACS, and thereby meet the requirements of industrial networks. © 2020 IEEE.Item NeST: Network Stack Tester(2020) Rai S.S.; Narayan G.; Dhanasekhar M.; Monis L.; Tahiliani M.P.Linux network namespaces are a cost-effective and scalable alternative to physical systems for the design and experimental evaluation of network protocols. These evaluations are required for a practical understanding of how various networking algorithms would perform in the real world. However, manually setting up testbeds and obtaining results in the desired format using network namespaces can be quite cumbersome and error-prone. Although writing scripts could make these tasks easier, it becomes tedious and impractical if the network under consideration is large and complex. In this paper, we propose a python based package called NeST (Network Stack Tester) to perform tests for different congestion control algorithms and queue disciplines. It uses Linux network namespaces and provides APIs to create complex emulated networks, run tests and extract the statistics using iproute2 and netperf in a single python script. We validate the results obtained from NeST against those obtained from a physical testbed, and a virtual testbed setup manually by using network namespaces. The experiments with NeST are easy to reproduce because it is a wrapper around the existing tools and does not introduce new system dependencies. © 2020 ACM.Item Linux-like Loss Detection Techniques for ns-3 TCP(2019) Bakshi S.; Sahoo A.P.; Keerthana P.; Bhalekar D.; Tahiliani M.P.Recent Acknowledgment (RACK) is a new loss detection technique for TCP proposed by Google and described in an Internet Draft in TCP Maintenance Working Group (tcpm) of IETF. It is the default loss detection technique in Linux kernel. RACK internally uses Forward Acknowledgement (FACK) and Duplicate Selective Acknowledgement (DSACK) loss detection techniques, and leverages the benefits of Tail Loss Probe (TLP). This paper describes the implementation and evaluation of FACK, DSACK and TLP loss detection techniques for TCP model in ns-3. The goal is to provision prerequisite loss detection techniques in ns-3 for implementing RACK. The implementation of FACK, DSACK and TLP in Linux is used as a reference for this work. Our implementation of these techniques in ns-3 is verified by evaluating their performance in respective scenarios and ensuring that they exhibit their key characteristics. © 2019 IEEE.Item Leveraging named data networking for industrial automation: Opportunities and challenges(2020) Nagaraj A.H.; Tahiliani M.P.; Tandur D.; Satheesh H.Information Centric Networking (ICN) has emerged as a promising solution to address the data dissemination and scalability concerns with the traditional TCP/IP Internet architecture. Named Data Networking (NDN) is a prominent ICN architecture that supports in-network caching, easy-to-deploy multicast communication and in-built security mechanisms, thus facilitating efficient, secure and timely delivery of content to the end users. These attributes make NDN a promising solution to meet the requirements of IoT applications. Consequently, NDN based IoT deployments for health care to smart home systems, smart buildings and smart cities are on the rise. Nonetheless, the feasibility of using NDN for building next generation Industrial Automation and Control Systems (IACS) has not been studied. Industrial automation requires an efficient and scalable networking infrastructure that facilitates data sharing to drive operational improvements and develop business intelligence. Ethernet based networking solutions are becoming increasingly popular in IACS to provide seamless integration of industrial networks with the Internet. In this paper, we explore the possibilities to leverage the benefits of NDN to provide an efficient, flexible and scalable networking solution for IACS. We study the impact of making this radical change to industrial networks, highlighting the opportunities and challenges for future research and development in this direction. © 2020 IEEE.Item Group based Publisher-Subscriber Communication Primitives for ndnSIM(2020) Ananthakrishnan S.; Tahiliani M.P.; Tandur D.; Satheesh H.Named Data Networking (NDN), an information centric network architecture, aims to provide an efficient and scalable alternative for content distribution. Modern Internet applications have more emphasis on data, and are less concerned about the location from where data is retrieved. NDN's communication primitives being strictly pull based help in bringing this focus shift from location of data to the data itself. A concern for NDN is to enable the support of other popular communication primitives, such as the Publisher-Subscriber (Pub-Sub) model, seamlessly. The Pub-Sub model is well suited for applications that require periodic retrieval of data or event triggered data, such as in Industrial Automation Control Systems (IACS). This paper presents the design and development of a model to support the group based Pub-Sub communication primitives in ndnSIM, a popular ns-3 based network simulator for NDN. The functionality of the proposed model is tested by developing an end-to-end simulation environment in ndnSIM that is representative of the popular use cases of Pub-Sub communication primitives. © 2020 IEEE.Item Geometric Sequence Technique for Effective RTO Estimation in CoAP(2020) Rathod V.J.; Tahiliani M.P.Internet of Things (IoT) is a network where physical objects with Internet connectivity can interact and exchange information with other connected objects. IoT devices are constrained in terms of power and memory, and have limited communication capabilities. The Constrained Application Protocol (CoAP) is a lightweight messaging protocol which is widely used by various IoT applications in low power and lossy wireless networks. CoAP provides reliability and minimal congestion control via a fixed Retransmission TimeOut (RTO) and Binary Exponential Backoff (BEB). It does not maintain end-to-end connection information and therefore, cannot adapt RTO based on the network conditions. Moreover, CoAP resets the RTO to its default value after having received the ACK for the retransmitted packet. This approach of resetting the RTO degrades the performance in a network with high latency and leads to spurious retransmissions. In this paper, we propose a Geometric Sequence Technique (GST) for effective RTO estimation in CoAP. GST retains the previous RTO value after having received the ACK for the retransmitted packet and eventually returns to the default value by decreasing the RTO depending on the number of consecutive successful transmissions. The proposed technique is implemented in Contiki OS and validated against the existing mechanisms. The experiments have been conducted using the Cooja simulator and the FIT/IoT-LAB testbed to verify the effectiveness of the proposed technique. The results show that GST minimizes the Flow Completion Times (FCT), reduces the number retransmissions and improves the network throughput. © 2020 IEEE.Item FQ-PIE Queue Discipline in the Linux Kernel: Design, Implementation and Challenges(2019) Ramakrishnan G.; Bhasi M.; Saicharan V.; Monis L.; Patil S.D.; Tahiliani M.P.Proportional Integral controller Enhanced (PIE) is an Active Queue Management (AQM) mechanism to address the bufferbloat problem. AQM mechanisms tackle bufferbloat by dropping or marking packets before the buffers fill up, but typically do not ensure fairness between responsive and unresponsive flows that share the same bottleneck link i.e., unresponsive flows can starve responsive flows when they co-exist. Recently, there has been an active interest in integrating flow protection mechanisms with AQM mechanisms to collectively tackle the problem of bufferbloat and fairness. There exist two such algorithms: Flow Queue Controlled Delay (FQ-CoDel) and Flow Queue Proportional Integral Controller Enhanced (FQ-PIE) that integrate flow protection with AQM mechanisms. Flow protection is achieved by dividing the incoming flows into separate queues and then applying CoDel/PIE algorithm on respective queues. Although FQ-CoDel is available in the mainline of Linux, there does not exist a model for FQ-PIE. In this paper, we discuss the design and implementation of FQ-PIE in the Linux kernel. We test and evaluate our proposed model of FQ-PIE in different scenarios by comparing the results obtained from it to those obtained for PIE and FQ-CoDel. Besides evaluating the fairness among responsive and unresponsive flows, we also evaluate the fairness among different types of responsive flows, such as when CUBIC TCP shares the same bottleneck link as TCP BBR. We also assess the benefits of integrating flow protection with AQM mechanisms in terms of reducing the latency for thin, latency sensitive flows when they coexist with thick, latency tolerant flows. © 2019 IEEE.