Papers by Chung-horng Lung
Science of Computer Programming, Aug 1, 2010
Selecting the architecture that meets the requirements, both functional and nonfunctional, is a c... more Selecting the architecture that meets the requirements, both functional and nonfunctional, is a challenging task, especially at the early stage when more uncertainties exist. Architectural prototyping is a useful approach in supporting the evaluation of alternative architectures and balancing different architectural qualities. Generative programming has gained increasing attention, but it mostly deals with lower-level artifacts; hence, it usually supports lower degrees of software automation. This paper proposes an architecturecentric generative approach in facilitating architectural prototyping and evaluation. We also present our empirical experience in raising the level of abstraction to the architecture layer for distributed and concurrent systems using GenVoca. GenVoca is a generative programming approach that is used here to support the generation or instantiation of a particular architectural pattern in distributed computing based on user's selection. As a result, it can support rapid architectural prototyping and evaluation of both functional and non-functional requirements and encourage greater degrees of software automation and reuse. Lessons learned from the empirical study are also reported and could be applied to other areas.
arXiv (Cornell University), Apr 4, 2020
Google's QUIC (GQUIC) is an emerging transport protocol designed to reduce HTTP latency. Deployed... more Google's QUIC (GQUIC) is an emerging transport protocol designed to reduce HTTP latency. Deployed across its platforms and positioned as an alternative to TCP+TLS, GQUIC is feature rich: offering reliable data transmission and secure communication. It addresses TCP+TLS's (i) Head of Line Blocking (HoLB), (ii) excessive round-trip times on connection establishment, and (iii) entrenchment. Efforts by the IETF are in progress to standardize the next generation of HTTP's (HTTP/3, or H3) delivery, with their own variant of QUIC. While performance benchmarks have been conducted between GQUIC and HTTP/2-over-TCP (H2), few analyses, to our knowledge, have taken place between H2 and H3. In addition, past studies rely on Page Load Time as their main, if not only, metric. The purpose of this article is to benchmark the latest draft specification of H3 and dig into a user's Quality of Experience (QoE) by using Lighthouse: an open source (and metric diverse) auditing tool. Our findings show that, for one of H3's early implementations, H3 is mostly worse but achieves a higher average throughput.
In-network Caching for ICN-based IoT (ICN-IoT): A Comprehensive Survey
IEEE Internet of Things Journal, 2023
Position-Error Based Schemes for Bilateral Teleoperation with Time Delay: Theory and Experiments
Abstract The problem of stable bilateral teleoperation with position error based force feedback i... more Abstract The problem of stable bilateral teleoperation with position error based force feedback in presence of time-varying possibly unbounded communication delay is addressed. Two stabilization schemes are proposed that guarantee" independent of delay ...
Lecture Notes in Computer Science, 2010
Real-time Radio Control (RC) systems require instantaneous response in the controlled device. RC ... more Real-time Radio Control (RC) systems require instantaneous response in the controlled device. RC systems have wide applications, including ad hoc networks. Imperfections in the wireless channel (noise and interference) result in randomly fluctuating latency in the response of the system. A lag occurs when the system latency exceeds a specified real-time threshold. System Lag Occurrence Probability (SLOP) is the probability of lag occurrence and is derived as the performance metric to characterize user experience in real-time radio control systems. Frequency hopping is used to mitigate interference effects. Uniform serial acquisition and N-state lock detection are used to simplify the derivation of SLOP. Simulation results are presented to verify the derivation of SLOP.
Segment Routing Green Spine Switch Management Systems for Data Center Networks
The datacenter is the core of most IT industries today. As a result, due to the exponential incre... more The datacenter is the core of most IT industries today. As a result, due to the exponential increase of information and services demanded from datacenters, it is projected that 14.1 zeta bytes of bandwidth would be needed to meet the current demands of data by the end of 2020 in USA datacenters alone. This would require a tremendous amount of energy to run those datacenters. Hence, in recent years there is more focus on reducing the energy consumption in a datacenter. This paper approaches this problem by utilizing Segment Routing based Green Spine Switch Management System (SR-GSSMS) to have an efficient bandwidth usage. The proposed approach makes it possible to deactivate some low utilized spine switches and links on the datacenter networks, which results in energy savings. The system also maintains the network performance when congestion or a link failure occurs. Our experimental results yielded up to a 78% energy savings on the spine-links, while maintaining the same traffic requirements.
Load Balancing Using ECMP in Multi-Stage Clos Topology in a Datacenter
Datacenters most widely use 3-stage Clos topologies such as fat-tree topology and leaf-spine topo... more Datacenters most widely use 3-stage Clos topologies such as fat-tree topology and leaf-spine topology. These network topologies may face challenges at scale in future as the datacenters are growing at a large scale. Therefore, 5-stage Clos topology is explored in this paper with its benefits. The paper focuses on load balancing of elephant flows using Equal-Cost Multi-Path (ECMP) algorithm. This leads to improvement in performance of 5-stage Clos topology which leads to decrease in number of failures that are incurred because of unavailable bandwidth. This performance is compared to 3-stage Clos in terms of time taken by a request for completion and the number of failures appeared in the network during the simulation experiments. The experimental results show that the 5-stage Clos topology performs better in terms of network reliability due to less number of packet losses and it also shows comparable results to 3-stage Clos for the processing time taken by the requests inside the datacenter network.
A Comparative Evaluation of Computation Offloading for Mobile Applications
ABSTRACT With the increasing availability of mobile applications, the usage of mobile devices, es... more ABSTRACT With the increasing availability of mobile applications, the usage of mobile devices, especially smart phones and tablets, has become popular nowadays. However, mobile devices are limited in battery, memory, storage, and processing capabilities. These constraints prevent mobile devices from widely running all kinds of rich mobile applications. Computation offloading is believed to be a potential solution to the hardware limitations of mobile devices for higher performance and/or energy savings. Computation offloading is often conducted on a remote server, typically in the cloud as reported in the literature. In this paper, we consider that it is not always necessary to offload all computations to a remote cloud server from a mobile device mainly due to the high communication delay that may be generated. Instead, the offloading target can be a nearby PC via a lower communication cost mechanism, such as Bluetooth, depending on the computation demands. We have examined the impact of various factors, such as computation workload, file size, and wireless communication protocols, and have investigated insights on power consumption through offloading to a nearby PC or a remote cloud server. Finally, we develop an adaptive algorithm based on experimental results to automatically select a computation resource which has capacity to execute computationally intensive applications and thus to save energy consumption of mobile devices.
Computer Communications, Jul 1, 2016
XML-based publish/subscribe (pub/sub) systems have been receiving a great deal of attention from ... more XML-based publish/subscribe (pub/sub) systems have been receiving a great deal of attention from the academic community and the industry. This research focuses on efficient pub/sub systems and considers the devising of XML-based pub/sub systems from the perspective of subscription/query and publication message delivery. Existing research mainly focuses on the efficiency of XML publication message filtering algorithms. Not much research, however, has considered using the system or the communication model in the context of XML publication messages delivery. This paper presents innovative XML delivery techniques, the cross-layer model and the peer model; both techniques make use of publisher and customer edge brokers for efficient XML subscription aggregation and publication message delivery. The primary contribution of the proposed models is the reduction of the number of XML publication message filtering and XPath query aggregation operations performed in the conventional filter-based XML multicast model, which has a high computational overhead. The main idea is to store user subscriptions at customer and publisher edge brokers which are either directly connected or close to the subscribers and the publisher, respectively. We have performed a number of experiments within a controlled local area network (LAN) environment for demonstration of the basic concepts underlying the techniques and in the Amazon cloud environment that emulates the wide area network (WAN). Both the cross-layer and the peer models can reduce the end-to-end (E2E) delay in message delivery. For example, the results obtained from experiments in a LAN demonstrate several-fold performance improvement in E2E delay for both the cross-layer and the peer models compared to the conventional filter-based XML multicast model, and the results using the cloud show an improvement as high as 64% in E2E delay for the peer model over the multicast model.
IFAC Proceedings Volumes, 2011
A new general stability result for networked bilateral teleoperator systems with projection-based... more A new general stability result for networked bilateral teleoperator systems with projection-based force reflection is presented. Contrary to the previous results, the human operator dynamics are taken into account in the stability analysis. The human dynamics are assumed to satisfy a general condition formulated in terms of finite distance between output of the human dynamics and a certain "passivity-based" stabilizing set. In particular, this assumption allows for both passive and non-passive behaviour of the human operator. The overall stability of the bilateral teleoperator system is proven in the presence of network-induced communication constraints for an arbitrary projection-based force reflection algorithm from a broad class.
Tracking Per-Flow State – Binned Duration Flow Tracking
Journal of Networks, 2012
Abstract Recent advances in network monitoring have increasingly focused on obtaining per-flow ... more Abstract Recent advances in network monitoring have increasingly focused on obtaining per-flow information, such as flow state. Tracking the state of network flows opens up a new dimension of information gathering for network operators, allowing previously unattainable data to ...
QoS routing solutions can be classified into two categories, state-dependent and time-dependent, ... more QoS routing solutions can be classified into two categories, state-dependent and time-dependent, according to their awareness of the future traffic demand in the network. Compared with representative state-dependent routing algorithms, a time-dependent variation of WSP-TDWSP-is proposed in this paper to study the role of traffic forecasting in QoS routing, by customizing itself for a range of traffic demands. Our simulation results confirm the feasibility of traffic forecasting in the context of QoS routing, which empowers TDWSP to achieve better routing performance and to overcome QoS routing difficulties, even though completely accurate traffic prediction is not required. The case study involving TDWSP further reveals that even a static forecast can remain effective over a large area in the solvable traffic demand space, if the network topology and the peak traffic value are given. Thus, the role of traffic forecasting in QoS routing becomes more prominent.
Performance optimization of big data in mobile networks
Smart phones are ubiquitous nowadays and have taken over the bulk data transfers in mobile networ... more Smart phones are ubiquitous nowadays and have taken over the bulk data transfers in mobile networks. The next generation phones are even more powerful to handle voice, video and data catering real time multimedia experience. Unfortunately, the increase in service provider capacity has not kept up with the user demand for more bandwidth. It is becoming very expensive for service providers to cater higher bandwidth without investing on new technology or expansion. In this paper we propose a bandwidth optimization algorithm based on cache coherency where the user data transfer is optimized without compromising the user expectation or the need for service providers to expand their capacity. The proposed algorithm is compared with existing data transfer techniques (such as GIT [9]) and we show the efficiency of the proposed method through simulation. We validate the simulation results through two year measured result.
Proceedings of the ... American Control Conference, Jul 1, 2007
This thesis is concerned with the analysis of the control design to the nonlinear networked contr... more This thesis is concerned with the analysis of the control design to the nonlinear networked control systems (NCSs). Ignoring the network connection and cascading actuators, the plant and sensors together, a sampled-data system is obtained. The stabilization problem of nonlinear sampled-data systems is considered under the low measurement rate constraint. Dual-rate control schemes based on the emulation design and discrete-time design approaches respectively are proposed that utilize a numerical integration model to approximately predict the current state of the plant. It is shown that using the dual-rate control schemes, input-to-state stability property will be preserved for the closed loop sampled-data system in a practical sense. On the other hand, the networked realization of nonlinear control systems is studied and a model-based control scheme is addressed as a solution to reduce the network traffic and resultantly, to attain a higher performance. The NCSs are modeled as continuous-time systems and sampled-data systems, respectively. Under the proposed control scheme, a tradeoff between satisfactory control performance and reduction of network traffic can be achieved. It is shown that by using the estimated values, generated by the plant model, instead of true values of the plant, a significant saving in the required bandwidth is achieved and this makes possible stabilization of the plant even under slow network conditions. ACKNOWLEDGEMENT I am grateful to all the people who made this thesis possible. First and foremost, I would like to express my deep and sincere gratitude to my supervisors, Prof. Yanping Lin and Prof. Haracio J. Marquez. During my doctoral study, they never failed to be friendly and supportive and always had time to help and discuss problems. Their enthusiasm and inspiration helped to make research life smooth and rewarding for me. Without their patience and generous support, as well as the critical comments during the thesis writing period, this thesis would not have come as far as it did. I thank my mathematics teachers in China and Canada and departmental support staff who have assisted me in many different ways. Special thanks to Prof. Tongwen Chen for one course which increased my knowledge in the area of sampled-data control. I am also very grateful to my doctoral committee members for the valuable comments and recommendations eliminated many ambiguous points from the thesis.
Current methods ,for software architecture analysis often ,fall short qf providing objective and ... more Current methods ,for software architecture analysis often ,fall short qf providing objective and quantitative performance information. The paper describes how to bring together techniques in software performance engineering and software architecture analysis in order to supportper-,formance-oriented software architecture engineering. The paper presents a systematic approach derived from empirical case studies in real-time telecommunications applications. The approach has been successfully applied to these case studies to help product teams analyze and improve the performance and other quality factors qf their systems.
Network capacity region of multi-queue multi-server queueing system with random ON-OFF connectivi... more Network capacity region of multi-queue multi-server queueing system with random ON-OFF connectivities and stationary arrival processes is derived in this paper. Specifically, the necessary and sufficient conditions for the stability of the system are derived under general arrival processes with finite first and second moments. In the case of stationary arrival processes, these conditions establish the network capacity region of the system. It is also shown that AS/LCQ (Any Server/Longest Connected Queue) policy stabilizes the system when it is stabilizable. Furthermore, an upper bound for the average queue occupancy is derived for this policy.
A Green Computing Based Architecture Comparison and Analysis
With rising energy cost and growing environmental concerns, green computing is receiving more and... more With rising energy cost and growing environmental concerns, green computing is receiving more and more attention. Software and system architectures ( in terms of concurrency patterns) play a crucial role in both computing and telecommunication systems, and they have been analyzed for performance, reliability, maintainability, and security. Yet, little work on analysis based on the amount of energy that the CPU/processor will consume has been reported. Since most communication systems have to run 24/7 (e.g., most server farms, servers in a cloud computing infrastructure), the energy consumption of a system based on a specific software architecture is of great importance. For example, high energy consuming always leads to higher operational cost of the system. High energy consumption also implies more heat produced, thus, more power is required for cooling-down. In this paper, we propose a model that can be used to analyze the energy consumption of a software architecture, given the CPU/processor that will be used. Thus, this model gives one an additional dimension to compare competing software architectures. We present a case study using a generic sender/receiver communication system to evaluate two architectural alternatives with emphasis on energy consumption.
Smart Proactive Caching: Empower the Video Delivery for Autonomous Vehicles in ICN-Based Networks
IEEE Transactions on Vehicular Technology, Jul 1, 2020
The recent advances in vehicular communications and networking are bringing self-driving vehicles... more The recent advances in vehicular communications and networking are bringing self-driving vehicles closer to reality. Once full automation (i.e., levels 4 and 5 as defined in the five levels of autonomous driving) becomes a reality, entertainment services for both drivers and passengers will shift from listening to radios (or music) to watching videos. Hence, how to improve the quality of experience (QoE) for autonomous vehicle (AV) users, and to reduce the network load will become a crucial problem. Information-centric networking (ICN) is seen as one of the potential paradigms for next-generation networks, and could potentially be used for content distribution in vehicular networks. Caching, an important feature in ICN-based networks, is an efficient way to reduce network load, and to improve QoE for users. However, traditional reactive caching approaches are inefficient for AV users due to the high delay caused by their high speed. In this paper, we propose a novel hierarchical proactive caching approach that considers both users’ future demands and AV user mobility. The proposed approach uses the non-negative matrix factorization (NMF) technique to predict user's preferences which are then used to predict users’ future demands by considering the historical popularity of videos. A user mobility prediction model is used to predict the AV users’ next location based on the current location and the planned route information which can be retrieved from the self-driving system. Based on the predicted users’ future demands and locations of AVs, the proposed caching approach can proactively cache videos that are likely to get requested at the next base station (BS) or roadside unit (RSU) that the users are moving to. The proposed approach has been evaluated under two scenarios: a highway scenario and a grid street scenario. Results show that the proposed approach can significantly improve the efficiency of caching in terms of hit ratio and the average number of hops.
DOAJ (DOAJ: Directory of Open Access Journals), Aug 1, 2004
Reengineering is to reorganize and modify existing systems to enhance them or to make them more m... more Reengineering is to reorganize and modify existing systems to enhance them or to make them more maintainable. Reengineering is usually necessary as systems evolve due to changes in requirements, technologies, and/or personnel. Design patterns capture recurring structures and dynamics among software participants to facilitate reuse of successful designs. Design patterns are common and well studied in network systems. In this project, we reengineer part of a network system with some design patterns to support future evolution and performance improvement. We start with reverse engineering effort to understand the system and recover its high level architecture. Then we apply concurrent and networked design patterns to restructure the main subsystem. Those patterns include Half-Sync/Half-Async, Monitor Object, and Scoped Locking idiom. The resulting system is more maintainable and has better performance.
Uploads
Papers by Chung-horng Lung