Département Informatique

Computer Science Department

New article “Real-Time Tracking and Mining of Users’ Actions over Social Media ” at Computer Science and Information Systems

Ejub Kajan, Noura Faci, Zakaria Maamar, Mohamed Sellami, Emir Ugljanin, et al.. Real-time tracking and mining of users’ actions over social media. Computer Science and Information Systems, ComSIS Consortium, In press, pp.2-2. ⟨10.2298/CSIS190822002K⟩⟨hal-02514060⟩

Abstract. With the advent of Web 2.0 technologies and social media, companiesare actively looking for ways to know and understand what users think and say about their products and services. Indeed, it has become the practice that users go online using social media like Facebook to raise concerns, make comments, and share recommendations. All these actions can be tracked in real-time and then mined using advanced techniques like data analytics and sentiment analysis. This paper discusses such tracking and mining through a system called Social Miner that allows companies to make decisions about what, when, and how to respond to users’ actions over social media. Questions that Social Miner allows to answer include what actions were frequently executed and why certain actions were executed more than others.

New paper “Towards an Approach for Validating the Internet-of-Transactional-Things” at AINA’2020

Z. Maamar, M. Sellami, N.C. Narendra, I. Guidara, E. Ugljanin, and B. Banihashemi. Towards an Approach for Validating the Internet-of-Transactional-Things. In the 34-th International Conference on Advanced Information Networking and Applications (AINA-2020)

 

Abstract. This paper examines the impact of transactional properties, known as pivot, retriable, and compensatable, on Internet-of-Things (IoT). Despite the ever-growing number of things in today’s cyber-physical world, a limited number of studies examine this impact while considering things’ particularities in terms of reduced size, restricted connectivity, continuous mobility, limited energy, and constrained storage. To address this gap, this paper proceeds first, with exposing things’ duties, namely sensing, actuating, and communicating. Then, it examines the appropriateness of each transactional property for each duty. During the performance of transactional things, (semi)-atomicity criterion is adopted allowing to approve when these things’ duties could be either canceled or compensated. A system that runs a set of what-if experiments is presented in the paper allowing to demonstrate the technical doability of transactional things.

New paper “A Model based Toolchain for the Cosimulation of Cyber-physical Systems with FMI” at MODELSWARD’2020

A Model based Toolchain for the Cosimulation of Cyber-physical Systems with FMI by D. Oudart, J. Cantenot, F. Boulanger and S. Chabridon

Abstract Smart Grids are cyber-physical systems that interface power grids with information and communication technologies in order to monitor them, automate decision making and balance production and consumption. Cosimulation with the Functional Mock-up Interface standard allows the exploration of the behavior of such complex systems by coordinating simulation units that correspond to the grid part, the communication network and the information system. However, FMI has limitations when it comes to cyber-physical system simulation, particularly because discrete-event signals exchanged by cyber components are not well supported. In addition, industrial projects involve several teams with different skills and methods that work in parallel to produce all the models required by the simulation, which increases the risk of inconsistency between models. This article presents a way to exchange discrete-event signals between FMI artifacts, which complies with the current 2.0 version of the standard. We developed a DSL and a model-based toolchain to generate the artifacts that are necessary to run the cosimulation of the whole system, and to
detect potential inconsistencies between models. The approach is illustrated by the use case of an islanded grid implementing diesel and renewable sources, battery storage and intelligent control of the production.

New paper “State-Machine Replication for Planet-Scale Systems” to be presented at Eurosys’20

New paper “State-Machine Replication for Planet-Scale Systems” to be presented at Eurosys’20. Congrat to Pierre and Tuanir!

New paper “Using differential execution analysis to identify thread interference”. To appear in IEEE Transactions on Parallel and Distributed Systems

Abstract Understanding the performance of a multi-threaded application is difficult. The threads interfere when they access the same shared resource, which slows down their execution. Unfortunately, current profiling tools report the hardware components or the synchronization primitives that saturate, but they cannot tell if the saturation is the cause of a performance bottleneck. In this paper, we propose a holistic metric able to pinpoint the blocks of code that suffer interference the most, regardless of the interference cause. Our metric uses performance variation as a universal indicator of interference problems. With an evaluation of 27 applications we show that our metric can identify interference problems caused by 6 different kinds of interference in 9 applications. We are able to easily remove 7 of the bottlenecks, which leads to a performance improvement of up to 9 times

https://hal.archives-ouvertes.fr/hal-02179717v1

New paper “ScalOMP: analyzing the Scalability of OpenMP applications” to be presented at IWOMP’19

Anton Daumen will present his work “ScalOMP: analyzing the Scalability of OpenMP applications” at IWOMP’19. 

His paper is available online: https://hal.archives-ouvertes.fr/hal-02179726

Abstract : Achieving good scalability from parallel codes is becoming increasingly difficult due to the hardware becoming more and more complex. Performance tools help developers but their use is sometimes complicated and very iterative. In this paper we propose a simple methodology for assessing the scalability and for detecting performance problems in an OpenMP application. This methodology is implemented in a performance analysis tool named ScalOMP that relies on the capabilities of OMPT for analyzing OpenMP applications. ScalOMP reports the code regions with scalability issues and suggests optimization strategies for those issues. The evaluation shows that ScalOMP incurs low overhead and that its suggestions lead to significant performance improvement of several OpenMP applications