MASCOTS 2020

Please click here to download the program in pdf file

DAY 1

9:00 Welcome

9:10 Opening Keynote

Gabriele Kotsis, Johannes Kepler University , Linz (Austria) Intelligence? Smartness? Emotion? What do we expect from future computing machinery? (certificate)

Bio: Dr. Gabriele Kotsis is a Professor at the Johannes Kepler University (Linz, Austria) and she is the ACM President since July 2020. During her career, she has been appointed to many prestigious positions, including among the others, the Vice-Rector for research at JKU University and the President of the Austrian Computer Society. Professor Kotsis holds a PhD from the University of Vienna. Her research interests include performance evaluation and workload characterization applied to many different application domains.

10:20 Traffic and network modelling (1h)

• Shaojun Zhang, Chen Wang and Albert Zomaya Adversarial attacks in a deep reinforcement learning based cluster scheduler (best paper award)
  
• Huibo Bi, Erol Gelenbe and Yanyan Chen Incentive mechanism for collective coordination in an urban intelligent transportation system using G-networks
• Beom-Su Kim and Ki-Il Kim A Priority-based Dynamic Link Scheduling Algorithm Using Multi-criteria Decision Making in Wireless Body Area Networks

• Harsha Sharma, Wenfei Wu and Bangwen Deng Symbolic Execution for Network Functions withTime-Driven Logic
• Christian Helm, Soramichi Akiyama and Kenjiro Taura Reliable Reverse Engineering of Intel DRAM Addressing Using Performance Counters
• June-Hyung Kim, Youngjae Kim, Safdar Jamil and Sungyong Park A NUMA-aware NVM File System Design for Manycore Server Application (short)

12:15 Lunch

13:00 Performance Optimization (1h 20min)

• Fuad Aleskerov, Anna Rezyapova, Alina Roman and Vyacheslav Yakuba New Centrality Measures in Networks and their Applications to the International Trade and Migration Networks
• Erol Gelenbe, Pawel Boryszko, Miltiadis Siavvas and Joanna Domanska  Optimum Checkpoints for Time and Energy
• Pedro Mendes, Maria Casimiro, Paolo Romano and David Garlan  TrimTuner: Efficient Optimization of Machine Learning Jobs in the Cloud via Sub-Sampling
• Anil Kirmaz, Diomidis S. Michalopoulos, Irina Balan and Wolfgang Gerstacker Mobile Network Traffic Forecasting Using Artificial Neural Networks

14:20 Session Data driven systems (1h 20min)

• Renato Cunha and Luiz Chaimowicz Towards a common environment for learning scheduling algorithms
• Malek Musleh, Roberto Penaranda, Allister Alemania,
  Pedro Yebenes Segura, Gene Wu, Jan Zielinkski, Krzysztof Raszkowski, Nan
  Ni, Scott Diesing, Anupama Kurpad, Ram Huggahalli, Curt Bruns, Steven
  Miller and Sujoy Sen Fabsim-X: A simulation Framework for the
  Analysis of Large-Scale Topologies and Congestion Control Protocols in
  Data Center Networks
• Vaibhav Saxena, K. R. Jayaram, Saurav Basu, Ashish Verma and Yogish Sabharwal Effective Elastic Scaling of Deep Learning Workloads
• Jehan-Francois Paris and Thomas Schwarz Merkle Hash Grids instead of Merkle Trees

15:40 Coffee Break

16:00 Hardware/storage (1h 20min)

• Mohammad Hossein Hajkazemi, Mania Abdi and Peter Desnoyers μCache: a mutable cache for SMR translation layer (best paper award)
  
• Onkar Patil, Frank Mueller, Latchesar Ionkov, Jason Lee and Michael Lang Symbiotic HW Cache and SW DTLB Prefetching for DRAM/NVM Hybrid Memory
• Roman Pletka, Nikolaos Papandreou, Radu Stoica, Haris Pozidis, Nikolas Ioannou, Tim Fisher, Aaron Fry, Kip Ingram and Andrew Walls Improving NAND flash performance with read heat separation
• Maher Kachmar and David Kaeli  A Smart Background Scheduler for Storage Systems

Day 2

9:00 Performance evaluation (1h 40min)

• Tatsuhiro Chiba and Takeshi Yoshimura Investigating Genome Analysis Pipeline Performance on GATK with Cloud Object Storage
• Fabiana Rossi, Valeria Cardellini and Francesco Lo Presti Self-adaptive Threshold-based Policy for Microservices Elasticity
• Alim Ul Gias and Giuliano Casale COCOA: Cold Start Aware Capacity Planning for Function-as-a-Service Platforms
• Thomas Prantl, Peter Ten, Lukas Iffländer, Alexandra Dmitrenko, Samuel Kounev and Chrisitian Krupitzer Evaluating the Performance of a State-of-the-Art Group-oriented Encryption Scheme for Dynamic Groups in an IoT Scenario
• Andrea Gulino, Arif Canakoglu, Stefano Ceri and Danilo Ardagna Performance Prediction for Data-driven Workflows on Apache Spark

10:40 Coffee Break

11:00 Keynote

Ernesto Damiani, University of Milano (Italy) and Khalifa University (United Arab Emirates) Mobile network data for risk assessment in the time of COVID: the PLACES approach (certificate)

Abstract: The talk discusses the approach used in the PLACES project, relying on mobile network data such as CDRs to derive spatio-temporal probabilistic models. PLACES allows to study the dynamics of contagious diseases, providing a sound basis to (i) predicting the evolution of collective phenomena and (ii) assessing collective risk. The talk also discusses potential coupling of PLACES models to the encounter-based systems implemented as apps that use device-to-device channels to exchange opaque codes.

Bio: Dr. Ernesto Damiani is the Senior Director of Robotics and Intelligent Systems Institute at Khalifa University, Abu Dhabi, UAE. He is also a Director of the Khalifa University Center for Cyber Physical Systems (C2PS). Dr. Damiani is the Chair of the Information Security Program and a Research Professor in EBTIC. He leads the SESAR research lab at the Department of Computer Science, Università degli Studi di Milano, Italy. He is also the President of the Italian Consortium of Computer Science Universities (CINI). Ernesto’s research interests include secure service-oriented architectures, privacy-preserving Big Data analytics and Cyber-Physical Systems security. 

12:00 Lunch

13:00: Cloud/fog (1h 35min)

• Ali Fahs, Guillaume Pierre and Erik Elmroth Voilà: Tail-Latency-Aware Fog Application Replicas Autoscaler        
• Johannes Grohmann, Daniel Seybold, Simon Eismann, Mark Leznik, Samuel Kounev and Jörg Domaschka Baloo: Measuring and Modeling the Performance Configurations of Distributed DBMS              
• Mulugeta Ayalew Tamiru, Guillaume Pierre, Johan Tordsson and Erik Elmroth Instability in Geo-Distributed Kubernetes Federation: Causes and Mitigation
• William Viktorsson, Cristian Klein and Johan Tordsson    Security-Performance Trade-offs of Kubernetes Container Runtimes (Short)
• Moiz Arif, M. Mustafa Rafique, Seung-Hwan Lim and Zaki Malik Infrastructure-Aware TensorFlow for Heterogeneous Datacenters

14:35 Coffee Break

15:00 Performance modeling (1h 55min)

• Ayat Zaki-Hindi, Salah-Eddine Elayoubi and Tijani Chahed Model-Aided Learning for URLLC Transmission in Unlicensed Spectrum
• Emilio Incerto, Annalisa Napolitano and Mirco Tribastone Statistical Learning of Markov Chains of Programs
• Ioannis Koukoutsidis Age of Information in an Overtake-Free Network of Quasi-Reversible Queues
• Youssef Ait El Mahjoub, Hind Castel and Jean-Michel Fourneau Energy Packet Networks with general service time distribution
• Marek Molęda, Alina Momot and Dariusz Mrozek Concept Drift and Avoiding its Negative Effects in Predictive Modeling of Failures of Electricity Production Units in Power Plants
• Eitan Bachmat and Josu Doncel Non-Asymptotic Performance Analysis of Size-Based Routing Policies (Short)