*LESS’23

Held in person on March 13, 2023, in Atlanta, GA (USA).

LinkedIn report

Technical Program Committee

  • Cristina L. Abad, Escuela Superior Politecnica del Litoral, Ecuador
  • Luciano Baresi, Politecnico di Milano, Italy
  • Christian Becker, Universität Mannheim, Germany
  • Priscilla Benedetti, University of Perugia, Italy
  • Giuseppe Coviello, NEC Laboratories America, USA
  • Ginés Garcia-Aviles, i2CAT, Spain
  • Marco Gramaglia, University Carlos III of Madrid, Spain
  • Shashikant Illgar, Vienna University of Technology, Austria
  • Bahman Javadi, Western Sydney University, Australia
  • Slawomir Kuklinski, Orange, Poland
  • Ignacio Labrador Pavon, Atos Research and Innovation, Spain
  • Eduard Marin, Telefonica Research, Spain
  • Enzo Mingozzi, University of Pisa, Italy
  • Peter Pietzuch, Imperial College London, UK
  • Jesús Pérez-Valero, University Carlos III of Madrid, Spain
  • Maria Rodriguez Read, The University of Melbourne, Australia
  • Marco Savi, University of Milano-Bicocca, Italy
  • Peter Sbarski, A Cloud Guru, Australia
  • Davide Taibi, Tampere University, Finland
  • Antonio Virdis, University of Pisa, Italy

Technical Program

  • Keynote: The Elastic AI Ecosystem – Towards A Holistic Pervasive System for Adaptive Artificial Intelligence, Gregor Schiele (University of Duisburg-Essen, Germany)
  • Meet-the-author with Gabriele Russo Russo (Università Roma Tre, Italy)
  • Measuring the Edge: A Performance Evaluation of Edge Offloading, Heiko Bornholdt and Kevin Röbert (Universität HambGermany); Martin Breitbach (University of Mannheim, Germany); Mathias Fischer and Janick Edinger (Universität Hamburg, Germany)
  • OS3: The Art and the Practice of Searching for Open-Source Serverless Functions, Sarvesh Bhatnagar and Zhengquan Li (University of MichiDearborn, USA); Zheng Song (University of Michigan at Dearborn, USA); Eli Tilevich (Virginia Tech, USA)
  • Reducing the Cost of GPU Cold Starts in Serverless Deep Learning Inference Serving, Justin San Juan and Bernard Wong (University of WaterCanada)

Keynote

Title: The Elastic AI Ecosystem - Towards A Holistic Pervasive System for Adaptive Artificial Intelligence

Speaker: Gregor Schiele

Abstract: The structure of pervasive application systems that use artificial intelligence (AI) is getting more complicated. On the one hand, AI is moving closer to its data sources, often sensors embedded in the environment or worn by users. Originally, sensors would stream their data to remote Cloud centers, where it would be processed and stored. In recent years, Edge systems have emerged, moving the AI-based processing of sensor data into the local environment of a pervasive system. Now, data is processed directly on the embedded sensor devices, allowing for tight control of data privacy and latency but at the same time forcing the AI to work with minimal compute power, memory, and energy. On the other hand, not everything can be done locally, either because the necessary AI models are too resource hungry or because they integrate data from many sources and locations. Therefore, it is often necessary and beneficial to connect local AI with Edge and Cloud AI and to let the system determine dynamically which part of the overall system should be executed where. To address this, we are working on a holistic system for pervasive AI. We call it the Elastic AI ecosystem. It consists of (i) a hardware platform for exchangeable AI accelerators, (ii) an accelerator generation tool, and (iii) a distributed runtime system. AI-based applications can be composed out of differ- ent building blocks such as sensor data sources, preprocessing algorithms, and neural networks. Building blocks are executed locally (i.e., embedded), on nearby Edge or remote Cloud servers and compositions can be adapted at runtime, e.g., to switch from a Cloud AI to a local AI in case of networking problems. In this keynote I present the current state of the Elastic AI ecosystem, how we use it in different application scenarios, and what challenges we face in the future.

Speaker bio: Gregor Schiele is a full Professor for Computer Science in the Department of Computer Science and Applied Cognitive Science at the University of Duisburg-Essen (UDE) in Germany. He leads the Embedded Systems Lab at UDE and is the Chairman of the examination board of the master program for Cyber Physical Systems. His current research interests are self- organizing pervasive systems, embedded machine learning, and customized hardware accelerators. He received his Ph.D. in 2007 from the University of Stuttgart, Germany and is a long time PerCom regular, having served in different roles over the years. Gregor is an on-off member of the ACM, the IEEE and the German Informatics Society.

*LESS’22

Event held in a fully virtual manner on March 21st, 2022. Times below in CET.

Technical Program Committee

  • Luciano Baresi, Politecnico di Milano, Italy
  • Christian Becker, Universität Mannheim, Germany
  • Marco Fiore, IMDEA Networks Institute, Spain
  • Xenofon Foukas, Microsoft, United Kingdom
  • Ginés Garcia-Aviles, i2CAT, Spain
  • Andres Garcia-Saavedra, NEC Laboratories Europe GmbH, Germany
  • Marco Gramaglia, University Carlos III of Madrid, Spain
  • Shashikant Illgar, Vienna University of Technology, Austria
  • Bahman Javadi - Western Sydney University, Australia
  • Eduard Marin, Telefonica Research, Spain
  • Redowan Mahmud, RMIT University, Australia
  • Enzo Mingozzi, University of Pisa, Italy
  • Peter Pietzuch, Imperial College London, UK
  • Maria Rodriguez Read, The University of Melbourne, Australia
  • Peter Sbarski, A Cloud Guru, Australia
  • Davide Taibi, Tampere University, Finland

Technical Program

  • FedTour: Participatory Federated Learning of Tourism Object Recognition Models with Minimal Parameter Exchanges between User Devices, Shusaku Tomita and Jose Talusan (Nara Institute of Science and Technology, Japan); Yugo Nakamura (Kyushu University, Japan); Hirohiko Suwa and Keiichi Yasumoto (Nara Institute of Science and Technology, Japan)
  • Reinforcement Learning Applicability for Resource-Based Auto-scaling in Serverless Edge Applications, Priscilla Benedetti (University of Perugia, Italy & Vrije Universiteit Brussels, Belgium); Gianluca Reali and Mauro Femminella (University of Perugia, Italy); Kris Steenhaut (Vrije Universiteit Brussel, Belgium)
  • Keynote: Designing Serverless Platforms to Support Emerging Applications]_, Cristina L. Abad (Escuela Superior Politécnica del Litoral Guayaquil, Ecuador)
  • Serverless Computing for NFV: Is it Worth it? A Performance Comparison Analysis (teaser video), Marco Savi and Alessandro Banfi (University of Milano-Bicocca, Italy); Alessandro Tundo (University of Milano - Bicocca, Italy); Michele Ciavotta (University of Milano-Bicocca, Italy)
  • Testing Approaches And Tools For AWS Lambda Serverless-Based Applications, Eetu Rinta-Jaskari, Christopher Allen, Tamara Meghla and Davide Taibi (Tampere University, Finland)
  • Towards QoS-Aware Function Composition Scheduling in Apache OpenWhisk (teaser video), Gabriele Russo Russo and Alfredo Milani (University of Rome Tor Vergata, Italy); Stefano Iannucci (Mississippi State University, USA); Valeria Cardellini (University of Rome Tor Vergata, Italy)
  • DataXe: A system for application self-optimization in serverless edge computing environments, Giuseppe Coviello (NEC Laboratories America, Inc., USA); Kunal Rao (NEC Laboratories America Inc., USA); Biplob Debnath (NEC Labs America, USA); Oliver Po (NEC Laboratories America Inc., USA); Srimat Chakradhar (NEC Research Labs, USA) |

    Keynote

Title: Designing Serverless Platforms to Support Emerging Applications

Speaker: Cristina L. Abad

Abstract: Serverless computing offerings from cloud providers have gained significant traction in recent years due to the advantages that these platforms bring with their flexible pricing models, built-in scalability, and minimal operational requirements. In a recent survey of serverless use cases, we found a wide variety of applications that depend on these services, including implementing the core functionality at the backend of mobile applications, automating the DevOps tasks of complex distributed applications, real-time processing of IoT streaming data, and scientific applications. To properly support these applications, the platforms should be fast, self-managing, and provide support for diverse QoS requirements. As a result, novel improvements to serverless platforms are rapidly being proposed and adopted. Evaluating these solutions necessitates application-based, workload-aware benchmarking tools that the community can rely on. This talk addresses these challenges and our research efforts on tackling them, presenting a performance engineering perspective about the current state and future challenges of serverless computing research. I will describe our solutions in autonomic resource management for serverless platforms, focusing on solutions that improve performance or reduce costs via scheduling, caching, and right-sizing of resources, along with our ongoing efforts in developing an application-driven serverless benchmark.

Speaker’s bio: Cristina L. Abad is a Professor in the Department of Electrical Engineering and Computer Science at Escuela Superior Politécnica del Litoral in Guayaquil-Ecuador, where she leads the Distributed Systems Research Lab and co-directs the Big Data Research Group. She received her Ph.D. in 2014 from the University of Illinois at Urbana-Champaign. For three years during her PhD, she was a member of the Hadoop Core Team at Yahoo, where she worked on workload modeling and evaluation of the HDFS and had the opportunity to contribute to the Apache Hadoop codebase. Her research interests lie at the intersection of Distributed Systems and Performance Engineering. In particular, her contributions focus on designing and building distributed systems that can self-adapt to workload changes and maximize performance. She is particularly interested in improving the systems supporting Big Data applications and cloud computing architectures. Her international funding sources have included VLIR-UOS, Google, Microsoft, Amazon Web Services, and AT&T Labs Research. She has received a Fulbright Fellowship, a UIUC CS Excellence Fellowship, and two Google Faculty Research Awards. Cristina is a member of IEEE, ACM, SPEC RG, and Usenix, and is the elected Secretary of the SPEC RG.