- I am a Research Software Engineer in the Software and Data Systems (SDSR) Research Lab at Nokia Bell Labs. [Curriculum Vitae].
- I work on a wide-range of research topics on IoT, edge computing, cloud.
- I like to code.
- I am self-motivated.
- Elasticity for data stream processing applications.
- Modelling distributed system behaviours.
- Mono- and multi-objective optimisation problems.
- (Near) Real-time solutions for big data analytics.
- Machine Learning (ML) and Reinforcement Learning (RL).
- IoT, Cloud, Edge and Fog computing.
- Federated Learning and Transfer Learning.
- Digital Twins.
- 2022 - present: Research Software Engineer.
- Leader: Lieven Trappeniers.
- Hosting team: Federated Stream Processing in Software & Data Systems (SDSR) Research Lab at the Nokia Bell Labs.
- Description: A wide variety of applications in Industry 4.0, AR/VR, IoT, smart technologies, connected cars, among others, produce massive amounts of data that must be analyzed seamlessly and near real-time fashion. In such scenarios, systems must handle strong computing and storage requirements, which change over time. This happens because in these scenarios the data generation is skewed, resources are prone to failures, network is unstable, etc. When any of these changes appeared in the system, it must adjust the deployment gracefully without affecting SLAs such as throughput, end-to-end latency, cost, etc. Additionally, these complex systems bring together the needs of multiple actors from data producers offering their data (e.g., sensors, cameras, monitoring systems) on a marketplace to a wide range of end-user applications running on a multi-tenant environment. Federated stream processing systems have emerged as a powerful tool to address these needs. These systems combine federation – which assumes that multiple companies operate sites for providing data, computation, storage, and communication – and distributed approach – which corresponds to spreading computing and data across a network topology. My research aims to answer questions like:
- How to provide data lineage/provenance/sovereignty?
- How to orchestrate/mediate the utilization of multiple services, platforms (public, private, and hybrid), and data providers?
- How to achieve an elastic, scalable, seamless and low-cost execution environment?
- How to enhance stream processing applications by allowing users to define their needs related to resilience, privacy, and robustness?
I am also working in research projects in the domains of distributed file systems, distributed storage systems, DNN partitioning, DNN early-exit, DNN placement, etc.
- 2019 - 2022: Postdoctoral Fellow in the Departament of Computer Science.
- Leader: Professor Eyal de Lara.
- Hosting team: Computer Systems and Networks Group in Department of Computer Science at the University of Toronto.
- Description: Distributed Stream Processing (DSP) applications are increasingly used in new pervasive services that process enormous amounts of data in a seamless and near real-time fashion. Edge computing has emerged as a means to minimise the time to handle events by enabling processing (i.e., operators) to be offloaded from the Cloud to the edges of the Internet, where the data is often generated. Deciding where to execute such operations (i.e., edge or cloud) during application deployment or at runtime is not a trivial problem. One of my goals is to improve performance metrics by introducing mechanisms for deploying DSP applications across Cloud and edge resources. I also participate in the research projects of Professor Eyal de Lara.
- 2016 - 2019: Ph.D. in Computer Science.
- Title: Algorithms for big data analytics. Thesis here
- Advisors: Marcos Dias de Assunção and Laurent Lefèvre.
- Hosting team: AVALON-Team in LIP at ENS-Lyon.
- Defense: September 2019.
- Approach: I worked on the subject “Algorithms for Elastic Big-Data Stream Analytics” where I developed QoS-aware mechanisms for (re)configuring data stream processing applications across edge and cloud resources. During my investigation, I introduced models and solutions for placing (near)real-time applications on heterogeneous infrastructures addressing single and multiple performance metrics. The techniques and methods covered by the research include: queueing theory, Markov Decision Process (MDP), Reinforcement Learning (RL), series-parallel graphs, Monte-Carlo Tree Search (MCTS), Temporal Difference Tree Search (TDTS), Q-learning and greedy algorithms.
- 2012 - 2014: M.Sc. in Computer Science.
Technnical Program Committees
- ICFEC 2023, IEEE SmartComp 2023, DEBS 2023, ICDCS 2023, EuroDW 2022, CCGrid 2022, ICFEC 2022, IEEE SmartComp 2022, ICPP 2020, InterCloud-HPC 2020, and ACM/IEEE Symposium on Edge Computing 2020.
- IEEE Transactions on Cloud Computing, IEEE Transactions on Parallel and Distributed Systems, Information Sciences, ACM Transactions on Autonomous and Adaptive Systems, Computing, Cybernetics and Systems, and Communications of the ACM.
- CCGrid 2017, and IEEE Globecom 2017 and 2018.
- February 10, 2023: I was invited to give a talk at UCLouvain - title: “Computation and Storage Systems for the Edge”.
- December 17, 2022: I was invited to join the TPC of SMARTCOMP 2023.
- December 17, 2022: I was invited to join the TPC of ICFEC 2023.
- November 14, 2022: I was invited to serve as a workshop chair at IC2E 2023.
- October 20, 2022: I was invited to be a PC member of DEBS 2023.
- October 7, 2022: I was invited to be a PC member of ICDCS 2023.
- August 16, 2022: I started at Nokia Bell Labs (Antwerp, Belgium) as a Research Software Engineer.
- April 5, 2022: A paper titled “Shepherd: Seamless Stream Processing on the Edge” was accepted at SEC 2022.
- March 12, 2022: A paper titled “Combining DNN Partitioning and Early Exit” was accepted at EdgeSys 2022.
- November 9, 2021: I was invited to be a PC member of CCGrid 2022.
- November 4, 2021: I gave a seminar entitled “Data Stream Processing on the Edge” at McGill University.
- October 27, 2021: I was invited to be a PC member of IEEE SmartComp 2022.
- October 26, 2021: I was invited to be a PC member of ICFEC 2022.
- July 20, 2021: A paper titled “Latency-Aware Strategies for Deploying Data Stream Processing Applications on Large Cloud-Edge Infrastructure” was accepted at IEEE Transactions on Cloud Computing.
- December 14, 2020: A paper titled “Scalable Joint Optimization of Placement and Parallelism of Data Stream Processing Applications on Cloud-Edge Infrastructure” was accepted at International Conference on Service Oriented Computing (ICSOC 2020).
- December 2, 2020: A paper titled “Boosting Big Data Streaming Applications in Clouds with BurstFlow” was accepted at IEEE Access.
- July 6, 2020: A paper titled “An Optimal Model for Optimizing the Placement and Parallelism of Data Stream Processing Applications on Cloud-Edge Computing” was accepted at The International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2020).
- July 6, 2020: A paper titled “Scalable Joint Optimization of Placement and Parallelism of Data Stream Processing Applications on Cloud-Edge Infrastructure” was accepted at the International Conference on Service Oriented Computing (ICSOC 2020)(ICSOC).
- July, 2020: I was invited to be a PC member of ACM/IEEE Symposium on Edge Computing.
- June 17, 2020: I was invited to be a PC member of InterCloud-HPC 2020.
- September 23, 2019: Ph.D. defense [talk] entitled “Quality of Service Aware Mechanisms for (Re)Configuring Data Stream Processing Applications on Highly Distributed Infrastructure”.
- August 30, 2019: I was invited to be a PC member of ICPP 2020.
- August 21-23, 2019: I visited the University of Toronto and I also gave a talk [talk].
- August 7, 2019: A paper titled “Monte-Carlo Tree Search and Reinforcement Learning for Reconfiguring Data Stream Processing on Edge Computing” was accepted at The International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2019).
- August 5, 2019: I attended the 48th International Conference on Parallel Processing and I gave a [talk].
- May 21, 2019: A paper entitled “Multi-Objective Reinforcement Learning for Reconfiguring Data Stream Analytics on Edge Computing” was accepted at 48th International Conference on Parallel Processing (ICPP).
- May 17, 2019: I attended the 19th Annual IEEE/ACM International Symposium in Cluster, Cloud, and Grid Computing held in Larnaca, Cyprus and I gave a [talk].
- February 16, 2019: A paper entitled “Distributed Operator Placement for IoT Data Analytics Across Edge and Cloud Resources” was accepted at IEEE/ACM International Symposium in Cluster, Cloud, and Grid Computing (CCGrid 2019).
- November 12-15, 2018: I attended the 16th International Conference on Service Oriented Computing held in Hangzhou, Zhejiang, China and I gave a [talk].
- July 23, 2018: A paper entitled “Latency-Aware Placement of Data Stream Analytics on Edge Computing” was accepted at International Conference on Service Oriented Computing (ICSOC).
- July 10, 2018: I attended the USENIX Workshop on Hot Topics in Edge Computing held in Boston, MA, USA.
- June 23, 2018: A poster entitled “Latency-Aware Strategies for Placing Data Stream Analytics onto Edge Computing” was accepted at USENIX Workshop on Hot Topics in Edge Computing (HotEdge ‘18).
- June 28-30, 2017: I attended the COMPAS held in Sophia-Antipolis, France and I gave a [talk].
- April 29, 2017: A paper was accepted in the Conférence d’informatique en Parallélisme, Architecture et Système(COMPAS).
- October 7, 2017: A paper titled “Resource Elasticity for Distributed Data Stream Processing: A Survey and Future Directions” was accepted at the Journal of Network and Computer Applications.
- July 27-30, 2015: I attended the 21st International Conference on Parallel, and Distributed Processing Techniques and Applications held in Las Vegas, Nevada, US and I gave a [talk]
- April 30, 2015: A paper entitled “BSPonP2P: Towards Running Bulk-Synchronous Parallel Applications on P2P Desktop Grids” was accepted at International Conference on Parallel and Distributed Processing Techniques and Applications.