Brief overview of my research interestsMy research broadly includes topics in parallel and distributed computing. In my PhD, I worked on models for cooperative parallelism (algorithm portfolio, poly-algorithms) and their applications to the resolution of SAT and CSP. In my post-docs, I worked on service and cloud computing problems: the service selection problem, the service composition problem, energy-efficient scheduling for green and distributed clouds' middlewares.
I still continue to work on these topics with my research group at LIPN. But in addition, at Qarnot computing, I started to work on models for ambiant intelligence in smart-buildings. This includes: demand response, machine learning for the recognition of daily life activities, frameworks for context-aware and in-situ machine learning for smart-buildings (application to event detection). I also work on edge computing models for IoT. This includes: Conceptual models of the role edge in IoT, work sharing models between cloud and edge, theoretical formulation of scheduling problems in edge computing.
Some recent works
- Yanik Ngoko and Denis Trystram: Revisiting Flynn’s classification: The portfolio approach, Euro-Edupar(Europar Workshop) 2017, Santiago de Compostela, Spain.
- Amaury Durand, Yanik Ngoko, Christophe Cérin: Distributed and In-situ Machine Learning for Smart-Homes and Buildings: Application to Alarm Sounds Detection. IPDPS Workshops 2017, Orlando, USA.
- Yanik Ngoko: Heating as a Cloud-Service, A Position Paper (Industrial Presentation). Euro-Par 2016, Grenoble, France.
- Thar Baker, Bandar Aldawsari, Hissam Tawfik, D. Reid, Yanik Ngoko: GreeDi: An energy efficient routing algorithm for big data on cloud. Ad Hoc Networks 35. 2015