Research
DEBS 2020 Grand Challenge
Co-Authored with Manuja DeSilva
This paper entails the technical details of an approach to the challenge presented by the DEBS 2020 committee, regarding Non-Intrusive Load Monitoring (NILM) and its relevance in the area of data streaming. Our project highlights how the open source project Apache Flink can provide an efficient solution for processing large data-sets. Furthermore, we implement a version of DBSCAN, a data clustering algorithm, and we present an effective approach for handling out of order events in a data stream. We observe that our approach strikes a balance between optimization, usability, and accuracy with room for future work. We propose a complete solution that is capable of detecting appliance power events and energy consumption by using a stream of voltage and current data.
View Paper | View VideoObsidian (Formally Manos)
Co-Authored with Manuja DeSilva
RocksDB is an in-memory database that is built using the concept of LSM(log-structured-merge) trees. Out of the box, it provides some guarantees, such as that its default parameters are suitable for most workloads. In addition, it also provides many knobs that can be adjusted by the user (the developer) to optimize the database for their particular workloads. We will build a command line interface tool to simulate workloads on RocksDB. We will include the ability to tune various parameters of RocksDB and run workloads using those parameter configurations. In addition to running these aggregated style experiments, we will also provide the ability to run individual experiments testing the affects of tuning one parameter on a dataset. Lastly, we will run our own benchmarking experiments using the CLI tool we built to test the guarantees cited by RocksDB, and devise our own parameter configurations that we predict are suitable for different types of workloads.
View Paper | View Code