Papaemmanouil Receives Funding from Huawei Technologies

Olga PapaemmanouilShenzhen-based Huawei Technologies, the largest manufacturer of telecom equipment in the world, is supporting Associate Professor of Computer Science Olga Papaemmanouil‘s efforts to develop machine learning approaches for managing the performance of data management systems. The grant will support research on workload management, that is the task of query placement, query scheduling and resource allocation for database applications. Workload management is an extremely critical task for database systems as it can impact the execution time of incoming processing tasks as well as the overall perceived performance of the database  and the quality of the service (QoS) offered to end-clients. The complexity of the problem increases for applications that involve dynamically changing workloads and concurrently executing queries sharing the same underlying resources, as well as applications that are deployed on data clusters with fluctuating resource availability.

Dr. Papaemmanouil’s research aims to design frameworks that can be trained on application-specific properties and performance metrics  to automatically learn how to effectively dispatch incoming queries across a cluster of servers, implicitly solving the resource allocation challenge. These techniques will rely on machine learning algorithms (reinforcement learning and deep learning)  that model the interaction of concurrently running queries  as well as the relationship between database performance and the underlying resource availability in the cluster. The project will lead the way towards the development of workload management solutions that eliminate ad-hoc heuristics often used by database administrators to address these challenges and replace them with software modules capable of providing custom workload management strategies to end-clients.

Natural Language Annotation for Machine Learning

From the Computer Science Department blog:

James Pustejovsky and his student Amber Stubbs have a new book “Natural Language Annotation for Machine Learning” out from O’Reilly Books and Media: “Systems exist for analyzing existing corpora, but making a new corpus can be extremely complex. To help you build a foundation for your own machine learning goals, this easy-to-use guide includes case studies that demonstrate four different annotation tasks in detail. You’ll also learn how to use a lightweight software package for annotating texts and adjudicating the annotations.”

Protected by Akismet
Blog with WordPress

Welcome Guest | Login (Brandeis Members Only)