Title: Distributed Annotations Database Performance Measurements Author: Keisuke Ito (keisukei) Abstract: The Annotations Database (ADB) is a Javelin-II system (open domain Question Answering system from LTI) component that enables retrieval, navigation and organization of a space of Documents based on semantic information. The Distributed Annotations Database (DADB) system is currently being developed to support parallel annotation for large corpora. In order to achieve the performance goals of the DADB system, there has to be a performance measurement infrastructure in place to evaluate the prototype systems when they become available (towards the end of the semester). I intend to learn about existing performance measurement techniques and create an infrastructure to actually measure performance of prototype systems of the DADB. Expected result: I hope to learn about existing techniques for measuring performance of distributed (database) systems, and apply them to create an infrastructure for the DADB performance measurement evaluations. (And use it to actually optimize the prototype DADB system for better performance during summer as my master's practicum)