TY - GEN
T1 - Adaptive exploration for large-scale protein analysis in the molecular dynamics database
AU - Nutanong, Sarana
AU - Carey, Nick
AU - Ahmad, Yanif
AU - Szalay, Alex S.
AU - Woolf, Thomas B.
PY - 2013
Y1 - 2013
N2 - Molecular dynamics (MD) simulations generate detailed time-series data of all-atom motions. These simulations are leading users of the world's most powerful supercomputers, and are standard-bearers for a wide range of high- performance computing (HPC) methods. However, MD data exploration and analysis is in its infancy in terms of scalability, ease-of-use, and ultimately its ability to answer 'grand challenge' science questions. This demonstration introduces the Molecular Dynamics Database (MDDB) project at Johns Hopkins, to study the co-design of database methods for deep on-the-fly exploratory MD analyses with HPC simulations. Data exploration in MD suffers from a "human bottleneck", where the laborious administration of simulations leaves little room for domain experts to focus on tackling science questions. MDDB exploits the data-rich nature of MD simulations to provide adaptive control of the exploration process with machine learning techniques, specifically reinforcement learning (RL). We present MDDB's data and queries, architecture, and its use of RL methods. Our audience will co-operate with our steering algorithm and science partners, and witness MDDB's abilities to significantly re- duce exploration times and direct computation resources to where they best address science questions.
AB - Molecular dynamics (MD) simulations generate detailed time-series data of all-atom motions. These simulations are leading users of the world's most powerful supercomputers, and are standard-bearers for a wide range of high- performance computing (HPC) methods. However, MD data exploration and analysis is in its infancy in terms of scalability, ease-of-use, and ultimately its ability to answer 'grand challenge' science questions. This demonstration introduces the Molecular Dynamics Database (MDDB) project at Johns Hopkins, to study the co-design of database methods for deep on-the-fly exploratory MD analyses with HPC simulations. Data exploration in MD suffers from a "human bottleneck", where the laborious administration of simulations leaves little room for domain experts to focus on tackling science questions. MDDB exploits the data-rich nature of MD simulations to provide adaptive control of the exploration process with machine learning techniques, specifically reinforcement learning (RL). We present MDDB's data and queries, architecture, and its use of RL methods. Our audience will co-operate with our steering algorithm and science partners, and witness MDDB's abilities to significantly re- duce exploration times and direct computation resources to where they best address science questions.
UR - http://www.scopus.com/inward/record.url?scp=84883047083&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84883047083&partnerID=8YFLogxK
U2 - 10.1145/2484838.2484872
DO - 10.1145/2484838.2484872
M3 - Conference contribution
AN - SCOPUS:84883047083
SN - 9781450319218
T3 - ACM International Conference Proceeding Series
BT - SSDBM 2013 - Proceedings of the 25th International Conference on Scientific and Statistical Database Management
T2 - 25th International Conference on Scientific and Statistical Database Management, SSDBM 2013
Y2 - 29 July 2013 through 31 July 2013
ER -