Andrew Pavlo

Creating a large-scale database application is easier now than it ever has been, in part due to the proliferation of distributed system tools, cloud-computing platforms, and affordable mobile sensors. But now the processing and storage needs of Internet-scale, "Big Data" applications are surpassing the limitations of legacy database management systems (DBMSs). As a result, I am interested in the research and development of new DBMS technologies for these modern high-volume and data-intensive applications.

In particular, my research is focused on novel distributed and parallel DBMS architectures for transaction processing applications (OLTP), analytical/business intelligence workloads (OLAP), and scientific computing. Much of my work is in applying techniques from machine learning and optimization research to enable these distributed DBMSs to execute workloads that are beyond what single-node systems can support. I am also interested in studying the performance characteristic's of non-volatile memory devices in the context of Big Data systems in order to build the groundwork for new DBMS architectures that can take advantage of these emerging technologies.