My name is Ippokratis Pandis and I like to build data processing systems. My work focuses on efficient, scalable data warehousing and transaction processing in modern hardware and on the Cloud. Currently, I am principal engineer at Amazon Web Services (AWS) and I am working on AWS Redshift. AWS Redshift is Amazon's fully managed, petabyte-scale data warehouse.
Before AWS, I was a software engineer working on the Impala project at Cloudera.
Impala is an open source, MPP SQL query processing engine that runs natively in Apache Hadoop.
Before Cloudera, I was a member of the Research Staff (RSM) at IBM Almaden Research Center in San Jose, CA. At IBM I was actively involved in the BLU project. My IBM webpage can be found here.
DB2 BLU is the column-store (in) DB2. It was first released as part of IBM DB2 LUW v10.5 with BLU Acceleration. BLU deeply integrates innovative techniques for processing column-organized tables that speed read-mostly queries by 10 to 100 times and improve storage savings by 3 to 10 times, compared to traditional row-organized tables, without the complexity of defining indexes or materialized views. A thorough technical description of DB2 BLU can be found here.
Prior joining IBM, I was a Ph.D. candidate of the Electrical and Computer Engineering Department at Carnegie Mellon University. My advisor was Prof. Anastassia Ailamaki.
During my PhD I worked on the StagedDB/CMP project. This project studied the behavior of modern DBMSs on the emerging multi-core (CMP) and multi-processor (SMP) environments, and tried to enhance their performance by proposing new designs both in the software, and hardware level. For more details: Staged Database Systems
As part of our research, our team has implemented Shore-MT. Shore-MT is a scalable multi-threaded port of the Shore storage manager. It can be downloaded from here.
My PhD thesis focused on the scalability of transaction processing systems and introduces a data-oriented transaction execution architecture (or DORA). A presentation about data-oriented transaction execution and Shore-MT can be found here.
I am co-organizing the Tenth International Workshop on Data Management on New Hardware, or DaMoN 2014. DaMoN is a one day workshop co-located with SIGMOD. SIGMOD 2014 will be held in Snowbird, UT.
I am the PC area chair for the Main-memory, Parallel and Distributed Database systems area of the DB Track of ACM CIKM 2014. CIKM 2014 will be held in Shanghai, China.
Also: Publications, Google Scholar profile, and DBLP Entry