Owen O'Malley Email & Phone Number

Wilmington, Delaware, Us

VP ORC 2015-2021VP Hadoop 2008-2010PMC member: Hadoop, Hive, ORC, Tez, Giraph, AmbariApache MemberApache Mentor: Accumulo, Ambari, BigTop, Chukwa, Eagle, Falcon, Giraph, HCatalog, Helix, Iceberg, Kafka, Knox, Pig, Ranger, Reef, and Tajo

Sunnyvale, Ca, Us

Santa Clara, California, Us

Hortonworks merged with Cloudera in January 2019.► Implemented Apache ORC column encryption that enables users to control who has access to columns individually and store the data encrypted on disk. Users can specify which columns to encrypt and the key to use. Integrates with various Key Management Servers.► Gave presentations on ORC, Iceberg, and open source communities at: - DataWorks Summit - 2019 (Barcelona & Washington DC) - SCaLE - 2019 - Berlin Buzzwords - 2019 - Strata NYC - 2019 - ApacheCon - 2019

Santa Clara, California, Us

Responsible for architecture across the projects with a focus on the storage layer and security. Guided open source community development, management, and Apache processes.► Designed and implemented ORC file format in both Java and C++. ORC has become the fastest way to store columnar data in the big data ecosystem.► When our competitor announced fast SQL processing, we needed speed up Hive. I worked with our team and our partners at Microsoft to speed it up. We successfully sped it up by 10x over the course of a year. That work continued with LLAP for an additional 10x.► Led the design and development of adding ACID transactions to Hive. Previously, Hive tables could only have entire new partitions added and customers needed to be able to add new data or modify data atomically. By supporting ACID transactions, users could pick better partitions for their tables.► Worked deeply with a large customer on their usage of Hive and Spark, including a custom framework started in 2012 that allowed high-volume updates to records in Hive tables.► Designed and implemented benchmarks comparing file format performance from both Hive and Spark. Those results were used to drive our development goals for ORC.► Investigated Kubernetes and helped direct our architectural strategy toward embracing it.► Guided our Metron team through the process to become an Apache project and eventually to a top level project. I helped them build their development community to be inclusive of external developers. ► Worked with Netflix on the Iceberg project to add ORC support and championed it to become an Apache project.► Gave presentations on Hive, performance, Hadoop security, and ORC at: - Dataworks Summit - 2012 - 2018 - Strata - 2013, 2014, 2017, 2018 - ApacheCon - 2011, 2014, 2015, 2016, 2018 - Berlin Buzzwords - 2012, 2013, 2018 - Big Data Spain - 2017 - FOSS Backstage - 2017, 2018 - Various User Groups and Meetups

Sunnyvale, Ca, Us

Led Architect on the Apache Hadoop project, the original Open Source Big Data project. Responsible for taking Hadoop from a small project running on a handful of machines at Yahoo to production use on huge enterprise clusters across the world. ► First committer added to Hadoop.► Responsible for improving performance and stability as we scaled from 5 machines to thousands.► Won the 2008 Jim Gray Terabyte sort benchmark after implementing the terasort tools for Hadoop.► Won the 2009 Minute and Gray sort benchmarks after reimplementing the Map/Reduce shuffle. Sorted a petabyte on 3000 machines. ► Tech lead adding Kerberos-based security to Hadoop.► As VP of Hadoop, responsible for working with and encouraging the open source community.► Gave conference presentations about: Hadoop, Security, interacting with Apache. - Hadoop Summit - 2007, 2008, 2009, 2010 - Hadoop World - 2009 - ApacheCon - 2008, 2009, 2010, 2011 - IGT World Summit of Cloud Computing - 2008 - Kerberos Conference - 2010 - Various Hadoop User Groups►Part of the team that spun out Hortonworks from Yahoo in 2011.

Washington, Dc, Us

Worked in Java Path Finder group working on software model checking and static analysis.► Analyzed the software that managed failure scenarios on the Space Shuttle.► When the Mars Exploration Rover (MER) first landed on Mars, the planning software core dumped intermittently. The team couldn't find the problem. Worried that they would lose an entire day of science, they called me to urgently diagnose the problem. I found and fixed the problem.► Worked with NASA administration on the NASA Open Source Agreement.

Palo Alto, Ca, Us

Us

Phd, Information & Computer Science

Ms, Information & Computer Science

Bs, Math/Computer Science

Education record