Thomas S. Repantis

http://www.ninewhilenine.org

Interests

Highly-scalable, fault-tolerant, real-time, distributed systems.

Professional Experience

Akamai Technologies, Platform Engineering
Cambridge, MA, October 2008-Present
Principal Lead Software Engineer, January 2015-Present
Leading the team that develops the alerting infrastructure responsible for monitoring Akamai's platform with operational efficiency.
Principal Software Engineer, July 2013-December 2014
Carried out scalability projects in Query, a distributed, event-based system that continuously processes data from the entire Akamai platform. Mentored over half a dozen senior software and performance engineers.
Senior Software Engineer, February 2010-June 2013
Designed and implemented in C++ multi-threaded system software for real-time publication, aggregation, delivery, and processing of data across Akamai's distributed platform. Developed C, C++, Java, Python, and Perl interfaces used by both internal and customer-facing applications for monitoring, alerting, and reporting.
Senior Performance Engineer, October 2008-January 2010
Used and developed tools to measure and analyze the performance, robustness, and scalability of large distributed systems. Took end-to-end responsibility of complex systems.
University of California, Riverside
Distributed Real-Time Systems Laboratory
Designed, implemented in Java, and evaluated distributed protocols for sharing-aware component composition, load prediction and hot-spot alleviation, replica placement for high availability, and data dissemination in peer-to-peer systems. Led the development of the Synergy distributed stream processing middleware; supervised other student research projects that used the platform.
IBM Research
Advanced Enterprise Middleware, Watson Research Center, Hawthorne, NY, Summer 2007
Developed in Java a replication middleware for distributed, multi-tier, server architectures. Quantified the server replication and data partitioning performance benefits, as well as the consistency overhead, using the TPC-W transactional web commerce benchmark. Filed a patent application for the middleware's efficient, distributed, strong-consistency protocol.
Intel Research
Corporate Technology Group, Pittsburgh, PA, Summer 2006
Contributed to the reliable email project. Built in C++ an event-driven, collaborative spam filter that employed a distributed protocol to defend against sybil attacks.
Hewlett-Packard
Enterprise Storage & Servers, Colorado Springs, CO, Summer 2005
As a member of the replication team of an upcoming product of HP's grid storage portfolio, developed in C++ and documented a logging mechanism used for asynchronous replication in a distributed disk array.
University of Patras, Greece
High Performance Information Systems Laboratory
Implemented in C a protocol for dynamic memory page migration across the nodes of a Software Distributed Shared Memory System, as part of an inter-departmental diploma thesis. Dynamic page migration improved performance by increasing locality and adaptability, while remaining transparent to the application programmer.
FGAN e.V. (Fraunhofer FKIE)
Bonn, Germany, Summer 2000
Analyzed the H.323 protocol family, used for multimedia applications (VoIP) in packet switched networks, and summarized the results in a technical report, including detailed protocol description and performance evaluation of applications under IPv6 in Solaris.

Education

Ph.D. in Computer Science
University of California, Riverside, August 2008
Thesis: Synergy: Quality of Service Support for Distributed Stream Processing Systems
M.Sc. in Computer Science
University of California, Riverside, August 2005
Thesis: Adaptive Data Dissemination and Content-Driven Routing in Peer-to-Peer Systems
Diploma in Electrical & Computer Engineering
(5-year program)
University of Patras, Greece, March 2003
Thesis: Implementation of Page Forwarding on Clusters

tsr home
cd /home