Jan S. Rellermeyer
I received my MSc CS in Distributed Systems from ETH Zürich, Switzerland, in 2006 and completed my PhD in Computer Science in the Systems Group at ETH in 2011. My thesis work was on Modularity as a Systems Design Principle.
In 2010, I did an internship at Microsoft Research in Redmond, WA in the eXtreme Computing Group and worked on an operating system design and energy management for the Intel SCC (Rock Creek) architecture.
After graduation, I joined IBM Research in Austin, TX where I worked as an RSM in the Future Systems and Next Generation Datacenters Group. I was part of the team that built and released the first mobile app that IBM ever published, IBM Mobile Systems Remote. This effort lead sparked the IBM Mobile First initiative. I was also a co-lead of the winning 2013 Global Technology Outlook (GTO) topic Software-Defined Environments. After mobile systems were turned into a product group in 2015, I remained in Research and moved to workload-optimized systems for the POWER architecture where I leveraged features like coherently-attached flash through CAPI. This work is available on Github, In 2016, I worked on cloud and datacenter computing and in 2017 my group was assigned to work on machine learning and natural language processing.
In 2013, I became an Adjunct Assistant Professor in the Department of Computer Science at The University of Texas at Austin and taught the Programming Languages, Programming Languages Honors, and Principles of Computer Systems classes until 2017.
I was the Invited Researcher of the OSGi Alliance from 2008-2011 and contributed to several successful open source projects in the Eclipse and Apache Foundations. I am the project lead of the Eclipse Concierge project.
In 2017, my co-authors Gustavo Alonso, Timothy Roscoe, and I received the ACM/IFIP/USENIX Middleware Test of Time award for our 2007 paper R-OSGi: Distributed Applications through Software Modularization.
My research revolves around building and researching large-scale distributed systems for demanding applications like Big Data Processing and Distributed Machine Learning. I am interested in the architecture of systems and the resulting properties like performance, efficiency, scalability, reliability, and security.
If you are interested in writing your master's thesis under my supervision, please contact me through email.
Current PhD Students
PhD Committee Member
Do Le Quoc: Approximate Data Analytics Systems, TU Dresden, 2018. Committee: Christof Fetzer (TU Dresden, Adviser), Pramod Bhatotia (University of Edinburgh, Adviser), Jan S. Rellermeyer
Amiya K. Maji: Dependability Where the Mobile World Meets the Enterprise World. Purdue University, 2015. Committee: Saurabh Bagchi (Purdue, Adviser), Elisa Bertino (Purdue), Anand Raghunathan (Purdue), Jan S. Rellermeyer
Current Master Students
Dan Graur (thesis work partly conducted at ETH Zurich with Gustavo Alonso) - Distributed Machine Learning
Michalis Vrachasotakis (thesis work partly conducted at ARM Research, Cambridge) - Making Kubernetes Scale for Edge Computing Applications
Stefan Stojkovski (EIT Digital, thesis work conducted at Rabobank and co-supervised with Dick Epema) - Addressing the challenges of Cloud Computing Adoption in an Enterprise Environment
Apourva Parthasarathy - Reducing the JVM Startup Overhead in Big Data Systems
Mark Kluit - Reducing the Serialization Overhead in Big Data Systems
Gaomei Shi - Cooperative Censorship Circumvention
Past Master Students
Gabriel Vilén (EIT Digital, thesis conducted at Logical Clocks and co-supervised with Dick Epema) - Mounting External Storage in HopsFS
Master Thesis Committee Member
Danilo Verhaert, TU Delft, 2018: An Architecture-Agnostic Protection Interface for the Tock Operating System. Committee: Koen Langendoen, Jan S. Rellermeyer, Mitra Nasri.
Mingfeng Li, TU Delft, 2018: Early DNA Analysis Using Incomplete DNA Data. Committee: Zaid Al-Ars, Arjan van Genderen, Jan S. Rellermeyer
Tim Hegeman, TU Delft, 2018: Experimental Performance Analysis of Graph Analytics Frameworks. Committee: Alexandru Iosup, Jan S. Rellermeyer, Andy Zaidman.
Olaf Maas, TU Delft, 2018: Towards Language Parametric Web-Based Development Environments. Committee: Eelco Visser, Peter Mosses (Swansea University), Jan S. Rellermeyer
Bas van IJzendoorn, TU Delft, 2018: Communicating with low latency peers - Building a low latency overlay in P2P networks. Committee: Johan Pouwelse, Jan S. Rellermeyer, Matthijs Spaan
Christos Kyprianou, TU Delft, 2018: Permissionless Banking API. Committee: Johan Pouwelse, Jan S. Rellermeyer, Zeki Erkin
Huang-da Chi, TU Delft, 2017: Parallelizing a Video Filterchain for Multi- and Many-core Systems. Committee: Zaid Al-Ars, Mauricio Alvarez-Mesa (Spin Digital), Jan S. Rellermeyer
ChangLiang Luo, TU Delft, 2017: Peer Discovery With Transitive Trust in Distributed System. Committee: Johan Pouwelse, Neil Yorke-Smith, Jan S. Rellermeyer
Past Honors Thesis Students
Matthew Allen: UT Austin, 2016: Parametric Polymorphism in the Go Language.