Modern communication mechanisms, such as Remote Direct Memory Access (RDMA), have significantly lowered the CPU cost of large data transfers. However, these performance advantages can only be leveraged through interleaving the computation with the network communication and through careful management of the RDMA buffers used for sending and receiving data. In this project, we investigate how database operators, like join or aggregation algorithms, need to be designed in order to take full advantage of all the features provided by modern low-latency high-throughput networks such as InfiniBand. Our results show that hardware-concious distributed algorithms exhibit good performance and scalability.
Rack-scale computers are becoming the building blocks of modern data centers. They are characterized by new hardware technologies that promise terabytes of main memory and thousands of CPU cores connected with high-speed interconnects. These systems face a variety of challenges, one of which is the orchestration of the processing and communication resources in order to achieve maximum performance. In this project, we explore how to design a database for a rack-scale system. We investigate several approaches for distributing a state-of-the-art relational database on a modern InfiniBand cluster and evaluate these techniques in terms of performance, scalability and complexity.
|Fall 2016||G. Alonso||Advanced Systems Lab||Course Webpage|
|Fall 2016||R. Marti||Information Systems for Engineers||Course Webpage|
|Spring 2016||G. Alonso||Data Modelling and Databases||Course Webpage|
|Fall 2015||L. Fässler||Foundations of Computer Science||Course Webpage|
|Spring 2015||F. Friedrich||Computer Science II for Civil Engineers||Course Webpage|
|Fall 2014||G. Alonso||Advanced Systems Lab||Course Webpage|
|Spring 2014||D. Kossmann||Data Modelling and Databases||Course Webpage|
|Spring 2015||A. Dovis||Data Processing System Decoupling over RDMA/InfiniBand||J. Giceva
|Spring 2015||S. Dollberg||Persistence and Durability for Project FBX||J. Giceva
|ETH Zurich – Systems Group||October 2013 – present|
|Research & Teaching Assistant|
The ETH Systems Group is a research and teaching collaboration among several professors in the general area of systems. My research interest lies in the area of distributed query processing. In addition to research, my duties include teaching activities at both the under-graduate and graduate levels.
|Oracle Labs||April 2014 – September 2014|
|Redwood Shores, California, United States of America|
Internship as a research assistant at Oracle Labs, working on Project RAPID, a hardware-software co-design project targeting large-scale data management and analysis. Oracle Labs is the sole organization at Oracle which is devoted exclusively to research. Its mission is to identify and explore new technologies.
|IBM Research||October 2012 – January 2013|
Internship as a research assistant in the BlueZ Business Computing Group at the IBM Zurich Research Laboratory in Switzerland. IBM Research – Zurich is the European branch of IBM's worldwide research division and conducts fundamental and applied research that contributes to IBM products, services, and solutions.
|Root S.A.||June 2010 – September 2010|
Internship as a software engineer, working on company-internal software for datacenter operations, server management and internet-backbone monitoring. Root S.A. is one of the leading internet backbone and colocation providers in Luxembourg, specialized in dedicated servers, colocation services and web hosting.
|Doctor of Sciences (Dr. sc. ETH Zurich) – aspired degree||October 2013 – present|
|ETH Zurich, Zurich, Switzerland|
PhD studies in computer science focusing on distributed query processing. At present, I am exploring how distributed database operators, like join or aggregation algorithms, need to be designed in order to take advantage of modern communication mechanisms such as Remote Direct Memory Access (RDMA).
|Master of Science ETH in Computer Science (M.Sc. ETH CS)||September 2007 – August 2013|
|ETH Zurich, Zurich, Switzerland|
Bachelor and master studies in computer science with specialization in distributed systems. In August 2013, I completed my master thesis titled Join Algorithms on Hierarchical Network Topologies under the supervision of Prof. Dr. Donald Kossmann and Dr. Simon Loesing.
ETH Zurich – Systems Group
Department of Computer Science
CAB E 69
8092 Zurich, Switzerland