Proceedings of the Workshops of the EDBT/ICDT 2015 Joint Conference (EDBT/ICDT), Brussels, Belgium, March 27th, 2015

نویسندگان

  • Peter M. Fischer
  • Gustavo Alonso
  • Marcelo Arenas
  • Floris Geerts
چکیده

This talk describes a new graph-pattern engine called Jedi.Using a recent simplification of worst-case optimal join al-gorithms due to Ngo et al., Jedi translates join queries intoa series of set intersection and union operations. Such setoperations are ideally suited to modern CPUs that providessingle-instruction, multiple data (SIMD) instructions. Usingthese ideas, we demonstrate that Jedi outperforms special-ized graph engines by over an order of magnitude and rela-tional systems by over two orders of magnitude on standardgraph processing queries over real data.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Message from the EDBT Workshops Chairs

The International Conference on Extending Database Technology (EDBT) and the International Conference on Database Theory (ICDT) are two prestigious forums for the exchange of the latest research results in the technological elements and theoretical foundations of data management and database systems. This year, five exciting workshops that focus on emerging topics in data management are co-loca...

متن کامل

Proceedings of the Workshops of the EDBT/ICDT 2014 Joint Conference (EDBT/ICDT 2014), Athens, Greece, March 28, 2014

MapReduce framework is established as the standard ap-proach for parallel processing of massive amounts of data. Inthis work, we extend the model of MapReduce scheduling onunrelated processors (Moseley et al., SPAA 2011) and dealwith the practically important case of jobs with any numberof Map and Reduce tasks. We present a polynomial-time(32 + )-approximation algorithm ...

متن کامل

Assignment of Different-Sized Inputs in MapReduce

A MapReduce algorithm can be described by a mapping schema, which assigns inputs to a set of reducers, such that for each required output there exists a reducer that receives all the inputs that participate in the computation of this output. Reducers have a capacity, which limits the sets of inputs that they can be assigned. However, individual inputs may vary in terms of size. We consider, for...

متن کامل

Lower Bounds on the Communication of XPath Queries in MapReduce

We present two algorithms, each depending on a different data fragmentation of the XML tree. They both compute XPath queries in MapReduce, by first computing subqueries and then combining their results. We compute the replication rate of each algorithm and show it is less than 2.

متن کامل

What are the Most Important Research Challenges in Energy Data Management? (panel)

This panel paper aims at initiating discussion at the Fourth International Workshop on Energy Data Management (EnDM 2015) about what the most important research challenges within Energy Data Management are. The author is the panel organizer, extra panelists will be recruited from the workshop audience.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره 1330  شماره 

صفحات  -

تاریخ انتشار 2015