Inhalt des Dokuments
- © CC-BY-SA-3.0
- © Chen Xu
A Member from TU Berlin's Database Systems and Information Management Group recently presented a research paper at the 32nd IEEE International Conference on Data Engineering (ICDE 2016), which was held between May 16 and May 20 in Helsinki, Finland. The attendee was DIMA Postdoc, Dr. Chen Xu.
Key highlights include:
DIMA Research Paper
The accepted research paper at this year is: Efficient Fault-tolerance for Iterative Graph Processing on Distributed Dataflow Systems.
In this work, we propose a novel fault-tolerance mechanism for iterative graph processing on distributed data-flow systems with the objective to reduce the checkpointing cost and failure recovery time. In particular, in contrast to the typical unblocking checkpointing approaches (i.e., managing checkpoints independently for immutable datasets), we inject the checkpoints of mutable datasets into the iterative dataflow itself. Moreover, we achieve speedier recovery, i.e., confined recovery, by using the local log files on each node to avoid a complete re-computation from scratch.
 Efficient Fault-tolerance for Iterative Graph Processing on Distributed Dataflow Systems, Chen Xu, Markus Holzemer, Manohar Kaul and Volker Markl.