TY - GEN
T1 - Evaluation of parallel graph loading techniques
AU - Then, Manuel
AU - Kaufmann, Moritz
AU - Kemper, Alfons
AU - Neumann, Thomas
N1 - Publisher Copyright:
© 2016 Copyright held by the owner/author(s).
PY - 2016/6/24
Y1 - 2016/6/24
N2 - For many exploratory graph workloads, the initial loading and construction of the graph data structures makes up a significant part of the total runtime. Still, this topic is hardly analyzed in literature and often neglected in systems and their evaluations. In this paper we analyze the whole graph loading process, including parsing, dense vertex identifier relabeling, and writing the final in-memory data structures. We present various loading strategies that take into consideration the properties of the input graph, e.g., partitioning, and evaluate them through extensive experiments.
AB - For many exploratory graph workloads, the initial loading and construction of the graph data structures makes up a significant part of the total runtime. Still, this topic is hardly analyzed in literature and often neglected in systems and their evaluations. In this paper we analyze the whole graph loading process, including parsing, dense vertex identifier relabeling, and writing the final in-memory data structures. We present various loading strategies that take into consideration the properties of the input graph, e.g., partitioning, and evaluate them through extensive experiments.
UR - http://www.scopus.com/inward/record.url?scp=84985987085&partnerID=8YFLogxK
U2 - 10.1145/2960414.2960418
DO - 10.1145/2960414.2960418
M3 - Conference contribution
AN - SCOPUS:84985987085
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the 4th International Workshop on Graph Data Management Experiences and Systems, GRADES 2016
PB - Association for Computing Machinery
T2 - 4th International Workshop on Graph Data Management Experiences and Systems, GRADES 2016
Y2 - 24 June 2016
ER -