GraphX: a resilient distributed graph system on Spark

By Reynold Xin, Joseph E. Gonzalez, M. Franklin, Ion Stoica. International Workshop on Graph Data Management Experiences and Systems, 2013

704 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems

View Paper: https://www.semanticscholar.org/paper/31f27864950a6c417cf996927b2d5558f70d2b14

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Information Systems

Citations in the corpus

GridGraph: Large-Scale Graph Processing on a Single Machine Using 2-Level Hierarchical Partitioning (2015, methodology): “4xlarge instances (each with 8 cores, 68.4GB RAM, 2 840GB HDDs, $0.98/h), using the result from [28].”
EmptyHeaded: A Relational Engine for Graph Processing (2015, methodology): “ed queries, we compare against LogicBlox, a commercial system that implements a worst-case optimal join algorithm. We also compared our performance to SociaLite [52], Vertica, PostgreSQL, and Graph-X [58] and found that each of these systems was at least two orders of magnitude slower on all queries”
A survey on platforms for big data analytics (2014, neutral): “On the top of the stack are many Spark wrappers such as Spark Streaming (Large Scale real-time stream processing), Blink DB (queries with bounded errors and bounded response times on very large data) [23], GraphX (Resilient distributed Graph System on Spark) [24] and MLBase (distributed machine lear”
Big data analytics on Apache Spark (2016, methodology): “Several libraries have been built on top of Spark core for handling different workloads: Spark’s MLlib for machine learning [61], GraphX [33,83] for graph processing, Spark 19 http://mattturck.com/2016/02/01/big-data-landscape/.”
Hop: Heterogeneity-aware Decentralized Training (2019, background): “Comparison to Parameter Servers For decentralized algorithm, training is conducted on a ring-based topology (Figure 11); for PS, we adopt BSP and use one additional machine as the parameter server.”
GraphMat: High performance graph analytics made productive (2015, methodology): “Some other ways to process graphs include writing vertex programs as UDFs for use in a column store [18] and GraphX [33] (set of graph primitives intended to work with Spark [2]).”
BigDansing: A System for Big Data Cleansing (2015, methodology): “It then uses GraphX [37] to ﬁnd all connected components in the hypergraph.”
Knowledge Graphs (2020, neutral): “Scalable frameworks for graph analytics [80, 126, 148] can be leveraged for computing centrality, clustering, summarisation, and so on, to gain insights about the domain being described.”
Learning Graph Representations with Embedding Propagation (2017, background): “Whenever a computational process can be mapped to message exchanges between nodes, it is implementable in graph processing frameworks such as Pregel [29], GraphLab [23], and GraphX [44].”
Heterogeneity-aware Distributed Parameter Servers (2017, background): “Other prevailing distributed ML systems include Mahout [2], MLlib [3], Pregel [32], GraphLab [30], and GraphX [43].”
SNAP: A General Purpose Network Analysis and Graph Mining Library (2016, background): “In particular, research on graph processing in large-scale distributed environments [Gonzalez et al. 2012; Malewicz et al. 2010; Kang et al. 2009; Salihoglu and Widom 2013; Xin et al. 2013] provides efficient frameworks, but these frameworks implement only a handful of most common graph algorithms, ”
Naiad: a timely dataflow system (2013, methodology): “While one can build a Pregel-like implementation using LINQ-style operators [40], such a collection-oriented pattern makes it hard to support Pregel’s full semantics including aggregation and graph mutation.”
TicTac: Accelerating Distributed Deep Learning with Communication Scheduling (2018, background): “It is worth noting that deep learning systems with computational graphs are fundamentally different from graph processing systems (Malewicz et al., 2010; Hoque & Gupta, 2013; Xin et al., 2013).”