Kineograph: taking the pulse of a fast-changing and connected world

By Raymond Cheng, Ji Hong, Aapo Kyrola, Youshan Miao, Xuetian Weng, Ming Wu, Fan Yang, Lidong Zhou, Feng Zhao, Enhong Chen. European Conference on Computer Systems, 2012

305 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Statistical and Nonlinear Physics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/282bc59faefb734137d2ea978cb1eb5699e67c7c

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Statistical and Nonlinear Physics

In Signal Processing

Citations in the corpus

PowerLyra: Differentiated Graph Computation and Partitioning on Skewed Graphs (2019, background): “There are also a few systems considering GPU [45, 80, 81], NUMA [100], RDMA [64, 83], stream processing [17, 55, 96], and temporal graphs [27].”
PowerGraph: Distributed Graph-Parallel Computation on Natural Graphs (2012, unknown): “Alternatively, Kineograph [11] presents a graph-parallel framework for time-evolving graphs which mixes features from both GraphLab and Pic-colo.”
Mizan: a system for dynamic load balancing in large-scale graph processing (2013, background): “Kineograph [5] is a distributed system that targets the processing of fast changing graphs.”
GraphX: Graph Processing in a Distributed Dataflow Framework (2014, methodology): “of the earlier examples is Kineograph [9], a distributed graph processing system that constructs incremental snapshots of the graph for offline static graph analysis.”
From "Think Like a Vertex" to "Think Like a Graph" (2013, background): “As a result, a number of distributed/parallel graph processing systems have been proposed, including Pregel [15], its open source implementation Apache Giraph [1], GraphLab [14], Kineograph [6], Trinity [20], and Grace [23].”
GraphChi: Large-Scale Graph Computation on Just a PC (2012, unknown): “This setting was recently studied by Cheng et. al ., who proposed Kineograph [15], a distributed sys-tem for processing a continuous in-ﬂow of graph updates, while simultaneously running advanced graph mining algorithms.”
Naiad: a timely dataflow system (2013, methodology): “When run on the same Twitter stream as Kineograph, using 32 computers and ingesting 1,000 tweets per epoch on each machine, the average through-put over ﬁve runs is 482,988 t/s with no fault-tolerance, 322,439 t/s with checkpoints every 100 epochs, and 273,741 t/s with continual logging.”
GraphX: a resilient distributed graph system on Spark (2013, background): “In response to the growing size and importance of graph data, a range of new large-scale distributed graph-parallel frameworks ( e.g., Pregel [9], PowerGraph [6], and others [5, 3, 11]) have emerged.”
X-Stream: edge-centric graph processing using streaming partitions (2013, background): “The fact that X-Stream starts from an unordered edge list means that it can easily handle growing graphs - similar to distributed systems such as Kineograph [25].”