Streaming graph challenge: Stochastic block partition

Kao, Edward K.; Gadepally, Vijay; Hurley, Michael B.; Jones, Michael; Kepner, Jeremy; Mohindra, Sanjeev; Monticciolo, Paul; Reuther, Albert; Samsi, Siddharth; Song, William; Staheli, Diane; Smith, Steven T.

doi:10.1109/hpec.2017.8091040

Cited by 62 publications

(61 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the Sparse DNN Challenge different DNNs were created with different numbers of neurons per layer. The RadiX-Net parameters used to create the base DNNs are given in Table I [16,16,16,16,16,16,16] produces a 6 layer, 1024 neurons per layer DNN with 32 connections per neuron.…”

Section: Neural Network Datamentioning

confidence: 99%

“…The non-zero values are written as triples to a .tsv file where each row corresponds to a different image, each column is the nonzero pixel location and the value is 1. [16,16,16,16,16,16,16,16,16,16,16,16,16] 12 65536 0.0005 -0.45 • Use an implementation that could work on real-world data • Create compressed binary versions of inputs to accelerate reading the data • Split inputs and run in data parallel mode to achieve higher performance (this requires replicating weight matrices on every processor and can require a lot of memory) • Split up layers and run in a pipeline parallel mode to achieve higher performance (this saves memory, but requires communicating results after each group of layers) • Use other reasonable optimizations that would work on real-world data Avoid • Exploiting the repetitive structure of weight matrices, weight values, and bias values • Exploiting layer independence of results • Using optimizations that would not work on real-world data…”

Section: Input Data Setmentioning

confidence: 99%

“…The Subgraph Isomorphism Graph Challenge [15] and the Stochastic Block Partition Challenge [16] have enabled a new generation of graph analysis systems by highlighting the benefits of novel innovations in these systems. Similarly, the proposed Sparse DNN Challenge seeks to highlight innovations that are applicable to emerging sparse AI and machine learning.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sparse Deep Neural Network Graph Challenge

Kepner

Alford²,

Gadepally

et al. 2019

2019 IEEE High Performance Extreme Computing Conference (HPEC)

Self Cite

View full text Add to dashboard Cite

The MIT/IEEE/Amazon GraphChallenge.org encourages community approaches to developing new solutions for analyzing graphs and sparse data. Sparse AI analytics present unique scalability difficulties. The proposed Sparse Deep Neural Network (DNN) Challenge draws upon prior challenges from machine learning, high performance computing, and visual analytics to create a challenge that is reflective of emerging sparse AI systems. The Sparse DNN Challenge is based on a mathematically well-defined DNN inference computation and can be implemented in any programming environment. Sparse DNN inference is amenable to both vertex-centric implementations and array-based implementations (e.g., using the GraphBLAS.org standard). The computations are simple enough that performance predictions can be made based on simple computing hardware models. The input data sets are derived from the MNIST handwritten letters. The surrounding I/O and verification provide the context for each sparse DNN inference that allows rigorous definition of both the input and the output. Furthermore, since the proposed sparse DNN challenge is scalable in both problem size and hardware, it can be used to measure and quantitatively compare a wide range of present day and future systems. Reference implementations have been implemented and their serial and parallel performance have been measured. Specifications, data, and software are publicly available at GraphChallenge.org.

show abstract

Section: Neural Network Datamentioning

confidence: 99%

Section: Input Data Setmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sparse Deep Neural Network Graph Challenge

Kepner

Alford²,

Gadepally

et al. 2019

2019 IEEE High Performance Extreme Computing Conference (HPEC)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The Graph Challenge consists of three challenges • Pre-challenge: PageRank pipeline [58] • Static graph challenge: subgraph isomorphism [59] • Streaming graph challenge: stochastic block partition [60] The static graph challenge is further broken down into triangle counting and k-truss. Triangle counting received the most submissions and is the focus of this paper.…”

Section: Triangle Countingmentioning

confidence: 99%

GraphChallenge.org: Raising the Bar on Graph Analytic Performance

Samsi¹,

Gadepally²,

Hurley³

et al. 2018

2018 IEEE High Performance Extreme Computing Conference (HPEC)

Self Cite

View full text Add to dashboard Cite

The rise of graph analytic systems has created a need for new ways to measure and compare the capabilities of graph processing systems. The MIT/Amazon/IEEE Graph Challenge has been developed to provide a well-defined community venue for stimulating research and highlighting innovations in graph analysis software, hardware, algorithms, and systems. GraphChallenge.org provides a wide range of preparsed graph data sets, graph generators, mathematically defined graph algorithms, example serial implementations in a variety of languages, and specific metrics for measuring performance. Graph Challenge 2017 received 22 submissions by 111 authors from 36 organizations. The submissions highlighted graph analytic innovations in hardware, software, algorithms, systems, and visualization. These submissions produced many comparable performance measurements that can be used for assessing the current state of the art of the field. There were numerous submissions that implemented the triangle counting challenge and resulted in over 350 distinct measurements. Analysis of these submissions show that their execution time is a strong function of the number of edges in the graph, Ne, and is typically proportional to N 4/3 e for large values of Ne. Combining the model fits of the submissions presents a picture of the current state of the art of graph analysis, which is typically 10 8 edges processed per second for graphs with 10 8 edges. These results are 30 times faster than serial implementations commonly used by many graph analysts and underscore the importance of making these performance benefits available to the broader community. Graph Challenge provides a clear picture of current graph analysis systems and underscores the need for new innovations to achieve high performance on very large graphs.

show abstract

“…[5], [6], traditional graph-based spectral clustering and cuts partitioning, discuss numerical issues of related large scale computations for Big Data spectral clustering, and describe our approach. Partition Challenge Datasets with Known Truth Partitions [7] suitable for our techniques are reviewed in §III. Our numerical results appear in §IV.…”

Section: Introductionmentioning

confidence: 99%

Preconditioned spectral clustering for stochastic block partition streaming graph challenge (Preliminary version at arXiv.)

Zhuzhunashvili

Knyazev

2017

2017 IEEE High Performance Extreme Computing Conference (HPEC)

View full text Add to dashboard Cite

Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) is demonstrated to efficiently solve eigenvalue problems for graph Laplacians that appear in spectral clustering. For static graph partitioning, 10-20 iterations of LOBPCG without preconditioning result in˜10x error reduction, enough to achieve 100% correctness for all Challenge datasets with known truth partitions, e.g., for graphs with 5K/.1M (50K/1M) Vertices/Edges in 2 (7) seconds, compared to over 5,000 (30,000) seconds needed by the baseline Python code. Our Python code 100% correctly determines 98 (160) clusters from the Challenge static graphs with 0.5M (2M) vertices in 270 (1,700) seconds using 10GB (50GB) of memory. Our single-precision MATLAB code calculates the same clusters at half time and memory. For streaming graph partitioning, LOBPCG is initiated with approximate eigenvectors of the graph Laplacian already computed for the previous graph, in many cases reducing 2-3 times the number of required LOBPCG iterations, compared to the static case. Our spectral clustering is generic, i.e. assuming nothing specific of the block model or streaming, used to generate the graphs for the Challenge, in contrast to the base code. Nevertheless, in 10-stage streaming comparison with the base code for the 5K graph, the quality of our clusters is similar or better starting at stage 4 (7) for emerging edging (snowballing) streaming, while the computations are over 100-1000 faster.

show abstract

Streaming graph challenge: Stochastic block partition

Cited by 62 publications

References 24 publications

Sparse Deep Neural Network Graph Challenge

Sparse Deep Neural Network Graph Challenge

GraphChallenge.org: Raising the Bar on Graph Analytic Performance

Preconditioned spectral clustering for stochastic block partition streaming graph challenge (Preliminary version at arXiv.)

Contact Info

Product

Resources

About