An effective framework for asynchronous incremental graph processing

Although many graph processing systems have been proposed, graphs in the real-world are often dynamic. It is important to keep the results of graph computation up-todate. Incremental computation is demonstrated to be an efficient solution to update calculated results. Recently, many incremental graph processing systems have been proposed to handle dynamic graphs in an asynchronous way and are able to achieve better performance than those processed in a synchronous way. However, these solutions still suffer from suboptimal convergence speed due to their slow propagation of important vertex state (important to convergence speed) and poor locality. In order to solve these problems, we propose a novel graph processing framework. It introduces a dynamic partition method to gather the important vertices for high locality, and then uses a priority-based scheduling algorithm to assign them with a higher priority for an effective processing order. By such means, it is able to reduce the number of updates and increase the locality, thereby reducing the convergence time. Experimental results show that our method reduces the number of updates by 30%, and reduces the total execution time by 35%, compared with state-of-the-art systems.     

图计算加速架构综述

在大数据时代,图被用于各种领域表示具有复杂联系的数据.图计算应用被广泛用于各种领域,以挖掘图数据中潜在的价值.图计算应用特有的不规则执行行为,引发了不规则负载、密集读改写更新操作、不规则访存和不规则通信等挑战.现有通用架构无法有效地应对上述挑战.为了克服加速图计算应用面临的挑战,大量的图计算硬件加速架构设计被提出.它们...     

A disk I/O optimized system for concurrent graph processing jobs

In order to analyze and process the large graphs with high cost efficiency, researchers have developed a number of out-of-core graph processing systems in recent years based on just one commodity computer. On the other hand, with the rapidly growing need of analyzing graphs in the real-world, graph processing systems have to efficiently handle massive concurrent graph processing (CGP) jobs. Unfortunately, due to the inherent design for single graph processing job, existing out-of-core graph processing systems usually incur unnecessary data accesses and severe competition of I/O bandwidth when handling the CGP jobs. In this paper, we propose GraphCP, a disk I/O optimized out-of-core graph processing system that efficiently supports the processing of CGP jobs. GraphCP proposes a benefit-aware sharing execution model to share the I/O access and processing of graph data among the CGP jobs and adaptively schedule the graph data loading based on the states of vertices, which efficiently overcomes above challenges faced by existing out-of-core graph processing systems. Moreover, GraphCP adopts a dependency-based future-vertex updating model so as to reduce disk I/Os in the future iterations. In addition, GraphCP organizes the graph data with a Source-Sorted Sub-Block graph representation for better processing capacity and I/O access locality. Extensive evaluation results show that GraphCP is 20.5× and 8.9× faster than two out-of-core graph processing systems GridGraph and GraphZ, and 3.5× and 1.7× faster than two state-of-art concurrent graph processing systems Seraph and GraphSO.     

Real-time continuous intersection joins over large sets of moving objects using graphic processing units

The Multiple Time Bucket Join (MTB-join) algorithm is the state of the art for processing the continuous intersection join (CI-join) query over moving objects. It considerably outperforms alternatives, but still falls short of real-time application performance requirements for large sets of moving objects. In this paper, we achieve real-time performance for the CI-join query over large sets of moving objects by exploiting the computational power of commodity graphics processing units (GPUs). We first analyze how the main characteristics of the MTB-join algorithm make it ill suited to GPUs and identify key challenges in designing efficient GPU-based algorithms for the query. We then address these challenges by developing the multi-layered grid join (MLG-join) algorithm which has the following key features: (i) memory locality friendly indexing, (ii) no dynamic memory allocation, (iii) in-place object updates, (iv) lock-free concurrent updates, and (v) massive parallelism. These features unleash the full potential of the memory bandwidth and parallel processing of GPUs. Furthermore, we conduct a theoretical analysis which can predict the pruning power of the MLG-join algorithm given certain parameter values used in the algorithm. This allows us to select optimal parameter values. Through extensive experimental results, we show that our analysis accurately models the MLG-join algorithm’s sensitivity to parameter values. The proposed MLG-join algorithm outperforms the MTB-join algorithm, and a GPU-based nested-loops join algorithm, by up to two orders of magnitude, and achieves real-time performance for CI-join queries on large sets of moving objects.     

A survey of graph processing on graphics processing units

Graphics processing units (GPUs) have become popular high-performance computing platforms for a wide range of applications. The trend of processing graph structures on modern GPUs has also attracted an increasing interest in recent years. This article aims to review research works on adapting the massively parallel architecture of GPUs to accelerate the performance of fundamental graph operations. Despite their merits, some factors such as the unique architecture of GPUs, limited programming models, and irregular structures of graphs prevent GPU implementations from achieving high performance. Thus, this survey also discusses challenges and optimization techniques used by recent studies to fully utilize the GPU capability. A categorization of the existing research works is also presented based on the specific issues these attempted to solve.     

iGraph: an incremental data processing system for dynamic graph

With the popularity of social network, the demand for real-time processing of graph data is increasing. However, most of the existing graph systems adopt a batch processing mode, therefore the overhead of maintaining and processing of dynamic graph is significantly high. In this paper, we design iGraph, an incremental graph processing system for dynamic graph with its continuous updates. The contributions of iGraph include: 1) a hash-based graph partition strategy to enable fine-grained graph updates; 2) a vertexbased graph computing model to support incremental data processing; 3) detection and rebalance methods of hotspot to address the workload imbalance problem during incremental processing. Through the general-purpose API, iGraph can be used to implement various graph processing algorithms such as PageRank. We have implemented iGraph on Apache Spark, and experimental results show that for real life datasets, iGraph outperforms the original GraphX in respect of graph update and graph computation.     

Efficient computation of distance labeling for decremental updates in large dynamic graphs

Since today’s real-world graphs, such as social network graphs, are evolving all the time, it is of great importance to perform graph computations and analysis in these dynamic graphs. Due to the fact that many applications such as social network link analysis with the existence of inactive users need to handle failed links or nodes, decremental computation and maintenance for graphs is considered a challenging problem. Shortest path computation is one of the most fundamental operations for managing and analyzing large graphs. A number of indexing methods have been proposed to answer distance queries in static graphs. Unfortunately, there is little work on answering such queries for dynamic graphs. In this paper, we focus on the problem of computing the shortest path distance in dynamic graphs, particularly on decremental updates (i.e., edge deletions). We propose maintenance algorithms based on distance labeling, which can handle decremental updates efficiently. By exploiting properties of distance labeling in original graphs, we are able to efficiently maintain distance labeling for new graphs. We experimentally evaluate our algorithms using eleven real-world large graphs and confirm the effectiveness and efficiency of our approach. More specifically, our method can speed up index re-computation by up to an order of magnitude compared with the state-of-the-art method, Pruned Landmark Labeling (PLL).     

KGDB: Knowledge Graph Database System with Unified Model and Query Language

Knowledge graph is an important cornerstone of artificial intelligence, which currently has two main data models: RDF graphs and property graphs. There are several query languages on these two data models, including SPARQL on RDF graphs and Cypher on property graphs. Over the last decade, various communities have developed different data management methods for RDF graphs and property graphs. Inconsistent data models and query languages hinder the wider application of knowledge graphs. In this paper, we propose a knowledge graphy database (KGDB) system with unified data model and query language. (1) We work out a unified storage scheme based on the relational model that supports the efficient storage of RDF graphs and property graphs, catering to the smooth storage and query of knowledge graph data. (2) The characteristic set-based clustering is used in KGDB for the storage of typeless entities. (3) It realizes the interoperability of SPARQL and Cypher by enabling them to operate on the same knowledge graph. Extensive experiments on real-world datasets and synthetic datasets reveal that KGDB is more efficient than existing knowledge graph database management systems in storage management and query efficiency. KGDB saves 30% of the storage space on average compared with gStore and Neo4j. In addition, KDGB is two orders of magnitude faster than gStore and Neo4j in the query of the real-world datasets, seen from experiments on the query of basic graph pattern matching.     

Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces

Graphs are a flexible and general formalism providing rich models in various important domains, such as distributed computing, intelligent tutoring systems or social network analysis. In many cases, such models need to take changes in the graph structure into account, that is, changes in the number of nodes or in the graph connectivity. Predicting such changes within graphs can be expected to yield important insight with respect to the underlying dynamics, e.g. with respect to user behaviour. However, predictive techniques in the past have almost exclusively focused on single edges or nodes. In this contribution, we attempt to predict the future state of a graph as a whole. We propose to phrase time series prediction as a regression problem and apply dissimilarity- or kernel-based regression techniques, such as 1-nearest neighbor, kernel regression and Gaussian process regression, which can be applied to graphs via graph kernels. The output of the regression is a point embedded in a pseudo-Euclidean space, which can be analyzed using subsequent dissimilarity- or kernel-based processing methods. We discuss strategies to speed up Gaussian processes regression from cubic to linear time and evaluate our approach on two well-established theoretical models of graph evolution as well as two real data sets from the domain of intelligent tutoring systems. We find that simple regression methods, such as kernel regression, are sufficient to capture the dynamics in the theoretical models, but that Gaussian process regression significantly improves the prediction error for real-world data.     

A Taxonomy and Survey of Dynamic Graph Visualization

Dynamic graph visualization focuses on the challenge of representing the evolution of relationships between entities in readable, scalable and effective diagrams. This work surveys the growing number of approaches in this discipline. We derive a hierarchical taxonomy of techniques by systematically categorizing and tagging publications. While static graph visualizations are often divided into node‐link and matrix representations, we identify the representation of time as the major distinguishing feature for dynamic graph visualizations: either graphs are represented as animated diagrams or as static charts based on a timeline. Evaluations of animated approaches focus on dynamic stability for preserving the viewer's mental map or, in general, compare animated diagrams to timeline‐based ones. A bibliographic analysis provides insights into the organization and development of the field and its community. Finally, we identify and discuss challenges for future research. We also provide feedback from experts, collected with a questionnaire, which gives a broad perspective of these challenges and the current state of the field.     

An asynchronous traversal engine for graph-based rich metadata management

Rich metadata in high-performance computing (HPC) systems contains extended information about users, jobs, data files, and their relationships. Property graphs are a promising data model to represent heterogeneous rich metadata flexibly. Specifically, a property graph can use vertices to represent different entities and edges to record the relationships between vertices with unique annotations. The high-volume HPC use case, with millions of entities and relationships, naturally requires an out-of-core distributed property graph database, which must support live updates (to ingest production information in real time), low-latency point queries (for frequent metadata operations such as permission checking), and large-scale traversals (for provenance data mining).Among these needs, large-scale property graph traversals are particularly challenging for distributed graph storage systems. Most existing graph systems implement a “level-synchronous” breadth-first search algorithm that relies on global synchronization in each traversal step. This performs well in many problem domains; but a rich metadata management system is characterized by imbalanced graphs, long traversal lengths, and concurrent workloads, each of which has the potential to introduce or exacerbate stragglers (i.e., abnormally slow steps or servers in a graph traversal) that lead to low overall throughput for synchronous traversal algorithms. Previous research indicated that the straggler problem can be mitigated by using asynchronous traversal algorithms, and many graph-processing frameworks have successfully demonstrated this approach. Such systems require the graph to be loaded into a separate batch-processing framework instead of being iteratively accessed, however.In this work, we investigate a general asynchronous graph traversal engine that can operate atop a rich metadata graph in its native format. We outline a traversal-aware query language and key optimizations (traversal-affiliate caching and execution merging) necessary for efficient performance. We further explore the effect of different graph partitioning strategies on the traversal performance for both synchronous and asynchronous traversal engines. Our experiments show that the asynchronous graph traversal engine is more efficient than its synchronous counterpart in the case of HPC rich metadata processing, where more servers are involved and larger traversals are needed. Moreover, the asynchronous traversal engine is more adaptive to different graph partitioning strategies.     

Implementing QR factorization updating algorithms on GPUs

Linear least squares problems are commonly solved by QR factorization. When multiple solutions need to be computed with only minor changes in the underlying data, knowledge of the difference between the old data set and the new can be used to update an existing factorization at reduced computational cost. We investigate the viability of implementing QR updating algorithms on GPUs and demonstrate that GPU-based updating for removing columns achieves speed-ups of up to 13.5× compared with full GPU QR factorization. We characterize the conditions under which other types of updates also achieve speed-ups.     

知识图谱划分算法研究综述

知识图谱是人工智能的重要基石,因其包含丰富的图结构和属性信息而受到广泛关注.知识图谱可以精确语义描述现实世界中的各种实体及其联系,其中顶点表示实体,边表示实体间的联系.知识图谱划分是大规模知识图谱分布式处理的首要工作,对知识图谱分布式存储、查询、推理和挖掘起基础支撑作用.随着知识图谱数据规模及分布式处理需求的不断增长,...     

Exploiting graphical processing units for data‐parallel scientific applications

Graphical processing units (GPUs) have recently attracted attention for scientific applications such as particle simulations. This is partially driven by low commodity pricing of GPUs but also by recent toolkit and library developments that make them more accessible to scientific programmers. We discuss the application of GPU programming to two significantly different paradigms—regular mesh field equations with unusual boundary conditions and graph analysis algorithms. The differing optimization techniques required for these two paradigms cover many of the challenges faced when developing GPU applications. We discuss the relevance of these application paradigms to simulation engines and games. GPUs were aimed primarily at the accelerated graphics market but since this is often closely coupled to advanced game products it is interesting to speculate about the future of fully integrated accelerator hardware for both visualization and simulation combined. As well as reporting the speed‐up performance on selected simulation paradigms, we discuss suitable data‐parallel algorithms and present code examples for exploiting GPU features like large numbers of threads and localized texture memory. We find a surprising variation in the performance that can be achieved on GPUs for our applications and discuss how these findings relate to past known effects in parallel computing such as memory speed‐related super‐linear speed up. Copyright © 2009 John Wiley & Sons, Ltd.     

图数据分析系统计算模型综述

随着大数据时代的到来,图作为一种表示和分析大数据的有效方法,正成为学术界和工业界广泛关注的焦点.图数据具有规模巨大、耦合性强、动态变化等特点,需要新的计算模型支持高效的图计算系统对大规模图数据进行处理.对图计算系统计算模型的研究现状进行了调研和综述,首先介绍图计算系统的产生和发展,然后将主流图计算系统中的计算模型按照计算对象分为:结点中心计算模型、边中心计算模型、路径中心计算模型和子图计算模型四类,重点介绍结点中心模型的应用和性能,最后对图计算模型的发展过程进行总结,并展望图计算模型未来发展方向.     

支持深度学习的视觉数据库管理系统研究进展

计算机视觉因其强大的学习能力,在各种真实场景中得到了广泛应用.随着数据库的发展,利用数据库中成熟的数据管理技术来处理视觉分析应用,已成为一种日益增长的研究趋势.图像、视频和文本等多模态数据的相互融合处理,也促进了视觉分析应用的多样性和准确性.近年来,因深度学习的兴起,支持深度学习的视觉分析应用开始受到广泛关注.然而,传统的数据库管理技术在深度学习场景下面临着复杂视觉分析语义难以表达、应用执行效率低等问题.因此,支持深度学习的视觉数据库管理系统得到了广泛关注.综述了目前视觉数据库管理系统的研究进展:首先,总结了视觉数据库管理系统在不同层面上面临的挑战,包括编程接口、查询优化、执行调度和数据存储;其次,分别探讨了上述4个层面上的相关技术;最后,对视觉数据库管理系统未来的研究方向进行了展望.     

基于深度学习的属性图异常检测综述

异常检测一直以来都是数据挖掘领域的研究热点之一,其任务是在海量数据中识别罕见的观测对象。随着图数据挖掘的发展,属性图异常检测在各个领域广受关注。然而,属性图因其复杂的拓扑结构和丰富的属性信息成为异常检测一大难点。深度学习方法在捕捉属性图复杂的信息中展现出优越性能,已被证实是解决属性图异常检测问题非常有效的方法。对普通图异常检测和属性图异常检测以及表示学习相关方法进行简要概述;其次从静态属性图和动态属性图两方面对最新深度学习异常检测方法进行介绍与分类;对常见数据集上的实验结果进行了对比、分析;对属性图异常检测的应用场景、存在的问题以及面临的挑战进行讨论,展望了未来的研究方向。     

Achieving Optimal Inter-Node Communication in Graph Partitioning Using Random Selection and Breadth-First Search

Processing large graph datasets represents an increasingly important area in computing research and applications. The size of many graph datasets has increased well beyond the processing capacity of a single computing node, thereby necessitating distributed approaches. As these datasets are processed over a distributed system of nodes, this leads to an inter-node communication cost problem that negatively affects system performance. Previously proposed algorithms implemented breadth-first search (BFS) for graph searching and focused on the execution, parallel performance and not the communication. In this paper a new methodology is proposed that combines BFS with random selection in order to partition large graph datasets and effectively minimize inter-node communication. The new method is discussed and applied to the single-source shortest path and PageRank algorithms using three graphs that are representative of real-world scenarios. Experimental results show that graph inter-node communication for canonical graphs representative of real-world data is improved up to 42 % in case of Powerlaw graph, up to 27 % in case of Random near K-regular graph (with low degree), and up to 7 % in case of Random near K-regular graph (with high degree).     

Scalable graph-based OLAP analytics over process execution data

In today’s knowledge-, service-, and cloud-based economy, businesses accumulate massive amounts of data from a variety of sources. In order to understand businesses one may need to perform considerable analytics over large hybrid collections of heterogeneous and partially unstructured data that is captured related to the process execution. This data, usually modeled as graphs, increasingly come to show all the typical properties of big data: wide physical distribution, diversity of formats, non-standard data models, independently-managed and heterogeneous semantics. We use the term big process graph to refer to such large hybrid collections of heterogeneous and partially unstructured process related execution data. Online analytical processing (OLAP) of big process graph is challenging as the extension of existing OLAP techniques to analysis of graphs is not straightforward. Moreover, process data analysis methods should be capable of processing and querying large amount of data effectively and efficiently, and therefore have to be able to scale well with the infrastructure’s scale. While traditional analytics solutions (relational DBs, data warehouses and OLAP), do a great job in collecting data and providing answers on known questions, key business insights remain hidden in the interactions among objects: it will be hard to discover concept hierarchies for entities based on both data objects and their interactions in process graphs. In this paper, we introduce a framework and a set of methods to support scalable graph-based OLAP analytics over process execution data. The goal is to facilitate the analytics over big process graph through summarizing the process graph and providing multiple views at different granularity. To achieve this goal, we present a model for process OLAP (P-OLAP) and define OLAP specific abstractions in process context such as process cubes, dimensions, and cells. We present a MapReduce-based graph processing engine, to support big data analytics over process graphs. We have implemented the P-OLAP framework and integrated it into our existing process data analytics platform, ProcessAtlas, which introduces a scalable architecture for querying, exploration and analysis of large process data. We report on experiments performed on both synthetic and real-world datasets that show the viability and efficiency of the approach.     

基于属性嵌入与图注意力网络的实体对齐算法

实体对齐旨在找到位于不同知识图谱中的等效实体,是实现知识融合的重要步骤.当前主流的方法是基于图神经网络的实体对齐方法,这些方法往往过于依赖图的结构信息,导致在特定图结构上训练得到的模型不能拓展应用于其他图结构中.同时,大多数方法未能充分利用辅助信息,例如属性信息.为此,本文提出了一种基于图注意力网络和属性嵌入的实体对齐方法,该方法使用图注意力网络对不同的知识图谱进行编码,引入注意力机制从实体应用到属性,在对齐阶段将结构嵌入和属性嵌入进行结合实现实体对齐效果的提升.在现实世界的3个真实数据集上对本文模型进行了验证,实验结果表明提出的方法在很大程度上优于基准的实体对齐方法.