Parallel graph processing on graphics processors made easy

Hybrid memory systems composed of DRAM and Non-Volatile Memory (NVM) can offer very large memory capacity for data-intensive applications such as graph processing. The performance of parallel graph processing is often affected by straggler tasks due to asymmetrical graph partitioning and sub-graph processing. In hybrid memory systems, the significant difference of performance between DRAM and NVM may exacerbate the load imbalance of parallel graph processing. Traditional load balancing schemes that only aim to balance computing loads among multiple processors are no longer effective in hybrid memory systems.In this paper, we first explore how hybrid memory systems affect the efficiency of existing graph processing systems. We make two observations: 1) Traditional work-stealing schemes may be not efficient enough in hybrid memory systems. Data migration combining with work-stealing can further improve the efficiency of parallel graph processing; 2) Interleaving accesses to different graph partitions due to data dependency have a significant impact on the effectiveness of data migration. To address these problems, we propose NVMGraph, a parallel graph processing scheme to mitigate memory access imbalance among partitions in hybrid memory systems. We first recognize data dependency among all partitions according to graph algorithms and then coalesce the randomly-accessed data required by a working thread in a single partition to mitigate data dependency. To reduce migration cost while still speeding up the straggler task, we only migrate randomly-accessed data blocks in the most frequently-accessed partition from NVM to DRAM. We implement NVMGraph based on Ligra and evaluate it in a hybrid memory system using real NVM devices. Experimental results demonstrate that NVMGraph can improve the performance of graph processing by up to 40.3% compared with the state-of-the-art Ligra and Polymer.

Graphs are a natural fit for modeling concepts used in solving diverse problems in science, commerce, engineering, and governance. Responding to the variety of graph data and algorithms, many parallel and distributed graph processing systems exist. However, until now these platforms use a static model of deployment: they only run on a pre-defined set of machines. This raises many conceptual and pragmatic issues, including misfit with the highly dynamic nature of graph processing, and could lead to resource waste and high operational costs. In contrast, in this work we explore a dynamic model of deployment. We first characterize workload dynamicity, beyond mere active-vertex variability. Then, to conduct an in-depth elasticity study of distributed graph processing, we build a prototype, JoyGraph, which is the first such system that implements complex, policy-based, and fine-grained elasticity. Using the state-of-the-art LDBC Graphalytics benchmark and the SPEC Cloud Group's elasticity metrics, we show the benefits of elasticity in graph processing: (i) improved resource utilization, (ii) reduced operational costs, and (iii) aligned operation-workload dynamicity. Furthermore, we explore the cost of elasticity in graph processing. We identify a key drawback: although elasticity does not degrade application throughput, graph-processing workloads are sensitive to data movement while leasing or releasing resources.

In general, the performance of parallel graph processing is determined by three pairs of critical parameters, namely synchronous or asynchronous execution mode (Sync or Async), Push or Pull communication mechanism (Push or Pull), and Data-driven or Topology-driven traversing scheme (DD or TD), which increases the complexity and sophistication of programming and system implementation of GPU. Existing graph-processing frameworks mainly use a single combination in the entire execution for a given application, but we have observed their variable and suboptimal performance. In this paper, we present SEP-Graph, a highly efficient software framework for graph-processing on GPU. The hybrid execution mode is automatically switched among three pairs of parameters, with an objective to achieve the shortest execution time in each iteration. We also apply a set of optimizations to SEP-Graph, considering the characteristics of graph algorithms and underlying GPU architectures. We show the effectiveness of SEP-Graph based on our intensive and comparative performance evaluation on NVIDIA 1080, P100, and V100 GPUs. Compared with existing and representative GPU graph-processing framework Groute and Gunrock, SEP-Graph can reduce execution time up to 45.8 times and 39.4 times.

There is an increasing interest in executing complex analyses over large graphs, many of which require processing a large number of multi-hop neighborhoods or subgraphs. Examples include ego network analysis, motif counting, finding social circles, personalized recommendations, link prediction, anomaly detection, analyzing influence cascades, and others. These tasks are not well served by existing vertex-centric graph processing frameworks, where user programs are only able to directly access the state of a single vertex at a time, resulting in high communication, scheduling, and memory overheads in executing such tasks. Further, most existing graph processing frameworks ignore the challenges in extracting the relevant portions of the graph that an analysis task is interested in, and loading those onto distributed memory. This paper introduces NScale, a novel end-to-end graph processing framework that enables the distributed execution of complex subgraph-centric analytics over large-scale graphs in the cloud. NScale enables users to write programs at the level of subgraphs rather than at the level of vertices. Unlike most previous graph processing frameworks, which apply the user program to the entire graph, NScale allows users to declaratively specify subgraphs of interest. Our framework includes a novel graph extraction and packing (GEP) module that utilizes a cost-based optimizer to partition and pack the subgraphs of interest into memory on as few machines as possible. The distributed execution engine then takes over and runs the user program in parallel on those subgraphs, restricting the scope of the execution appropriately, and utilizes novel techniques to minimize memory consumption by exploiting overlaps among the subgraphs. We present a comprehensive empirical evaluation comparing against three state-of-the-art systems, namely Giraph, GraphLab, and GraphX, on several real-world datasets and a variety of analysis tasks. Our experimental results show orders-of-magnitude improvements in performance and drastic reductions in the cost of analytics compared to vertex-centric approaches.

Network Intrusion Detection Systems (NIDSs) are an increasingly important tool for the prevention and mitigation of cyber attacks. Over the past years, a lot of research efforts have aimed at leveraging the increasingly powerful models of Machine Learning (ML) for this purpose. A number of labelled synthetic datasets have been generated and made publicly available by researchers, and they have become the benchmarks via which new ML-based NIDS classifiers are being evaluated. Recently published results show excellent classification performance with these datasets, increasingly approaching 100 percent performance across key evaluation metrics such as Accuracy, F1 score, AUC, etc. Unfortunately, we have not yet seen these excellent academic research results translated into practical NIDS systems with such near-perfect performance. This motivated our research presented in this paper, where we analyse the statistical properties of the benign traffic in three of the more recent and relevant NIDS datasets, (CIC_IDS, UNSW_NB15, TON_IOT), by converting them into a common flow format. As a comparison, we consider two datasets obtained from real-world production networks, one from a university network and one from a medium size Internet Service Provider (ISP). Our results show that the two real-world datasets are quite similar among themselves in regards to most of the considered statistical features. Equally, the three synthetic datasets are also relatively similar within their group. However, and most importantly, our results show a distinct difference of most of the considered statistical features between the three synthetic datasets and the two real-world datasets. Since ML relies on the basic assumption of training and test datasets being sampled from the same distribution, this raises the question of how well the performance results of ML-classifiers trained on the considered synthetic datasets can translate and generalise to real-world networks. We believe this is an interesting and relevant question which provides motivation for further research in this space.

Graph processing workloads are memory intensive with irregular access patterns and large memory footprint resulting in low data locality. Their popular software implementations typically employ either Push or Pull style propagation of changes through the graph over multiple iterations that follow the Bulk Synchronous Model. The performance of these algorithms on traditional computing systems is limited by random reads/writes of vertex values, synchronization overheads, and additional overheads for tracking active sets of vertices or edges across iterations. In this paper, we present GraphPulse, a hardware framework for asynchronous graph processing with event-driven scheduling that overcomes the performance limitations of software frameworks. Event-driven computation model enables a parallel dataflow-style execution where atomic updates and active sets tracking are inherent to the model; thus, scheduling complexity is reduced and scalability is enhanced. The dataflow nature of the architecture also reduces random reads of vertex values by carrying the values in the events themselves. We capitalize on the update properties commonly present in graph algorithms to coalesce in-flight events and substantially reduce the event storage requirement and the processing overheads incurred. GraphPulse event-model naturally supports asynchronous graph processing, enabling substantially faster convergence by exploiting available parallelism, reducing work, and eliminating synchronization at iteration boundaries. The framework provides easy to use programming interface for faster development of hardware graph accelerators. A single GraphPulse accelerator achieves up to 74x speedup (28x on average) over Ligra, a state of the art software framework, running on a 12 core CPU. It also achieves an average of 6.2x speedup over Graphicionado, a state of the art graph processing accelerator.

In many global businesses of multinational enterprises, graph-structure data is usually geographically distributed in different regions to support low-latency services. Geo-distributed graph processing suffers from the Wide Area Networks (WANs) with scarce and heterogeneous bandwidth, thus essentially differs from traditional distributed graph processing. In this paper, we propose RAGraph, aRegion-Aware framework for geo-distributed graph processing.At the core of RAGraph, we design a region-aware graph processing framework that allows advancing inefficient global updates locally and enables sensible coordination-free message interactions. RAGraph also contains an adaptive hierarchical message interaction engine to switch interaction modes adaptively based on network heterogeneity and fluctuation, and a discrepancy-aware message filtering strategy to filter important messages. Finally, the experiments show that RAGraph can achieve 1.69X - 40.53X speedup and 20.9% - 97% WAN cost reduction compared with state-of-the-art systems.

With continuously growing data, clusters also need to grow periodically to accommodate the increased demand of data processing. This is usually done by addition of newer hardware, whose configuration might differ from the existing nodes. As a result, clusters are becoming heterogeneous in nature. For many real world machine learning and data mining applications, data is represented in the form of graphs. Most of the existing distributed graph processing frameworks such as Pregel and Graphlab assume that the computational nodes are homogeneous. These frameworks split the graph into approximately equal subgraphs, which is appropriate for homogeneous clusters. In heterogeneous clusters, these frameworks perform poorly in most of the scenarios. To the best of our knowledge, GraphIVE is the only heterogeneity-aware graph processing framework. It learns the relative capabilities of the nodes based on runtime metrics of previous jobs and partitions the graph proportionally. However, it may not perform well if a new job differs drastically in terms of resource requirements when compared to previous jobs executed on the cluster. To overcome this limitation, we propose GraphSteal, a dynamic graph re-partitioning policy for vertex-cut based graph processing frameworks on heterogeneous clusters. GraphSteal dynamically re-partitions the graph based on the runtime characteristics of the job. To avoid computational skew in the cluster, it migrates edges from slow nodes to fast nodes. To demonstrate our approach, we modify the source code of Graphlab to incorporate dynamic graph re-partitioning strategy. Experimental results show that GraphSteal significantly improves the performance over Graphlab.

The importance of high-performance graph processing to solve big data problems targeting high-impact applications is greater than ever before. Recent graph processing frameworks target different hardware platforms (e.g., shared memory systems, accelerators such as GPUs, and distributed systems) and differ with respect to the programming model they adopt (e.g., based on linear algebra formulations of graph algorithms or enabling direct access to the graph structure). To better understand the impact of these choices, this paper, presents a comparative study of five state-of-the-art graph processing frameworks: two CPU-only frameworks - GraphMat and Galois, two GPU-based frameworks - Nvgraph and Gunrock; and Totem, a hybrid (CPU+GPU) framework. We use three popular graph algorithms (PageRank, Single Source Shortest Path, and Breadth-First Search), and massive scale graphs with up to billions of edges. Our evaluation focuses on three performance metrics: (i) execution time, (ii) scalability and (iii) energy consumption.

Network signature matching is an important task in many applications such as network security or traffic analysis, which generally rely on a flexible signature matching system to extract important packet information from each processed packet. This task is computation and data intensive, and requires significant processing time in sequential manner. In order to accelerate signature matching of giga-bit network traffic, we aim to exploit the inherent parallelism of signature matching through the use of parallel graphics processor units. In this paper, we present detailed analysis of signature matching along with the system design for parallel graphics processors(GPUs). The signature matching schema proposed is based on port matching and keyword matching in each packet header. A real system on graphics processor units was implemented to evaluate the efficacy of our design. Experimental results proved that signature matching can be efficiently done on graphics processor units.

Parallel processing of large graphs

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.

Parallel graph algorithms have become one of the principal applications of high-performance computing besides numerical simulations and machine learning workloads. However, due to their highly unstructured nature, graph algorithms remain extremely challenging for most parallel systems, with large gaps between observed performance and theoretical limits. Further-more, most mainstream architectures rely heavily on single instruction multiple data (SIMD) processing for high floating-point rates, which is not beneficial for graph processing which instead requires high memory bandwidth, low memory latency, and efficient processing of unstructured data. On the other hand, we are currently observing an explosion of new hardware architectures, many of which are adapted to specific purposes and diverge from traditional designs. A notable example is the Graphcore Intelligence Processing Unit (IPU), which is developed to meet the needs of upcoming machine intelligence applications. Its design eschews the traditional cache hierarchy, relying on SRAM as its main memory instead. The result is an extremely high-bandwidth, low-latency memory at the cost of capacity. In addition, the IPU consists of a large number of independent cores, allowing for true multiple instruction multiple data (MIMD) processing. Together, these features suggest that such a processor is well suited for graph processing. We test the limits of graph processing on multiple IPUs by implementing a low-level, high-performance code for breadth-first search (BFS), following the specifications of Graph500, the most widely used benchmark for parallel graph processing. Despite the simplicity of the BFS algorithm, implementing efficient parallel codes for it has proven to be a challenging task in the past. We show that our implementation scales well on a system with 8 IPUs and attains roughly twice the performance of an equal number of NVIDIA V100 GPUs using state-of-the-art CUDA code.

Implementation of parallel processing in computational fluid dynamics (CFD) is shortening the time required to design products and systems, and is bringing once-elusive problems under a new measure of control. Parallel processing, which is making ever larger models practical, is based on an idea called domain decomposition. The software divides the flow domain into segments of roughly equal computational work. The technology is influencing automotive and aerospace testing by getting more and more work out of the wind tunnel and onto a computer, where it costs less to perform. DaimlerChrysler in Stuttgart, Germany, explored the aerodynamics of its Mercedes E-class sedan with a 10 million cell model. Besides saving time and money by parallel processing, CFD simulation permits review of arbitrary locations in a flow field, including particle tracking, and can rapidly highlight areas of concern about heat build-up. There are several factors that can inhibit a high degree of parallel scaling. Domain decomposition algorithms affect the load balancing between processors. Imbalances slow the system because the solution isn't in until the last processor is finished

A stochastic programming approach to surgery scheduling under parallel processing principle