ABSTRACT Although energy has become a major concern in data processing systems, it is usually hard to get a deep understanding of how performance and energy consumption relate to each other when planning how to configure a computing environment to execute a specific data‐oriented workload. In this paper, we propose a multi‐layered methodology to analyze the energy consumption of big data workloads executed using Apache Spark in virtualized cloud environments. The approach is structured into three layers: Resource provisioning, system‐level resource utilization, and application‐level resource utilization. Using direct energy measurements using a Power Distribution Unit (PDU) and detailed system monitoring, the study investigates how infrastructure choices and workload characteristics influence energy consumption. Results show that optimal virtual machine configurations depend on workload type and input size; while provisioning decisions affect energy consumption, system‐level metrics such as CPU utilization and disk I/O offer a deeper understanding of the final performance versus energy consumption results. By applying our methodology, our results reveal the impact of task distribution and resource under‐utilization on overall energy efficiency. The findings demonstrate that energy optimization in big data environments requires a comprehensive understanding of factors across infrastructure, system, and application layers. The proposed methodology serves as a practical guide for energy‐aware design and decision‐making in cloud‐based data processing systems.

Graph Pattern Mining (GPM) refers to a class of problems involving the processing of sub graphs extracted from larger graphs. Applications to GPM algorithms include querying subgraphs, identifying motif structures in biological networks, characterizing social media, among others. G PM algorithms are challenging to develop due to subroutines that include non-trivial graph theory concepts and methods such as isomorphism. General-purpose GPM systems have emerged as a solution to improve the user experience with such algorithms. However, existing general-purpose GPM systems are heterogeneous in terms of implementation details, hardware environment and algorithmic paradigms for sub graph exploration and thus, observations taken from the experimental results alone may not clearly identify when a particular paradigm prevails over another. In this work we present an experimentation analysis of popular paradigms used in existing GPM systems. In order to provide a fair and comprehensive evaluation of various algorithmic paradigms we implement all of them within a single GPM framework. Our results show that no single paradigm is best for every application scenario, and we believe that our findings may guide practitioner towards more optimized GPM systems in the future.

Graph Pattern Mining (GPM) is an important, rapidly evolving, and computation demanding area. GPM computation relies on subgraph enumeration, which consists in extracting subgraphs that match a given property from an input graph. Graphics Processing Units (GPUs) have been an effective platform to accelerate applications in many areas. However, the irregularity of subgraph enumeration makes it challenging for efficient execution on GPU due to typical uncoalesced memory access, divergence, and load imbalance. Unfortunately, these aspects have not been fully addressed in previous work. Thus, this work proposes novel strategies to design and implement subgraph enumeration efficiently on GPU. We support a depth-first search style search (DFS-wide) that maximizes memory performance while providing enough parallelism to be exploited by the GPU, along with a warp-centric design that minimizes execution divergence and improves utilization of the computing capabilities. We also propose a low-cost load balancing layer to avoid idleness and redistribute work among thread warps in a GPU. Our strategies have been deployed in a system named DuMato, which provides a simple programming interface to allow efficient implementation of GPM algorithms. Our evaluation has shown that DuMato is often an order of magnitude faster than state-of-the-art GPM systems and can mine larger subgraphs (up to 12 vertices).

In this paper we propose Fractal, a high performance and high productivity system for supporting distributed graph pattern mining (GPM) applications. Fractal employs a dynamic (auto-tuned) load-balancing based on a hierarchical and locality-aware work stealing mechanism, allowing the system to adapt to different workload characteristics. Additionally, Fractal enumerates subgraphs by combining a depth-first strategy with a from scratch processing paradigm to avoid storing large amounts of intermediate state and, thus, improves memory efficiency. Regarding programmer productivity, Fractal presents an intuitive, expressive and modular API, allowing for rapid compositional expression of many GPM algorithms. Fractal-based implementations outperform both existing systemic solutions and specialized distributed solutions on many problems - from frequent graph mining to subgraph querying, over a range of datasets.

Graph data mining is defined as searching in an input graph for all subgraphs that satisfy some property that makes them interesting to the user. Examples of graph data mining problems include frequent subgraph mining, counting motifs, and enumerating cliques. These problems differ from other graph processing problems such as PageRank or shortest path in that graph data mining requires searching through an exponential number of subgraphs. Most current parallel graph analytics systems do not provide good support for graph data mining. One notable exception is Arabesque, a system that was built specifically to support graph data mining. Arabesque provides a simple programming model to express graph data mining computations, and a highly scalable and efficient implementation of this model, scaling to billions of subgraphs on hundreds of cores. This demonstration will showcase the Arabesque system, focusing on the end-user experience and showing how Arabesque can be used to simply and efficiently solve practical graph data mining problems that would be difficult with other systems.

The increasing amount of data being stored and the variety of applications being proposed recently to make use of those data enabled a whole new generation of parallel programming environments and paradigms. Although most of these novel environments provide abstract programming interfaces and embed several run-time strategies that simplify several typical tasks in parallel and distributed systems, achieving good performance is still a challenge. In this paper we identify some common sources of performance degradation in the Spark programming environment and discuss some diagnosis dimensions that can be used to better understand such degradation. We then describe our experience in the use of those dimensions to drive the identification performance problems, and suggest how their impact may be minimized considering real applications.

Given the large amount of data from different sources that have become available to researchers in multiple fields, Data Science has emerged as a new paradigm for exploring and getting value from that data. In that context, new parallel processing environments with abstract programming interfaces, like Spark, were proposed to try to simplify the development of distributed programs. Although such solutions have become widely used, achieving the best performance with them is still not always straight-forward, despite the multiple run-time strategies they use. In this work we analyze some of the causes of performance degradation in such systems and, based on that analysis, we propose a tool to improve performance by dynamically adjusting data partitioning and parallelism degree in recurrent applications based on previous executions. Our results applying that methodology show consistent reductions in execution time for the applications considered, with gains of up to 50%.

Summary Most high‐performance data processing (a.k.a. big data) systems allow users to express their computation using abstractions (like MapReduce), which simplify the extraction of parallelism from applications. Most frameworks, however, do not allow users to specify how communication must take place: That element is deeply embedded into the run‐time system abstractions, making changes hard to implement. In this work, we describe Wathershed‐ng, our re‐engineering of the Watershed system, a framework based on the filter–stream paradigm and originally focused on continuous stream processing. Like other big‐data environments, Watershed provided object‐oriented abstractions to express computation (filters), but the implementation of streams was a run‐time system element. By isolating stream functionality into appropriate classes, combination of communication patterns and reuse of common message handling functions (like compression and blocking) become possible. The new architecture even allows the design of new communication patterns, for example, allowing users to choose MPI, TCP, or shared memory implementations of communication channels as their problem demands. Applications designed for the new interface showed reductions in code size on the order of 50%and above in some cases. The performance results also showed significant improvements, because some implementation bottlenecks were removed in the re‐engineering process. Copyright © 2016 John Wiley & Sons, Ltd.