Energy-Efficient Speculative Execution using Advanced Reservation for Heterogeneous Clusters

"Big data" has become a major force of innovation across enterprises of all sizes. New platforms with increasingly more features for managing big datasets are being announced almost on a weekly basis. Yet, there is currently a lack of any means of comparability among such platforms. While the performance of traditional database systems is well understood and measured by long-established institutions such as the Transaction Processing Performance Council (TCP), there is neither a clear definition of the performance of big data systems nor a generally agreed upon metric for comparing these systems. In this article, we describe a community-based effort for defining a big data benchmark. Over the past year, a Big Data Benchmarking Community has become established in order to fill this void. The effort focuses on defining an end-to-end application-layer benchmark for measuring the performance of big data applications, with the ability to easily adapt the benchmark specification to evolving challenges in the big data space. This article describes the efforts that have been undertaken thus far toward the definition of a BigData Top100 List. While highlighting the major technical as well as organizational challenges, through this article, we also solicit community input into this process.

Big data applications and analytics are employed in many sectors for a variety of goals: improving customers satisfaction, predicting market behavior or improving processes in public health. These applications consist of complex software stacks that are often run on cloud systems. Predicting execution times is important for estimating the cost of cloud services and for effectively managing the underlying resources at runtime. Machine Learning (ML), providing black box solutions to model the relationship between application performance and system configuration without requiring in-detail knowledge of the system, has become a popular way of predicting the performance of big data applications. We investigate the cost-benefits of using supervised ML models for predicting the performance of applications on Spark, one of today's most widely used frameworks for big data analysis. We compare our approach with Ernest (an ML-based technique proposed in the literature by the Spark inventors) on a range of scenarios, application workloads, and cloud system configurations. Our experiments show that Ernest can accurately estimate the performance of very regular applications, but it fails when applications exhibit more irregular patterns and/or when extrapolating on bigger data set sizes. Results show that our models match or exceed Ernest's performance, sometimes enabling us to reduce the prediction error from 126-187% to only 5-19%.

Many real world applications have problems with high dimensionality, which existing algorithms cannot overcome. A critical data preprocessing problem is feature selection, whereby its non-scalability negatively influences both the efficiency and performance of big data applications. In this research, we developed a new algorithm to reduce the dimensionality of a problem using graph-based analysis, which retains the physical meaning of the original high-dimensional feature space. Most existing feature-selection methods are based on a strong assumption that features are independent of each other. However, if the feature-selection algorithm does not take into consideration the interdependencies of the feature space, the selected data fail to correctly represent the original data. We developed a new feature-selection method to address this challenge. Our aim in this research was to examine the dependencies between features and select the optimal feature set with respect to the original data structure. Another important factor in our proposed method is that it can perform even in the absence of class labels. This is a more difficult problem that many feature-selection algorithms fail to address. In this case, they only use wrapper techniques that require a learning algorithm to select features. It is important to note that our experimental results indicates, this proposed simple ranking method performs better than other methods, independent of any particular learning algorithm used.

The execution time of a stage is extended by a few slow running tasks in Spark computing environments. To tackle this so-called straggler problem, Spark adopts speculative execution mechanism under which the scheduler speculatively launch additional backup for the straggler with the hope to complete early. However, due to the characteristics of tasks and the complexity of runtime environments, the Spark original speculative execution strategy and its improved versions cannot deal with this problem effectively. In this paper, we propose a novel strategy called ETWR to improve the efficiency of speculative execution in Spark. We consider the heterogeneous environment when we around to tackle the three key points of speculative execution: straggler identification, backup node selection and effectiveness guarantee. Based on the task type classification, first, we divide the task into sub-phases and use both the process speed and progress rate within a phase to find the straggler promptly. Second, we use the Locally Weighted Regression model to estimate the execution time of the task, which will be used to calculate the task's remaining time and backup time. Third, we present iMCP model to guarantee the effectiveness of speculative tasks, which can additionally keep load balancing for nodes. Finally, the factors of fast node and better location are considered when choosing proper backup nodes. Extensive experiments show that ETWR can reduce the job execution time by 23.8 percent, and improve the cluster throughput by 33.2 percent compared with Spark-2.2.0.

Nowadays, Big Data becomes a research focus in industrial, banking, social network, and other fields. In addition, the explosive increase of data and information require efficient processing solutions. Therefore, Spark is considered as a promising candidate of Large-Scale Distributed Computing Systems for big data processing. One primary challenge is the straggler problem that occurred due to the presence of heterogeneity where a machine takes an extra-long time to finish execution of a task, which decreases the system throughput. To mitigate straggler tasks, Spark adopts speculative execution mechanism, in which the scheduler launches additional backup to avoid slow task processing and achieve acceleration. In this paper, a new Optimized Straggler Mitigation Framework is proposed. The proposed framework uses a dynamic criterion to determine the closest straggler tasks. This criterion is based on multiple coefficients to achieve a reliable straggler decision. Also, it integrates the historical data analysis and online adaptation for intelligent straggler judgment. This guarantees the effectiveness of speculative tasks by improving cluster performance. Experimental results on various benchmarks and applications show that the proposed framework achieves 23.5% to 30.7% execution time reductions, and 25.4 to 46.3% increase of the cluster throughputs compared with spark engine.

Cloud computing and big data: Technologies and applications

Task stragglers dramatically impede parallel job execution of data-intensive computing in Cloud Datacenters Due to the uneven distribution of input data resulted from heterogeneous data nodes, resource contention situations, and network configurations, it causes delay failures due to the violation of job completion time. However, data-intensive computing frameworks, such as MapReduce or Hadoop, employ a mechanism called speculative execution to deal with the straggler issue, speculative execution provide limited effectiveness because in many cases straggler identification occurs too late within a job lifecycle. Identifying the straggler and the timing of identifying it is very important for Straggler mitigation in Data-intensive cloud computing. Speculative execution method is a widely adopted as a straggler identification and mitigation scheme but it has certain inherent limitations. In this paper, we strive to make Hadoop more efficient in cloud environments. We present Progress and Feedback based Speculative Execution Algorithm (PFSE), a new Straggler identification scheme to identify the straggler MapReduce tasks based on the feedback information received from completed tasks beside the progress of the currently processing task, our extensive simulation shows that PFSE can outperform the dynamic scheduling techniques like Self-Learning MapReduce scheduler (SLM) and LATE. PFSE can assist in enhancing straggler Identification and mitigation for tolerating late-timing failures within data intensive cloud computing.

Dynamic information flow tracking (also known as taint tracking) is an appealing approach to combat various security attacks. However, the performance of applications can severely degrade without hardware support for tracking taints. This paper observes that information flow tracking can be efficiently emulated using deferred exception tracking in microprocessors supporting speculative execution. Based on this observation, we propose SHIFT, a low-overhead, software-based dynamic information flow tracking system to detect a wide range of attacks. The key idea is to treat tainted state (describing untrusted data) as speculative state (describing deferred exceptions). SHIFT leverages existing architectural support for speculative execution to track tainted state in registers and needs to instrument only load and store instructions to track tainted state in memory using a bitmap, which results in significant performance advantages. Moreover, by decoupling mechanisms for taint tracking from security policies, SHIFT can detect a wide range of exploits, including high-level semantic attacks. We have implemented SHIFT using the Itanium processor, which has support for deferred exceptions, and by modifying GCC to instrument loads and stores. A security assessment shows that SHIFT can detect both low-level memory corruption exploits as well as high-level semantic attacks with no false positives. Performance measurements show that SHIFT incurs about 1% overhead for server applications. The performance slowdown for SPEC-INT2000 is 2.81X and 2.27X for tracking at byte-level and wordlevel respectively. Minor architectural improvements to the Itanium processor (adding three simple instructions) can reduce the performance slowdown down to 2.32X and 1.8X for byte-level and word-level tracking, respectively.

In the last fifteen years, services deployed via the Web have become a part of our everyday lives. Supporting Web technologies have also evolved very quickly to respond to the new services and features demanded by mobile and always connected users. Currently, the demand for services on the Web is still growing at a very fast pace as well as new techniques, methods and cutting-edge technologies that make the Web an increasingly complex environment.

The appearance of simultaneous multithreading (SMT) processors, offering instruction-level parallelism and increasing hardware resources utilization, permits the execution of various instructions streams at the same time. A particular case is the Intel SMT implementation (Hyper-Threading Technology) where a processor can execute two instructions streams independently. Due to hardware resources sharing in a Hyper-Threading processor, both instruction streams can contend for the same execution resources depending on the instruction execution units utilization.As other works in this area have demonstrated, we have observed that the system throughput can be improved when the execution resources used by each thread are contemplated in their scheduling. Starting from the evaluation results, we propose a Thread Coloring philosophy that designs a way to manage information about resources utilization (hardware and software resources) per each thread. This policy permits the scheduler to select threads that will perform better with less resource contention, besides including OS upper layers in the resource management. The proposed Thread Coloring scheme establishes the bases of processor and OS cooperation, sharing resources utilization information and providing a form of controlling logical processor thread assignment taking into account possible resource contention.

In a dynamic reordering superscalar processor, the front-end fetches instructions and places them in the issue queue. Instructions are then issued by the back-end execution core. Till recently, the front-end was designed to maximize performance without considering energy consumption. The front-end fetches instructions as fast as it can until it is stalled by a filled issue queue or some other blocking structure. This approach wastes energy: (i) speculative execution causes many wrong-path instructions to be fetched and executed, and (ii) back-end execution rate is usually less than its peak rate, but front-end structures are dimensioned to sustained peak performance. Dynamically reducing the front-end instruction rate and the active size of front-end structure (e.g. issue queue) is a required performance-energy trade-off. Techniques proposed in the literature attack only one of these effects. In this paper, we propose Speculative Instruction Window Weighting (SIWW), a fetch gating technique that allows to address both fetch gating and instruction issue queue dynamic sizing. A global weight is computed on the set of inflight instructions. This weight depends on the number and types of inflight instructions (non-branches, high confidence or low confidence branches, ...). The front-end instruction rate can be continuously adapted based on this weight. SIWW is shown to perform better than previously proposed fetch gating techniques. SIWW is also shown to allow to dynamically adapt the size of the active instruction queue.

The proliferation of Big Data processing environments such as Hadoop, Apache Spark, and HPCC Systems is driving the development of performance analysis tools in these distributed systems. The goal is to achieve high performance through the optimization of Big Data applications. However, tuning performance in a fine-grained manner is quite challenging due to the high complexity and massive size of the distributed systems. ECL-Watch is a data-flow based fine-grained comprehensive Big Data performance analysis tool utilizing the high level declarative dataflow programming language ECL in HPCC Systems. As a case study, we implement and optimize the Yinyang K-Means machine learning algorithm in ECL in HPCC Systems. The experimental results show that the performance of the native ECL version of the Yinyang K-Means algorithm increased significantly after tuning: from being about three times slower than the standard K-Means implementation in ECL, to become roughly 15% faster than standard K-Means.

This article explores methods for optimizing energy consumption in smart buildings using big data analysis and Internet of Things (IoT) technology. The research initiates with the extraction of pivotal attributes-including time, ambient temperature, humidity levels, and illumination intensity-via the processes of data cleansing and preparation. Following this, both descriptive and inferential statistical methodologies are applied to investigate the interplay between energy usage and a range of contributing elements. For predicting future energy consumption patterns of structures, the investigation employs the Nonlinear Autoregressive (NAR) neural network approach alongside the Long Short-Term Memory (LSTM) network to develop a predictive model. Central to this endeavor is the Intelligent Energy Management System (IEMS), which harnesses AI and machine learning (ML) techniques to facilitate real-time surveillance, precise forecasting, and smart optimization of a building's energy consumption metrics. By leveraging IoT technology, the IEMS gathers consumption data, applies advanced big data analytics combined with ML algorithms for processing and interpretation, and devises optimal strategies for energy utilization predicated on forecast outcomes and external influences. Empirical findings indicate that the deployment of these optimization tactics has led to a reduction in overall building energy consumption by approximately 20.35 %, with the most pronounced effect observed in air conditioning energy use, which saw a decrease of 25.23%. This research shows that the intelligent building energy consumption management system combining big data analysis and IoT technology has remarkable effects in improving energy utilization efficiency and reducing energy consumption, which provides strong technical support and practical reference for energy consumption optimization of intelligent buildings in the future.

Public and private cloud computing environments employ virtualization methods to consolidate application workloads onto shared servers. Modern servers typically have one or more sockets each with one or more computing cores, a multi-level caching hierarchy, a memory subsystem, and an interconnect to the memory of other sockets. While resource management methods may manage application performance by controlling the sharing of processing time and input-output rates, there is generally no management of contention for virtualization kernel resources or for the memory hierarchy and subsystems. Yet such contention can have a significant impact on application performance. Hardware platform specific counters have been proposed for detecting such contention. We show that such counters alone are not always sufficient for detecting contention. We propose a software probe based approach for detecting contention for shared platform resources and demonstrate its effectiveness. We show that the probe imposes low overhead and is remarkably effective at detecting performance degradations due to inter-VM interference over a wide variety of workload scenarios and on two different server architectures. The probe successfully detected virtualization-induced software bottleneck and memory contention on both server architectures. Our approach supports the management of workload placement on shared servers and pools of shared servers.

Nowadays, big data analytics has been widely applied in addressing the growing cybercrime threats. However, energy consumption is explosive increasing with the fast growth of big data processing in anti-cybercrime. In this paper, an energy-efficient framework for big data applications is proposed to reduce energy consumption while satisfying deadline constrains. First, the problem of energy-efficient tasks scheduling of a single Spark job is modeled as an integer program. We design an energy-efficient tasks scheduling algorithm to minimize the energy consumption for big data application in Spark. To avoid service-level agreement violations for execution time, we propose an optimal task scheduling algorithm with deadline constrains by tradingoff execution time and energy consumption. Experiments on a Spark cluster are performed to determine the energy consumption and execution time for several workloads from the HiBench benchmark suite. Our algorithms consume less energy on average than FIFO and FAIR under deadlines. The optimal algorithm is able to find near optimal tasks schedules to trade off energy consumed and response time benefit in small shuffle partitions.