Data Placement Approach Research Articles

Dynamic data replication and placement strategy in geographically distributed data centers

AbstractWith the evolution of geographically distributed data centers in the Cloud Computing landscape along with the amount of data being processed in these data centers, which is growing at an exponential rate, processing massive data applications become an important topic. Since a given task may require many datasets for its execution and the datasets are spread over several different data centers, finding an efficient way to manage the datasets storage across nodes of a Cloud system is a difficult problem. In fact, the execution time of a task might be influenced by the cost of data transfers, which mainly depends on two criterias. The first one is the initial placement of the input datasets during the build‐time phase, while the second is the replication of the datasets during the runtime phase. The replication is explicitly considered when datasets are being migrated over the data centers in order to make them locally available wherever needed. Data placement and data replication are important challenges in Cloud Computing. Nevertheless, many studies focus on data placement or data replication exclusively. In this paper, a combination of a data placement strategy followed by a dynamic data replication management strategy is proposed, with the purpose of reducing the associated cost of all data transfers between the (distant) data centers. Our proposed data placement approach considers the main characteristics of a data center such as storage capacity and read/write speeds to efficiently store the datasets, while our dynamic data replication management approach considers three parameters: the number of replicas in the system, the dependency between datasets and tasks and the storage capacity of data centers. The decision of when and whether to keep or to delete replicas is determined by the fulfillment of those three parameters. Our approach estimates the total execution time of the tasks as well as the monetary cost, considering the data transfers activity. Our experiments are conducted using Cloudsim simulator. The obtained results show that our proposed strategies produce an efficient data management by reducing the overheads of the data transfers, compared to both a data placement without replication (by 76%) and the selected data replication approach from Kouidri et al. (by 52%), and by improving the financial cost.

Optimal Data Placement and Replication Approach for SIoT with Edge

Social networks (SNs) are sources with extreme number of users around the world who are all sharing data like images, audio, and video to their friends using IoT devices. This concept is the so-called Social Internet of Things (SIot). The evolving nature of edge-cloud computing has enabled storage of a large volume of data from various sources, and this task demands an efficient storage procedure. For this kind of large volume of data storage, the usage of data replication using edge with geo-distributed cloud service area is suited to fulfill the user’s expectations with low latency. The major issue is the way to store the data and replicate these large data items optimally and allocate the request from the data center efficiently. For efficient storage of these data, we use edge server, which is part of the cloud server, in this study. Thus, the data are distributed and stored with quick access, which will reduce the latency with response. The proposed data placement approach learns with machine learning (ML) algorithm called radial basis kernel function assisted with support vector machine (RBF-SVM) to classify the data center for storing the user and friend’s data from the SIoT devices. These learning algorithms will be used to predict the workload of the data stored in the data center as either edge or cloud depending on the existing time slots. The data placement with dynamic nature is also optimized using the proposed dynamic graph partitioning (GP) method to meet the individual user’s demand of low latency with minimum costs. This way will keep the SIoT data placement efficient and effective over time. Accordingly, this proposed data placement and replication approach introduces three kinds of innovations compared with the existing data placement approach. (i) Rather than storing the user data in a single cloud, this study uses the edge server closest to the SIoT devices for faster access with reduced response time. (ii) The classification algorithm called RBF-SVM is used to find storage for user for reducing data replication. (iii) Dynamic GP is introduced for data placement with reduced latency and minimum cost to fulfil the dynamic nature of the SN. The simulation result of this approach obtains reduced latency of 130 ms and minimum cost compared with those of the existing data placement approaches. Therefore, our proposed data placement with ML-based learning on edge provides promising results in terms of efficiency, effectiveness, and performance with reduced latency and minimum cost.

RENDA: Resource and Network Aware Data Placement Algorithm for Periodic Workloads in Cloud

The Hadoop enabled cloud platforms are gradually becoming preferred computational environment to execute scientific big data workloads in a periodic manner. However, it is observed that the default data placement approach of such cloud platforms is not the efficient one and often ends up with significant data transfer overhead leading to degradation of the overall job completion time. In this article, a Resource and Network-aware Data Placement Algorithm (RENDA) is proposed to reduce the non-local executions and thereby reduce the overall job completion time for periodic workloads in the cloud environment. The entire job execution is modeled as a two-stage execution characterized as data distribution and data processing. The RENDA reduces the time of the stages as mentioned above by estimating the heterogeneous performance of the nodes on a real-time basis followed by careful allocation of data in several installments to participating nodes. The experimental results show that the proposed RENDA algorithm consistently outperforms over the recent state-of-the-art alternatives with as much as 28 percent reduction in data transfer overhead leading to 16 percent reduction in average job completion time with 27 percent average speedup on average job execution.

Reliability-aware Garbage Collection for Hybrid HBM-DRAM Memories

Emerging workloads in cloud and data center infrastructures demand high main memory bandwidth and capacity. Unfortunately, DRAM alone is unable to satisfy contemporary main memory demands. High-bandwidth memory (HBM) uses 3D die-stacking to deliver 4–8× higher bandwidth. HBM has two drawbacks: (1) capacity is low, and (2) soft error rate is high. Hybrid memory combines DRAM and HBM to promise low fault rates, high bandwidth, and high capacity. Prior OS approaches manage HBM by mapping pages to HBM versus DRAM based on hotness (access frequency) and risk (susceptibility to soft errors). Unfortunately, these approaches operate at a coarse-grained page granularity, and frequent page migrations hurt performance. This article proposes a new class of reliability-aware garbage collectors for hybrid HBM-DRAM systems that place hot and low-risk objects in HBM and the rest in DRAM. Our analysis of nine real-world Java workloads shows that: (1) newly allocated objects in the nursery are frequently written, making them both hot and low-risk, (2) a small fraction of the mature objects are hot and low-risk, and (3) allocation site is a good predictor for hotness and risk. We propose RiskRelief, a novel reliability-aware garbage collector that uses allocation site prediction to place hot and low-risk objects in HBM. Allocation sites are profiled offline and RiskRelief uses heuristics to classify allocation sites as DRAM and HBM. The proposed heuristics expose Pareto-optimal trade-offs between soft error rate (SER) and execution time. RiskRelief improves SER by 9× compared to an HBM-Only system while at the same time improving performance by 29% compared to a DRAM-Only system. Compared to a state-of-the-art OS approach for reliability-aware data placement, RiskRelief eliminates all page migration overheads, which substantially improves performance while delivering similar SER. Reliability-aware garbage collection opens up a new opportunity to manage emerging HBM-DRAM memories at fine granularity while requiring no extra hardware support and leaving the programming model unchanged.

Application and Storage-Aware Data Placement and Job Scheduling for Hadoop Clusters

As one of the most popular frameworks for large-scale analytics processing, Hadoop is facing two challenges: both applications and storage devices become heterogeneous. However, existing data placement and job scheduling schemes pay little attention to such heterogeneity of either application I/O requirements or I/O device capability, thus can greatly degrade system efficiencies. In this paper, we propose ASPS, an Application and Storage-aware data Placement and job Scheduling approach for Hadoop clusters. The idea is to place application data and schedule application tasks considering both application I/O requirements and storage device characteristics. Specifically, ASPS first introduces novel metrics to quantify I/O requirements of applications. Then, based on the quantification, ASPS places data of different applications to the preferred storage devices. Finally, ASPS tries to launch jobs with high I/O requirements on the nodes with the same type of faster devices to improve system efficiency. We have implemented ASPS in Hadoop framework. Experimental results show that ASPS can reduce the completion time of a single application by up to 36% and the average completion time of six concurrent applications by 27%, compared to existing data placement policies and job scheduling approaches.

QoS-Aware Data Placement for MapReduce Applications in Geo-Distributed Data Centers

With growing data volumes and the scaling of data center clusters, communication resources often become a bottleneck in service provisioning for many MapReduce applications (e.g., training machine learning models). Therefore, data placements that bring data blocks closer to data consumers (e.g., MapReduce applications) are seen as a promising solution. In this article, we propose an efficient data-placement technique that considers network traffic reduction as well as QoS guarantees for the data blocks to optimize the communication resources. We first formulate the joint optimization of the data-placement problem, propose a generic model for minimizing communication costs, and show that the joint data-placement problem is NP-hard. To solve this problem, we propose a heuristic algorithm considering traffic flows in the network topology of data centers by first seeking optimal QoS-aware data placement based on golden division on a Zipflike replica distribution, then transforming the joint data-placement problem into a block-dependence tree (BDT) construction problem, and finally reducing the BDT construction to a graph-partitioning problem. The experimental results demonstrate that our data-placement approach could effectively improve the performance of MapReduce jobs with lower communication costs and less job execution time for big-data processing.

Special Issue on Infrastructures and Algorithms for Scalable Computing

We are happy to present this special issue of the scientific journal Scalable Computing: Practice and Experience. In this special issue on Infrastructures and Algorithms for Scalable Computing (Volume 19, No 3 June 2018), we have selected four papers out of submitted nine, which gone through a peer review according to the journal policy. All papers represent novel results in the fields of distributed algorithms and infrastructures for scalable computing.
 The first paper presents present a novel approach for efficient data placement, which improves the performance of workflow execution in distributed datacenters. The greedy heuristic algorithm, which is based on a network flow optimization framework, minimizes the total storage cost, including efforts to move and store the data from different source locations and dependencies. The second paper evaluated the significance of different clustering techniques viz. k-means, Hierarchical Agglomerative Clustering and Markov Clustering in groupingawaredata placement for data-intensive applications with interest locality. The evaluation in Azure reported that Markov Clustering-based data placement strategy improves the local map execution and reduces the execution time compared to Hadoops Default Data Placement Strategy and other evaluated clustering techniques. This is more emphasized for data-intensive applications that have interest locality. The third paper presents an experimental evaluation of the openMP thread-mapping strategies in different hardware environments (IntelXeon Phi coprocessor and hybrid CPU-MIC platforms). The paper shows the optimal choice of thread affinity, the number of threads and the execution mode that can provide optimal performance of the LU factorization. In the fourth paper, the authors study the amount of memory occupied by sparse matrices split up into same-size blocks. The paper considers and statistically evaluates four popular storage formats and combinations among them. The conclusion is that block-based storage formats may significantly reduce memory footprints of sparse matrices arising from a wide range of application domains.
 We use this opportunity to thank all contributors to this Special Issue: all authors who submitted the results of their latest research and all reviewers for their valuable comments and suggestions for improvement. We would like to express our special gratitude for the Editor-in-Chief, Professor Dana Petcu, for her constant support during the whole process of this Special Issue.

Exact and Heuristic Data Workflow Placement Algorithms for Big Data Computing in Cloud Datacenters

Several big data-driven applications are currently carried out in collaboration using distributed infrastructure. These data-driven applications usually deal with experiments at massive scale. Data generated by such experiments are huge and stored at multiple geographic locations for reuse. Workflow systems, composed of jobs using collaborative task-based models, present new dependency and data exchange needs. This gives rise to new issues when selecting distributed data and storage resources so that the execution of applications is on time, and resource usage-cost-efficient. In this paper, we present an efficient data placement approach to improve the performance of workflow processing in distributed data centres. The proposed approach involves two types of data: splittable and unsplittable intermediate data. Moreover, we place intermediate data by considering not only their source location but also their dependencies. The main objective is to minimise the total storage cost, including the effort for transferring, storing, and moving that data according to the applications needs. We first propose an exact algorithm which takes into account the intra-job dependencies, and we show that the optimal fractional intermediate data placement problem is NP-hard. To solve the problem of unsplittable intermediate data placement, we propose a greedy heuristic algorithm based on a network flow optimisation framework. The experimental results show that the performance of our approach is very promising. We also show that even with divergent conditions, the cost ratio of the heuristic approach is close to the optimal solution.

A vibrant data placement approach for map reduce in diverse environments

Map reduce assumes that the computing capacity is same for each node in a cluster. Each node is assigned to the same load in homogeneous environment, hence it fully use the resources in the cluster. In such a cluster, there is likely to be various specifications of PCs or servers, which causes the abilities of the nodes to differ. If such a heterogeneous environment still uses the original Hadoop strategy that distributes data blocks into each node equally and the load is also evenly distributed to each node, then the overall performance of Hadoop may be reduced. The majorreasonis thatdifferentcomputing capacitiesbetweennodes causethetask executiontimeto differ so thatthefasterexecutionrate nodes processinglocal data blocks faster than other slower nodes do.The required data should be transferredfrom another node through the network.Becausewaitingforthedatatransmissiontimeincreasesthetask executiontime,it causestheentirejobexecution timeto becomeprolonged.

Data placement approach for scalable online social networks

Online social networks are attracting more and more users. Faced with hundreds of millions of users, how to store user data in a scalable manner has become a hot issue of focus for both social service providers and researchers. Currently, distributed key value store is widely used; it places user data across multiple storage servers based on a hash approach. However, it results in a huge amount of communication traffic inside a data center, and is not conducive to the scale of social networks. By considering user interaction characteristics, this paper proposes a data placement approach that combines both social graph partitioning and data replication. Considering the network topologies of data centers, we design the data placement for specific topologies. Furthermore, we discuss the incremental adjustment for social network growth and the distributed implementation of the proposed algorithms. Finally, experiments on real world traces indicate that the proposed algorithms can effectively reduce internal communication traffic, thereby enhancing the scalability of online social networks.

ExaPlan

Multi-tiered storage, where each tier consists of one type of storage device (e.g., SSD, HDD, or disk arrays), is a commonly used approach to achieve both high performance and cost efficiency in large-scale systems that need to store data with vastly different access characteristics. By aligning the access characteristics of the data, either fixed-sized extents or variable-sized files, to the characteristics of the storage devices, a higher performance can be achieved for any given cost. This article presents ExaPlan, a method to determine both the data-to-tier assignment and the number of devices in each tier that minimize the system’s mean response time for a given budget and workload. In contrast to other methods that constrain or minimize the system load, ExaPlan directly minimizes the system’s mean response time estimated by a queueing model. Minimizing the mean response time is typically intractable as the resulting optimization problem is both nonconvex and combinatorial in nature. ExaPlan circumvents this intractability by introducing a parameterized data placement approach that makes it a highly scalable method that can be easily applied to exascale systems. Through experiments that use parameters from real-world storage systems, such as CERN and LOFAR, it is demonstrated that ExaPlan provides solutions that yield lower mean response times than previous works. It supports standalone SSDs and HDDs as well as disk arrays as storage tiers, and although it uses a static workload representation, we provide empirical evidence that underlying dynamic workloads have invariant properties that can be deemed static for the purpose of provisioning a storage system. ExaPlan is also effective as a load-balancing tool used for placing data across devices within a tier, resulting in an up to 3.6-fold reduction of response time compared with a traditional load-balancing algorithm, such as the Longest Processing Time heuristic.

Efficient location-aware data placement for data-intensive applications in geo-distributed scientific data centers

Recent developments in cloud computing and big data have spurred the emergence of data-intensive applications for which massive scientific datasets are stored in globally distributed scientific data centers that have a high frequency of data access by scientists worldwide. Multiple associated data items distributed in different scientific data centers may be requested for one data processing task, and data placement decisions must respect the storage capacity limits of the scientific data centers. Therefore, the optimization of data access cost in the placement of data items in globally distributed scientific data centers has become an increasingly important goal. Existing data placement approaches for geo-distributed data items are insufficient because they either cannot cope with the cost incurred by the associated data access, or they overlook storage capacity limitations, which are a very practical constraint of scientific data centers. In this paper, inspired by applications in the field of high energy physics, we propose an integer-programming-based data placement model that addresses the above challenges as a Non-deterministic Polynomial-time (NP)-hard problem. In addition we use a Lagrangian relaxation based heuristics algorithm to obtain ideal data placement solutions. Our simulation results demonstrate that our algorithm is effective and significantly reduces overall data access cost.

Tology-Aware Optimal Data Placement Algorithm for Network Traffic Optimization

We propose a new optimal data placement technique to improve the performance of MapReduce in cloud data centers by considering not only the data locality but also the global data access costs. We first conducted an analytical and experimental study to identify the performance issues of MapReduce in data centers and to show that MapReduce tasks that are involved in unexpected remote data access have much greater communication costs and execution time, and can significantly deteriorate the overall performance. Next, we formulated the problem of optimal data placement and proposed a generative model to minimize global data access cost in data centers and showed that the optimal data placement problem is NP-hard. To solve the optimal data placement problem, we propose a topology-aware heuristic algorithm by first constructing a replica-balanced distribution tree for the abstract tree structure, and then building a replica-similarity distribution tree for detail tree construction, to construct an optimal replica distribution tree. The experimental results demonstrated that our optimal data placement approach can improve the performance of MapReduce with lower communication and computation costs by effectively minimizing global data access costs, more specifically reducing unexpected remote data access.

SWORD: workload-aware data placement and replica selection for cloud data management systems

Cloud computing is increasingly being seen as a way to reduce infrastructure costs and add elasticity, and is being used by a wide range of organizations. Cloud data management systems today need to serve a range of different workloads, from analytical read-heavy workloads to transactional (OLTP) workloads. For both the service providers and the users, it is critical to minimize the consumption of resources like CPU, memory, communication bandwidth, and energy, without compromising on service-level agreements if any. In this article, we develop a workload-aware data placement and replication approach, called SWORD, for minimizing resource consumption in such an environment. Specifically, we monitor and model the expected workload as a hypergraph and develop partitioning techniques that minimize the average query span, i.e., the average number of machines involved in the execution of a query or a transaction. We empirically justify the use of query span as the metric to optimize, for both analytical and transactional workloads, and develop a series of replication and data placement algorithms by drawing connections to several well-studied graph theoretic concepts. We introduce a suite of novel techniques to achieve high scalability by reducing the overhead of partitioning and query routing. To deal with workload changes, we propose an incremental repartitioning technique that modifies data placement in small steps without resorting to complete repartitioning. We propose the use of fine-grained quorums defined at the level of groups of data items to control the cost of distributed updates, improve throughput, and adapt to different workloads. We empirically illustrate the benefits of our approach through a comprehensive experimental evaluation for two classes of workloads. For analytical read-only workloads, we show that our techniques result in significant reduction in total resource consumption. For OLTP workloads, we show that our approach improves transaction latencies and overall throughput by minimizing the number of distributed transactions.

A LNS-based data placement strategy for data-intensive e-science applications

With the growth of data, the era of big data comes. Big data requires a huge amount of storage and new data placement methods. For data-intensive e-science applications in grid and utility computing, they need to process large data sets resided in different data centres. The data transfer across data centres leads to bandwidth cost and time delay. It is more important to solve the data placement problem for these applications. The existing data placement approaches can only reduce the data transfer frequency, but not improve overall performance. In this paper, we propose a data placement strategy based on large neighbourhood search (LNS). The strategy has initial strategy generation stage and relaxation and re-insertion stage. Our strategy offers the improvements in overall performance, especially in reducing time delay.

A Cloud‐Computing‐Based Data Placement Strategy in High‐Speed Railway

As an important component of China’s transportation data sharing system, high‐speed railway data sharing is a typical application of data‐intensive computing. Currently, most high‐speed railway data is shared in cloud computing environment. Thus, there is an urgent need for an effective cloud‐computing‐based data placement strategy in high‐speed railway. In this paper, a new data placement strategy named hierarchical structure data placement strategy is proposed. The proposed method combines the semidefinite programming algorithm with the dynamic interval mapping algorithm. The semi‐definite programming algorithm is suitable for the placement of files with various replications, ensuring that different replications of a file are placed on different storage devices, while the dynamic interval mapping algorithm ensures better self‐adaptability of the data storage system. A hierarchical data placement strategy is proposed for large‐scale networks. In this paper, a new theoretical analysis is provided, which is put in comparison with several other previous data placement approaches, showing the efficacy of the new analysis in several experiments.

A High-Speed Railway Data Placement Strategy Based on Cloud Computing

The application of high-speed railway data, which is an important component of China's transportation science data sharing, has embodied the typical characteristics of data-intensive computing. A reasonable and effective data placement strategy is needed to deploy and execute data-intensive applications in the cloud computing environment. Study results of current data placement approaches have been analyzed and compared in this paper. Combining the semi-definite programming algorithm with the dynamic interval mapping algorithm, a hierarchical structure data placement strategy is proposed. The semi-definite programming algorithm is suitable for the placement of files with various replications, ensuring that different replications of a file are placed on different storage devices. And the dynamic interval mapping algorithm could guarantee better self-adaptability of the data storage system. It has been proved both by theoretical analysis and experiment demonstration that a hierarchical data placement strategy could guarantee the self-adaptability, data reliability and high-speed data access for large-scale networks.

Distributed Real Time Architecture for Data Placement in Wireless Sensor Networks

Problem statement: Application scenarios of Wireless Sensor Networks (WSN) create interesting design challenges since the sensor nodes used to collect and communicate data may fail suddenly and unpredictably. The issue of real-time and reliable data delivery is extremely important for taking effective decisions in WSN. Approach: In this study we proposed architecture for reliable and real-time approach for data placement using sensor clusters. Instead of storing information in individual cluster heads as suggested in some protocols, in our architecture we suggest storing of information of all clusters within a cell in the corresponding base station. For data dissemination and action we have used Action and Relay Stations (ARS). Results: We developed programming model for formal specification of our architecture. Conclusion: Present model not only offers real time sensing and monitoring, but also provides real time dynamic decision making based on the sensed data. Moreover it prolongs network life time.

Exploiting sequential access when declustering data over disks and MEMS-based storage

Due to the large difference between seek time and transfer time in current disk technology, it is advantageous to perform large I/O using a single sequential access rather than multiple small random I/O accesses. However, prior optimal cost and data placement approaches for processing range queries over two-dimensional datasets do not consider this property. In particular, these techniques do not consider the issue of sequential data placement when multiple I/O blocks need to be retrieved from a single device. In this paper, we reevaluate the optimal cost of range queries by declustering two-dimensional datasets over multiple devices, and prove that, in general, it is impossible to achieve the new optimal cost. This is because disks cannot facilitate two-dimensional sequential access which is required by the new optimal cost. Then we revisit the existing data allocation schemes under the new optimal cost, and show that none of them can achieve the new optimal cost. Fortunately, MEMS-based storage is being developed to reduce I/O cost. We first show that the two-dimensional sequential access requirement can not be satisfied by simply modeling MEMS-based storage as conventional disks. Then we propose a new placement scheme that exploits the physical properties of MEMS-based storage to solve this problem. Our theoretical analysis and experimental results show that the new scheme achieves almost optimal I/O costs.

A simulated annealing approach for multimedia data placement

Multimedia applications are characterized by their strong timing requirements and constraints and thus multimedia data storage is a critical issue in the overall system's performance and functionality. This paper describes multimedia data representation models that effectively guide data placement towards the improvement of the Quality of Presentation for the considered multimedia applications. The performance of both constructive placement and iterative improvement placement algorithms is evaluated and discussed. Emphasis is given on placement schemes which are based on the simulated annealing optimization algorithm. A placement policy, based on a self-improving version of the simulated annealing (SISA) algorithm is applied and evaluated. Performance of the placement policies is experimentally evaluated on a simulated tertiary storage subsystem. As proven by the experimentation, the proposed approach shows considerable gain in terms of seek and service times. The improvements of the proposed SISA approach are in the range of 40% when compared to random placement and at the range of 15–35% when compared to the typical simulated annealing algorithm, depending a lot on the initial configuration and the neighborhood search.