Search

Although real-time analytics on the up-to-date dataset has become an emerging demand, many big data systems are still designed for offline analytics. Particularly, for critical applications like Fintech, transactional data ingestion ensures a timely, always-correct, and scalable dataset. To carry out append-only ingestion, existing OLTP/HTAP systems are based on strict transactions with imperfect scalability, while NoSQL-like systems support scalable but relaxed transactions. How to ensure essential transactional guarantees without harming scalability seems to be a non-trivial issue. This paper proposes Karst to bring transactional data ingestion for existing offline analytics. We notice that blocking two-phase commit (2PC) to resolve transactional data ingestion is a performance killer for the partitioned analytical systems. Karst introduces a scalable protocol called metadata-oriented commit (MOC) that converts each distributed transaction into multiple partial transactions to avoid 2PC. Moreover, to ingest massive data into plenty of partitions, Karst also employs lazy persistence, lightweight logging, and optimized data traffic. In experiments, Karst could achieve up to about 2x $\sim$ ∼ 10x performance over relevant systems and also shows remarkable scalability. [ABSTRACT FROM AUTHOR]

Nowadays, GPUs have achieved high throughput computing by running plenty of threads. However, owing to disjoint memory spaces of discrete CPU-GPU systems, exploiting CPU and GPU within a data processing pipeline is a non-trivial issue, which can only be resolved by the coarse-grained workflow of “copy-kernel-copy” or its variants in essence. There is an underlying bottleneck caused by frequent inter-processor invocations for fine-grained batch sizes. This article presents XeFlow that enables streamlined execution by leveraging hardware mechanisms inside new generation GPUs. XeFlow significantly reduces costly explicit copy and kernel launching within existing fashions. As an alternative, XeFlow introduces persistent operators that continuously process data through shared topics, which establish efficient inter-processor data channels via hardware page faults. Compared with the default “copy-kernel-copy” method, XeFlow shows up to $2.4\times \!\sim \!3.1\times$ 2. 4 × ∼ 3. 1 × performance advantages in both coarse-grained and fine-grained pipeline execution. To demonstrate its potentials, this article also evaluates two GPU-accelerated applications, including data encoding and OLAP query. [ABSTRACT FROM AUTHOR]

To fully utilize the resources of a main memory database cluster, we additionally take the independent parallelism into account to parallelize multiple pipelines of one query. However, scheduling resources to multiple pipelines is an intractable problem. Traditional static approaches to this problem may lead to a serious waste of resources and suboptimal execution order of pipelines, because it is hard to predict the actual data distribution and fluctuating workloads at compile time. In response, we propose a dynamic scheduling algorithm, List with Filling and Preemption (LFPS), based on two novel techniques. (1) Adaptive filling improves resource utilization by issuing more extra pipelines to adaptively fill idle resource “holes” during execution. (2) Rank-based preemption strictly guarantees scheduling the pipelines on the critical path first at run time. Interestingly, the latter facilitates the former filling idle “holes” with best efforts to finish multiple pipelines as soon as possible. We implement LFPS in our prototype database system. Under the workloads of TPC-H, experiments show our work improves the finish time of parallelizable pipelines from one query up to 2.5X than a static approach and 2.1X than a serialized execution. [ABSTRACT FROM AUTHOR]

With the emergence of the big data age, the issue of how to obtain valuable knowledge from a dataset efficiently and accurately has attracted increasingly attention from both academia and industry. This paper presents a Parallel Random Forest (PRF) algorithm for big data on the Apache Spark platform. The PRF algorithm is optimized based on a hybrid approach combining data-parallel and task-parallel optimization. From the perspective of data-parallel optimization, a vertical data-partitioning method is performed to reduce the data communication cost effectively, and a data-multiplexing method is performed is performed to allow the training dataset to be reused and diminish the volume of data. From the perspective of task-parallel optimization, a dual parallel approach is carried out in the training process of RF, and a task Directed Acyclic Graph (DAG) is created according to the parallel training process of PRF and the dependence of the Resilient Distributed Datasets (RDD) objects. Then, different task schedulers are invoked for the tasks in the DAG. Moreover, to improve the algorithm's accuracy for large, high-dimensional, and noisy data, we perform a dimension-reduction approach in the training process and a weighted voting approach in the prediction process prior to parallelization. Extensive experimental results indicate the superiority and notable advantages of the PRF algorithm over the relevant algorithms implemented by Spark MLlib and other studies in terms of the classification accuracy, performance, and scalability. With the expansion of the scale of the random forest model and the Spark cluster, the advantage of the PRF algorithm is more obvious. [ABSTRACT FROM PUBLISHER]

In the cloud platform, the startup security of guest virtual machines (VMs) can be guaranteed by existing techniques such as TBoot, however, how to monitor and guarantee their runtime security seems to be a non-trivial challenge, when they are exposed to the Internet. For a practical cloud system, security and performance are two important issues. In this paper, we propose a dynamic framework called CloudMon to detect kernel rootkits and guarantee the runtime security of guest VMs. CloudMon is transparent to a guest VM, neither requires its specific system information, nor has to one-on-one run with it. Meanwhile, CloudMon detects kernel rootkits through self-adjusting monitoring on memory with an acceptable overhead. A working prototype of CloudMon is implemented based on Xen. The case studies on security show that CloudMon is effective to detect kernel rootkits in guest VMs, while the performance experiments demonstrate that it brings a low performance overhead. [ABSTRACT FROM PUBLISHER]

The data-oriented applications have introduced increased demands on memory capacity and bandwidth, which raises the need to rethink the architecture of the current computing platforms. The logic-in-memory architecture is highly promising as future logic-memory integration paradigm for high throughput data-driven applications. From memory technology aspect, as one recently introduced nonvolatile memory device, domain-wall nanowire (or race-track) not only shows potential as future power efficient memory, but also computing capacity by its unique physics of spintronics. This paper explores a novel distributed in-memory computing architecture where most logic functions are executed within the memory, which significantly alleviates the bandwidth congestion issue and improves the energy efficiency. The proposed distributed in-memory computing architecture is purely built by domain-wall nanowire, i.e., both memory and logic are implemented by domain-wall nanowire devices. As a case study, neural network-based image resolution enhancement algorithm, called DW-NN, is examined within the proposed architecture. We show that all operations involved in machine learning on neural network can be mapped to a logic-in-memory architecture by nonvolatile domain-wall nanowire. Domain-wall nanowire-based logic is customized for in machine learning within image data storage. As such, both neural network training and processing can be performed locally within the memory. The experimental results show that the domain-wall memory can reduce 92% leakage power and 16% dynamic power compared to main memory implemented by DRAM; and domain-wall logic can reduce 31% both dynamic and 65% leakage power under the similar performance compared to CMOS transistor-based logic. And system throughput in DW-NN is improved by 11.6x and the energy efficiency is improved by 56x when compared to conventional image processing system. [ABSTRACT FROM PUBLISHER]

Virtualization plays a vital role in building the infrastructure of clouds, and isolation is considered as one of its important features. However, we demonstrate with practical measurements that there exist two kinds of isolation problems in current virtualized systems, due to cache interference in a multi-core processor. That is, one virtual machine could degrade the performance or obtain the load information of another virtual machine, which running on a same physical machine. Then we present a time-sensitive contention management approach (TSAC) for allocating resources dynamically in the virtual machine monitor, in which virtual machines are controlled to share some physical resources (e.g., CPU or page color) in a dynamical manner, in order to enforce isolation between the virtual machines without sacrificing performance of the virtualized system. We have implemented a working prototype based on Xen, evaluated the implemented prototype with experiments, and experimental results show that TSAC could significantly improve isolation of virtualization. Specifically, compared to the default Xen, TSAC could improve the performance of the victim virtual machine by up to about 78 percent, and perform well in blocking its cache-based load information leakage. [ABSTRACT FROM PUBLISHER]

As energy has become one of the key operating costs in running a data center and power waste commonly exists, it is essential to reduce energy inefficiency inside data centers. In this paper, we develop an innovative framework, called PowerTracer, for diagnosing energy inefficiency and saving power. Inside the framework, we first present a resource tracing method based on request tracing in multi-tier services of black boxes. Then, we propose a generalized methodology of applying a request tracing approach for energy inefficiency diagnosis and power saving in multi-tier service systems. With insights into service performance and resource consumption of individual requests, we develop (1) a bottleneck diagnosis tool that pinpoints the root causes of energy inefficiency, and (2) a power saving method that enables dynamic voltage and frequency scaling (DVFS) with online request tracing. We implement a prototype of PowerTracer, and conduct extensive experiments to validate its effectiveness. Our tool analyzes several state-of-the-practice and state-of-the-art DVFS control policies and uncovers existing energy inefficiencies. Meanwhile, the experimental results demonstrate that PowerTracer outperforms its peers in power saving. [ABSTRACT FROM PUBLISHER]

Recent cost analysis shows that the server cost still dominates the total cost of high-scale data centers or cloud systems. In this paper, we argue for a new twist on the classical resource provisioning problem: heterogeneous workloads are a fact of life in large-scale data centers, and current resource provisioning solutions do not act upon this heterogeneity. Our contributions are threefold: first, we propose a cooperative resource provisioning solution, and take advantage of differences of heterogeneous workloads so as to decrease their peak resources consumption under competitive conditions; second, for four typical heterogeneous workloads: parallel batch jobs, web servers, search engines, and MapReduce jobs, we build an agile system PhoenixCloud that enables cooperative resource provisioning; and third, we perform a comprehensive evaluation for both real and synthetic workload traces. Our experiments show that our solution could save the server cost aggressively with respect to the noncooperative solutions that are widely used in state-of-the-practice hosting data centers or cloud systems: for example, EC2, which leverages the statistical multiplexing technique, or RightScale, which roughly implements the elastic resource provisioning technique proposed in related state-of-the-art work. [ABSTRACT FROM AUTHOR]

As an important cornerstone for clouds, virtualization plays a vital role in building this emerging infrastructure. Virtual machines (VMs) with a variety of workloads may run simultaneously on a physical machine in the cloud platform. The scheduling algorithm used in Xen schedules virtual CPUs (VCPUs) of a VM asynchronously and guarantees the proportion of the CPU time allocated to the VM. This proportional sharing (PS) method is beneficial as it simplifies the implementation of CPU scheduling in the virtual machine monitor (VMM), and can deliver near-native performance for some workloads. However, when workloads in VMs are concurrent applications such as multithreaded programs with the synchronization operation, it has been demonstrated that this method in the VMM can reduce the performance, due to the negative impact of virtualization on synchronization. To address this issue, we present a hybrid scheduling framework for CPU management in the VMM to adapt to the diversity of VMs running simultaneously on a physical machine. We implement a hybrid scheduler based on Xen, and experimental results indicate that the hybrid CPU management method is feasible to mitigate the negative influence of virtualization on synchronization, and improve the performance of concurrent applications in the virtualized system, while maintaining the performance of high-throughput applications. [ABSTRACT FROM PUBLISHER]

Using virtualization techniques, a cloud administrator can consolidate, distribute, and manage computing resources through a network from a central console. Existing open source virtual machine monitors, such as Xen, do not support virtual disk snapshot and rollback, which is important for clouds, especially for virtual cloud storage and computing resource management. Although traditional snapshot technologies may be directly applied to virtual cloud storage, they are inefficient and not fully reliable. Most seriously, the techniques impose limitations on the cloud environments. In this article, we present SNPdisk (snapshot disk), a para-virtualization snapshot mechanism, which uses a novel sparse tree scheme to implement reliable and efficient snapshots. Experiments with a prototype system demonstrate its high performance; it is especially useful for private clouds. [ABSTRACT FROM PUBLISHER]

Cloud computing uses a virtual, nonlocal system for storage and computation. The creation of individual virtual machines (VMs) in the larger system is an important foundation for cloud computing, especially in the delivery of software as a service (SaaS). The biggest challenge for SaaS cloud-computing systems is maintaining user-level security in VMs. A proposed in-VM measuring framework confirms the status of user-level applications launched in guest VMs. A measurement module measures the executable and transfers the values to a trusted VM, which stores them in a measurement table. Comparing those values to a reference table containing the trusted measurement values of running executables verifies the executable's status. This framework relies on the system's Trusted Platform Module to guarantee the two tables' integrity. In addition, a memory watcher module confirms the measurement module's status. A working prototype of the framework based on the Xen VM monitor showed good efficiency. [ABSTRACT FROM PUBLISHER]