以下は負荷予測手法に関する関連論文3本。
Trace-based evaluation of job runtime and queue wait time predictions in gridshttp://portal.acm.org/citation.cfm?id=1551609.1551632
Large-scale distributed computing systems such as grids are serving a growing number of scientists. These environments bring about not only the advantages of an economy of scale, but also the challenges of resource and workload heterogeneity. A consequence of these two forms of heterogeneity is that job runtimes and queue wait times are highly variable, which generally reduces system performance and makes grids difficult to use by the common scientist. Predicting job runtimes and queue wait times have been widely studied for parallel environments. However, there is no detailed investigation on how the proposed prediction methods perform in grids, whose resource structure and workload characteristics are very different from those in parallel systems. In this paper, we assess the performance and benefit of predicting job runtimes and queue wait times in grids based on traces gathered from various research and production grid environments. First, we evaluate the performance of simple yet widely used time series prediction methods and the effect of applying them to different types of job classes (e.g., all jobs submitted by single users or to single sites). Then, we investigate the performance of two kinds of queue wait time prediction methods for grids. Last, we investigate whether prediction-based grid-level scheduling policies can have better performance than policies that do not use predictions.
Swift: Fast, Reliable, Loosely Coupled Parallel Computationhttp://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.92.519A common pattern in scientific computing involves the execution of many tasks that are coupled only in the sense that the output of one may be passed as input to one or more others—for example, as a file, or via a Web Services invocation. While such “loosely coupled” computations can involve large amounts of computation and communication, the concerns of the programmer tend to be different than in traditional high performance computing, being focused on management issues relating to the large numbers of datasets and tasks (and often, the complexities inherent in “messy ” data organizations) rather than the optimization of interprocessor communication. To address these concerns, we have developed Swift, a system that combines a novel scripting language called SwiftScript with a powerful runtime system based on CoG Karajan and Falkon to allow for the concise specification, and reliable and efficient execution, of large loosely coupled computations. Swift adopts and adapts ideas first explored in the GriPhyN virtual data system, improving on that system in many regards. We describe the SwiftScript language and its use of XDTM to describe the logical structure of complex file system structures. We also present the Swift system and its use of CoG Karajan, Falkon, and Globus services to dispatch and manage the execution of many tasks in different execution environments. We summarize application experiences and detail performance experiments that quantify the cost of Swift operations. 1.
Delay scheduling: a simple technique for achieving locality and fairness in cluster schedulinghttp://portal.acm.org/citation.cfm?id=1755940As organizations start to use data-intensive cluster computing systems like Hadoop and Dryad for more applications, there is a growing need to share clusters between users. However, there is a conflict between fairness in scheduling and data locality (placing tasks on nodes that contain their input data). We illustrate this problem through our experience designing a fair scheduler for a 600-node Hadoop cluster at Facebook. To address the conflict between locality and fairness, we propose a simple algorithm called delay scheduling: when the job that should be scheduled next according to fairness cannot launch a local task, it waits for a small amount of time, letting other jobs launch tasks instead. We find that delay scheduling achieves nearly optimal data locality in a variety of workloads and can increase throughput by up to 2x while preserving fairness. In addition, the simplicity of delay scheduling makes it applicable under a wide variety of scheduling policies beyond fair sharing.
Generating Adaptation Policies for Multi-tier Applications in Consolidated Server Environmentshttp://portal.acm.org/citation.cfm?id=1444674Creating good adaptation policies is critical to building complex autonomic systems since it is such policies that define the system configuration used in any given situation. While online approaches based on control theory and rule-based expert systems are possible solutions, each has its disadvantages. Here, a hybrid approach is described that uses modeling and optimization offline to generate suitable configurations, which are then encoded as policies that are used at runtime. The approach is demonstrated on the problem of providing dynamic management in virtualized consolidated server environments that host multiple multi-tier applications. Contributions include layered queuing models for Xen-based virtual machine environments, a novel optimization technique that uses a combination of bin packing and gradient search, and experimental results that show that automatic offline policy generation is viable and can be accurate even with modest computational effort.
