Jinhui Qin

Job co-allocation strategies for multiple high performance computing clusters

To more effectively use a network of high performance computing clusters, allocating multi-process jobs across multiple connected clusters becomes an attractive possibility. This allocation process entails dividing the processes of a job among several clusters, which we refer to as co-allocation. Co-allocation offers the possibility of more efficient use of computer resources, reduced turn-around time and computations using numbers of processes larger than processes on any single cluster. In order to realize these possibilities, effective co-allocation, ultimately, depends on the inter-cluster communication cost. In this paper, we introduce a scalable co-allocation strategy called the Maximum Bandwidth Adjacent cluster Set (MBAS) strategy. The strategy makes use of two thresholds to control allocation: one to control the limit on bandwidth on usable inter-cluster communication links and another to control how jobs are split. A simulator that can simulate the dynamic behavior of jobs running across multiple clusters was developed and used to examine the performance of the MBAS co-allocation strategy. Our results indicate that by adjusting the thresholds for link level control and chunk size control in splitting jobs, the MBAS co-allocation strategy can significantly improve both user satisfaction and system utilization.

Real-Time Earthquake Intensity Estimation Using Streaming Data Analysis of Social and Physical Sensors

Earthquake intensity is one of the key components of the decision-making process for disaster response and emergency services. Accurate and rapid intensity calculations can help to reduce total loss and the number of casualties after an earthquake. Modern intensity assessment procedures handle a variety of information sources, which can be divided into two main categories. The first type of data is that derived from physical sensors, such as seismographs and accelerometers, while the second type consists of data obtained from social sensors, such as witness observations of the consequences of the earthquake itself. Estimation approaches using additional data sources or that combine sources from both data types tend to increase intensity uncertainty due to human factors and inadequate procedures for temporal and spatial estimation, resulting in precision errors in both time and space. Here we present a processing approach for the real-time analysis of streams of data from both source types. The physical sensor data is acquired from the U.S. Geological Survey (USGS) seismic network in California and the social sensor data is based on Twitter user observations. First, empirical relationships between tweet rate and observed Modified Mercalli Intensity (MMI) are developed using data from the M6.0 South Napa, CAF earthquake that occurred on August 24, 2014. Second, the streams of both data types are analyzed together in simulated real-time to produce one intensity map. The second implementation is based on IBM InfoSphere Streams, a cloud platform for real-time analytics of big data. To handle large processing workloads for data from various sources, it is deployed and run on a cloud-based cluster of virtual machines. We compare the quality and evolution of intensity maps from different data sources over 10-min time intervals immediately following the earthquake. Results from the joint analysis shows that it provides more complete coverage, with better accuracy and higher resolution over a larger area than either data source alone.

Experimenter's portal: the collection, management and analysis of scientific data from remote sites

This paper describes an e-Science initiative to enable teams of scientists to run experiments with secure links at one or more advanced research facilities. The software provides a widely distributed team with a set of controls and screens via common browsers to operate, observe and record essential parts of an experiment and to access remote cloud-based analysis software to process the large data sets that are often involved in complex experiments. This paper describes the architecture of the software, the underlying web services used for remote access to research facilities and describes the cloud-based approach for data analysis. The core services are general and can be used as the basis for access to a variety of systems, though specific screen interfaces and analysis software must be tailored to a facility. For illustrative purposes, we focus on use of the system to access a single site - a synchrotron beamline at the Canadian Light Source. We conclude with a discussion of the generality and extensibility of the software and services.

Accelerated Synchrotron X-ray Diffraction Data Analysis on a Heterogeneous High Performance Computing System

The analysis of synchrotron X-ray Diffraction (XRD) data has been used by scientists and engineers to understand and predict properties of materials. However, the large volume of XRD image data and the intensive computations involved in the data analysis makes it hard for researchers to quickly reach any conclusions about the images from an experiment when using conventional XRD data analysis software. Synchrotron time is valuable and delays in XRD data analysis can impact decisions about subsequent experiments or about materials that they are investigating. In order to improve the data analysis performance, ideally to achieve near real time data analysis during an XRD experiment, we designed and implemented software for accelerated XRD data analysis. The software has been developed for a heterogeneous high performance computing (HPC) system, comprised of IBM PowerXCell 8i processors and Intel quad-core Xeon processors. This paper describes the software and reports on the improved performance. The results indicate that it is possible for XRD data to be analyzed at the rate it is being produced.

Dynamic resource matching for multi-clusters based on an ontology-fuzzy approach

One key aspect for the successful utilization of grid environments is how to efficiently schedule distributed and parallel applications to these configurations. It is also desirable to make the tlymatching operation of available resources as transparent as possible to the user. These aspects are especially important for grid environments formed by heterogeneous multi-cluster machines. In this paper, we present an approach that considers both computer resources and communication links. The approach is based on a combination of ontology and fuzzy logic. The ontology paradigm is employed as a standard interface to accept users’s requirements for desired resources. The fuzzy logic algorithms are used to compute parameters for matching based on dynamically monitored values of processor usage and communication. Experimental results indicate that the proposed approach is successful in terms of gathering dynamically more appropriate distributed resources and communication links in multi-cluster environments.

A Study on Job Co-Allocation in Multiple HPC Clusters

To more effectively use HPC clusters for even larger computations, improve turn-around times and better utilize compute resource, users are looking to interconnect multiple HPC clusters, creating a grid. To effectively use such grids, it may be desirable to split and co-allocate jobs requiring many processes across multiple clusters. While splitting a very large job across multiple clusters is an attractive possibility, the benefit, in terms of improving turn-around time, ultimately depends on the communication patterns between processes, workload on the communication links, and the maximum bandwidth of the links. The objective of this work is to understand the impact of communications on multi-processor jobs in order to develop scheduling strategies and job allocation algorithms for multi-cluster grids which can accommodate communication factors. In this paper we report on initial investigations of some co-allocation strategies. This evaluation is based on a simulator that has been implemented and validated experimentally across two HPC clusters.

An Evaluation of Communication Factors on an Adaptive Control Strategy for Job Co-allocation in Multiple HPC Clusters

To more effectively use a network of high performance computing clusters, allocating multi-process jobs across multiple connected clusters, i.e., job co-allocation, offers the possibility of more efficient use of computer resources, reduced turn-around time and computations using numbers of processes larger than processors on any single cluster. Effective co-allocation, ultimately, depends on the inter-cluster communication cost. We previously introduced a scalable co-allocation strategy – Maximum Bandwidth Adjacent cluster Set (MBAS) strategy. It made use of two thresholds to control job co-allocation – one dealing with inter-cluster links and one controlling job partitioning. We subsequently introduced the Adaptive Threshold Control System (ATCS), which used a fuzzy control approach to dynamically adjust these thresholds within MBAS. Results suggested that using ATCS during MBAS job co-allocation could achieve an overall performance improvement. However, these results only considered jobs that involved either master-slave or all-all communications among constituent processes. In this paper, we extend this analysis by also considering jobs that exhibit 2D-mesh communication patterns and evaluate ATCS further.

Big Data Challenges and Hazards Modeling

In this work we present an overview of the challenges presented by remote sensing and other big data sources for hazards modeling and response in the world today. Big data not only provides vital information for rapid and efficient assessment of the effects and impacts of natural and anthropogenic effects, but is also an important boundary object facilitating communication and interaction between the relevant scientific, business, and governmental organizations. To effectively serve that role, big data must be credible, salient, and legitimate. The characteristics of big data are examined and we conclude that the most important ones for this application are volume, velocity, variety, and value. We present two different applications from the fields of climate and the solid earth science that are designed to solve these challenges for big data science.