Javier Delgado

Cloud federation in a layered service model

We show how a layered Cloud service model of software (SaaS), platform (PaaS), and infrastructure (IaaS) leverages multiple independent Clouds by creating a federation among the providers. The layered architecture leads naturally to a design in which inter-Cloud federation takes place at each service layer, mediated by a broker specific to the concerns of the parties at that layer. Federation increases consumer value for and facilitates providing IT services as a commodity. This business model for the Cloud is consistent with broker mediated supply and service delivery chains in other commodity sectors such as finance and manufacturing. Concreteness is added to the federated Cloud model by considering how it works in delivering the Weather Research and Forecasting service (WRF) as SaaS using PaaS and IaaS support. WRF is used to illustrate the concepts of delegation and federation, the translation of service requirements between service layers, and inter-Cloud broker functions needed to achieve federation.

Transparent grid enablement of weather research and forecasting

The impact of hurricanes is so devastating throughout different levels of society that there is a pressing need to provide a range of users with accurate and timely information that can enable effective planning for and response to potential hurricane landfalls. The Weather Research and Forecasting (WRF) code is the latest numerical model that has been adopted by meteorological services worldwide. The current version of WRF has not been designed to scale out of a single organizations local computing resources. However, the high resource requirements of WRF for fine-resolution and ensemble forecasting demand a large number of computing nodes, which typically cannot be found within one organization. Therefore, there is a pressing need for the Grid-enablement of the WRF code such that it can utilize resources available in partner organizations. In this paper, we present our research on Grid enablement of WRF by leveraging our work in transparent shaping, GRID superscalar, profiling, code inspection, code modeling, meta-scheduling, and job flow management.

Efficiency Assessment of Parallel Workloads on Virtualized Resources

In cloud computing, virtual containers on physical resources are provisioned to requesting users. Resource providers may pack as many containers as possible onto each of their physical machines, or may pack specific types and quantities of virtual containers based on user or system QoS objectives. Such elastic provisioning schemes for resource sharing may present major challenges to scientific parallel applications that require task synchronization during execution. Such elastic schemes may also inadvertently lower utilization of computing resources. In this paper, we describe the elasticity constraint effect and ripple effect that cause a negative impact to application response time and system utilization. We quantify the impact using real workload traces through simulation. Then, we demonstrate that some resource scheduling techniques can be effective in mitigating the impacts. We find that a tradeoff is needed among the elasticity of virtual containers, the complexity of scheduling algorithms, and the response time of applications.

Performance prediction of weather forecasting software on multicore systems

Performance prediction is valuable in different areas of computing. The popularity of lease-based access to high performance computing resources particularly benefits from accurate performance prediction. Most contemporary processors are employing multiple computing cores, which complicates the task of performance prediction. In this paper, we describe the methodology employed for predicting the performance of a popular weather forecasting application on systems with between 4 and 256 processors. An average prediction error of less than 10% was achieved after testing on three different multi-node, multicore systems.

Paravirtualization for Scientific Computing: Performance Analysis and Prediction

Resource virtualization technologies have recently increased in popularity. The emergence of cloud computing, which requires provisioning isolated environments on shared resources, is one reason for this. Virtualization adds flexibility in terms of resource provisioning, but it can impact application performance. In this work, we analyze the performance of medical image processing and computational fluid dynamics applications when run on virtualized resources. We then apply the observed performance characteristics to a performance prediction model. We measure the impact of virtualization by performing several benchmarks on virtualized and non-virtualized resources. We evaluate the accuracy of the performance prediction model in this environment. We find that virtualization can slow down some applications by more than 200%, but usually the performance impact is below 15%. The overhead itself is predictable if general application characteristics are known. Execution time in a virtual environment can be predicted to within 13% using a simple mathematical prediction model.

New Metrics for Scheduling Jobs on Cluster of Virtual Machines

As the virtualization of resources becomes popular, the scheduling problem of batch jobs on virtual machines requires new approaches. The dynamic and sharing aspects of virtual machines introduce unique challenges and complexity for the scheduling problems of batch jobs. In this paper, we propose a new set of metrics, called potential capacity (PC) and equilibrium capacity (EC), of resources that incorporate these dynamic, elastic, and sharing aspects of co-located virtual machines. We then show that we mesh this set of metrics smoothly into traditional scheduling algorithms. We evaluate the performance in using the metrics in a widely used greedy scheduling algorithm and show that the new scheduler improves job speedup for various configurations when compared to a similar algorithm using traditional physical machine metrics such as available CPU capacity.

A learning and collaboration platform based on SAGE

In this paper, we describe the use of a tiled-display wall platform for use as a general purpose collaboration and learning platform. The main scenario of emphasis for this work is online learning by users in different countries. We describe the general efficacy of this platform for our purposes and describe its shortcomings for this purpose empirically. We discuss its advantages and also the shortcomings that we found. We also describe an enhancement made to make it more viable for our target usage scenario by implementing an interface for a modern human interface device.

Towards an Efficient and Extensible Grid-Based Data Storage Solution

The emergence of data-intensive applications in medical and other fields of study has created the need for increased storage space. It is often the case that these data should not or cannot be placed in the same domain. The data usually requires considerable computational power to process, which has inspired the area of Grid computing. This study proposes a new methodology for sharing storage volumes across distributed domains securely and making them accessible to all members of a grid. The proposed system was designed with the intent to integrate with the Globus Toolkit, which gives it the advantage of having useful features that are commonly necessary for data intensive applications.

Improving the Scalability of a Hurricane Forecast System in Mixed-Parallel Environments

The Hurricane Weather Research and Forecasting (HWRF) model is one of the premier models in NOAAs operational suite of severe weather forecasting systems. An axiom in numerical weather prediction suggests that modeling the environment at high resolution optimizes forecast accuracy. However, due to operational time constraints, only the region immediately surrounding a single hurricane can be modeled in high resolution. Currently, this is achieved by embedding a relatively small high resolution, storm-following pair of grids within a larger and coarser grid. In a previous work, we extended HWRF to support multiple such independent storm-following pair of grids. The result was improved forecast accuracy by virtue of modeling storm-to-storm interactions in high resolution. However, some shortcomings in the underlying WRF framework cause these independent pairs of grids to be simulated sequentially. This limits the models scalability and makes it impossible to harness this novel capability within the operational time constraints. In this paper, we address this issue by modifying the underlying WRF framework to simulate these independent pairs of storm-following grids in parallel. This is the first approach to be successfully implemented in the history of the WRF framework.

On the efficacy of present grid computing software for deploying a medical grid

Grid computing promises improvements in collaboration. This includes sharing of computational resources as well as improved collaboration amongst professionals of different areas of expertise. Several mature software applications are available for simplifying the deployment of an arbitrary grid. In this work, we share our experience with some of these applications for collaboration between a consortium of hospitals and our research lab, which specializes in neuroscience and image processing applications. We explain the suitability of the Grid tools through extensions and enhancements made to an existing Grid Computing Software platform and the visualization mechanisms of the display wall. This paper describes our grid computing prototype infrastructure, which uses Globus Toolkit-4 (GT4) and third-party components.