Research Roundup: Expanding the Science Cloud
This week’s hand-picked assortment focuses on advancements made to improve the performance of scientific applications in the cloud, touching on issues such as fault tolerance, workflow management, and 2D and 3D cellular simulation.
Cloud Service Fault Tolerance
Cloud computing presents a unique opportunity for science and engineering with benefits compared to traditional high-performance computing, especially for smaller compute jobs and entry-level users to parallel computing. However, according to researchers from RMIT University in Melbourne, doubts remain for production high-performance computing in the cloud, the so-called science cloud, as predictable performance, reliability and therefore costs remain elusive for many applications.
Their paper used parameterized architectural patterns to assist with fault tolerance and cost predictions for science clouds, in which a single job typically holds many virtual machines for a long time, communication can involve massive data movements, and buffered streams allow parallel processing to proceed while data transfers are still incomplete.
They utilized predictive models, simulation and actual runs to estimate run times with acceptable accuracy for two of the most common architectural patterns for data-intensive scientific computing: MapReduce and Combinational Logic. Run times were fundamental to understand fee-for-service costs of clouds.
These are typically charged by the hour and the number of compute nodes or cores used. The researchers evaluated their models using realistic cloud experiments from collaborative physics research projects and showed that proactive and reactive fault tolerance is manageable, predictable and composable, in principle, especially at the architectural level.
Cloud Computing and Cellular Automata Simulation
Cellular automata can be applied to solve several problems in a variety of areas, such as biology, chemistry, medicine, physics, astronomy, economics, and urban planning.
The automata are defined by simple rules that give rise to behavior of great complexity running on very large matrices. 2D applications may require more than 106 × 106 matrix cells, which are usually beyond the computational capacity of local clusters of computers.
A paper from Brazilian researchers out of Pontifical Catholic University of Rio de Janeiro and the Federal University of Espirito Santo presented a solution for traditional cellular automata simulations. They proposed a scalable software framework, based on cloud computing technology, which is capable of dealing with very large matrices.
The use of the framework facilitated the instrumentation of simulation experiments by non-computer experts, as it removed the burden related to the configuration of MapReduce jobs, so that researchers need only be concerned with their simulation algorithms.
Managing Computational Workflows in the Cloud
Scientists today are exploring the use of new tools and computing platforms to do their science. They are using workflow management tools to describe and manage complex applications and are evaluating the features and performance of clouds to see if they meet their computational needs, argue researchers out of the USC Information Sciences Institute.
Although today, hosting is limited to providing virtual resources and simple services, one can imagine that in the future entire scientific analyses will be hosted for the user. The latter would specify the desired analysis, the timeframe of the computation, and the available budget.
Hosted services would then deliver the desired results within the provided constraints. Their paper described current work on managing scientific applications on the cloud, focusing on workflow management and related data management issues.
Frequently, applications are not represented by single workflows but rather as sets of related workflow ensembles. Thus, hosted services need to be able to manage entire workflow ensembles, evaluating tradeoffs between completing as many high-value ensemble members as possible and delivering results within a certain time and budget.
Their paper gives an overview of existing hosted science issues, presents the current state of the art on resource provisioning that can support it, as well as outlines future research directions in this field.
Optimizing Data Analysis in the Cloud
A research team out of Duke University presented Cumulon, a system designed to help users rapidly develop and intelligently deploy matrix-based big-data analysis programs in the cloud.
Cumulon, according to the research, features a ﬂexible execution model and new operators especially suited for such workloads. In the paper, they show how to implement Cumulon on top of Hadoop/HDFS while avoiding limitations of MapReduce, and demonstrate Cumulon’s performance advantages over existing Hadoop-based systems for statistical data analysis.
To support intelligent deployment in the cloud according to time/budget constraints, Cumulon goes beyond database style optimization to make choices automatically on not only physical operators and their parameters, but also hardware provisioning and conﬁguration settings, according to the Duke researchers.
They applied a suite of benchmarking, simulation, modeling, and search techniques to support effective cost-based optimization over this rich space of deployment plans.
Business Integration as a Service: The Case Study of the University of Southampton
Finally, a paper out of the University of Southampton presented Business Integration as a Service (BIaaS) to allow two services to work together in the Cloud to achieve a streamline process. They illustrated this integration using two services; Return on Investment (ROI) Measurement as a Service (RMaaS) and Risk Analysis as a Service (RAaaS) in the case study at the University of Southampton.
The case study demonstrated the cost-savings and the risk analysis achieved, so two services can work as a single service. Advanced techniques were used to demonstrate statistical services and 3D Visualisation services under the remit of RMaaS and Monte Carlo Simulation as a Service behind the design of RAaaS.
Computational results were presented with their implications discussed. Different types of risks associated with Cloud adoption can be calculated easily, rapidly and accurately with the use of BIaaS. This case study confirmed the benefits of BIaaS adoption, including cost reduction and improvements in efficiency and risk analysis. Implementation of BIaaS in other organisations is also discussed.
Important data arising from the integration of RMaaS and RAaaS are useful for management and stakeholders of University of Southampton.