Reginald Cushing

Distributed Computing on an Ensemble of Browsers

In this article, the authors propose a new approach to distributed computing with Web browsers and introduce the WeevilScout prototype framework. The proliferation of Web browsers and the performance gains being achieved by current JavaScript virtual machines raises the question whether Internet browsers can become yet another middleware for distributed computing. With 2 billion users online, computing through Internet browsers has the potential to amass immense resources, thus transforming the Internet into a distributed computer ideal for common classes of distributed scientific applications such as parametric studies. As a proof of concept, the authors demonstrate how a cluster of globally distributed Internet browsers is used to compute thousands of bio-informatics tasks.

Enabling Web Services to Consume and Produce Large Datasets

Service-oriented architectures and Web services are well-established paradigms for developing distributed applications. However, Web services face problems when accessing, moving, and processing large datasets. To address this problem, the authors present ProxyWS, which uses myriad protocols to transport large amounts of data. ProxyWS undertakes data transfers on behalf of legacy Web services and can serve as an interface for developing new Web services that can stream data. Experiments show how this approach facilitates scalable data transports for two data-intensive applications.

Applying workflow as a service paradigm to application farming

Task farming is often used to enable parameter sweep for exploration of large sets of initial conditions for large scale complex simulations. Such applications occur very often in life sciences. Available solutions enable to perform parameter sweep by creating multiple job submissions with different parameters. This paper presents an approach to farm workflows, employing service oriented paradigms using the WS‐VLAM workflow manager, which provides ways to create, control, and monitor workflow applications and their components. We present two service‐oriented approaches for workflow farming: task level, whereby task harness acts as services by being invoked on which task to load, and data level, where the actual task is invoked as a service with different chunks of data to process. An experimental evaluation of the presented solution is performed with a biomedical application for which 3000 simulations were required to perform a Monte Carlo study. Copyright

Cloud Data Federation for Scientific Applications

Nowadays, data-intensive scientific research needs storage capabilities that enable efficient data sharing. This is of great importance for many scientific domains such as the Virtual Physiological Human. In this paper, we introduce a solution that federates a variety of systems ranging from file servers to more sophisticated ones used in clouds or grids. Our solution follows a client-centric approach that loosely couples a variety of data resources that may use different technologies such as Openstack-Swift, iRODS, GridFTP, and may be geographically distributed. It is implemented as a lightweight service which does not require installation of a software on the resources it uses. In this way we are able to efficiently use heterogeneous storage resources, reduce the usage complexity of multiple storage resources, and avoid vendor lock-in in case of cloud storage. To demonstrate the usability of our approach we performed a number of experiments that assess the performance and functionality of the developed system.

Automata-Based Dynamic Data Processing for Clouds

Big Data is a challenge in many dimensions one of which is its processing. The often complicated and lengthy processing requires specialized programming paradigms to distribute and scale the computing power. Such paradigms often focus on ordering tasks to fit an underlying architecture. In this paper, we propose a new and complementary way of reasoning about data processing by describing complex data processing as set of state transitions, specifically, a non-deterministic finite automata which captures the essence of complex data processing. Through a P2P implementation of this model we demonstrate the dynamism, intrinsic scalability and adeptness to Cloud architectures.

Towards a data processing plane

Data processing complexity, partitionability, locality and provenance play a crucial role in the effectiveness of distributed data processing. Dynamics in data processing necessitates effective modeling which allows the understanding and reasoning of the fluidity of data processing. Through virtualization, resources have become scattered, heterogeneous, and dynamic in performance and networking. In this paper, we propose a new distributed data processing model based on automata where data processing is modeled as state transformations. This approach falls within a category of declarative concurrent paradigms which are fundamentally different than imperative approaches in that communication and function order are not explicitly modeled. This allows an abstraction of concurrency and thus suited for distributed systems. Automata give us a way to formally describe data processing independent from underlying processes while also providing routing information to route data based on its current state in a P2P fashion around networks of distributed processing nodes. Through an implementation, named Pumpkin, of the model we capture the automata schema and routing table into a data processing protocol and show how globally distributed resources can be brought together in a collaborative way to form a processing plane where data objects are self-routable on the plane. New distributed data processing paradigm that describes transformations as automata.Automata provide a schema of data processing and also facilitate a P2P routing.Protocol for data processing with automata, data, code and state as part of packet.Data packet control (parallel, coalesce, backlog) is based on compute predictions.

Beyond Scientific Workflows: Networked Open Processes

The multitude of scientific services and processes being developed brings about challenges for future in silico distributed experiments. Choosing the correct service from an expanding body of processes means that the the task of manually building workflows is becoming untenable. In this paper we propose a framework to tackle the future of scientific collaborative distributed computing. We introduce the notion of Networked Open Processes whereby processes are exposed, published, and linked using semantics in the same way as is done with Linked Open Data. As part of the framework we introduce several novel concepts including Process Object Identifiers, Semantic Function Templates, and TReQL, a SQL-like language for querying networked open process graphs.

Workflow as a service: an approach to workflow farming

Task farming is often used to enable parameter sweep for exploration of large sets of initial conditions for large scale complex simulations. Available solutions enable to perform parameter sweep by creating multiple grid job submissions with different parameters. This paper presents an approach to farm grid workflows with a web service based workflow management system, called WS-VLAM, which provides ways to create, control and monitor workflows applications and their components. An approach to balance the workload between the workflow instances and to collect provenance data at both system and workflow levels is also discussed. Finally, an experimental evaluation of the presented solution is performed with a biomedical application for which 3000 simulations were required to perform a Monte-Carlo study.

Service level management for executable papers

Reproducibility of Science is considered as one of the main principles of the scientific method, and refers to the ability of an experiment to be accurately reproduced, by third person, in complex experiment every detail matters to ensure the correct reproducibility. In the context of the ICCS 2011, Elsevier organized the executable paper grand challenge a contest to improve the way scientific information is communicated and used. While during this contest the focus was on developing methods and technique to realize the idea of executable papers, in this paper we focus on the operational issues related to the creation a viable service with a predefined QoS.

Car2x with software defined networks, network functions virtualization and supercomputers technical and scientific preparations for the Amsterdam Arena telecoms fieldlab

In the invited talk “Car2x with SDN, NFV and supercomputers” we report about how our past work with SDN [1, 2] allows the design of a smart mobility fieldlab in the huge parking lot the Amsterdam Arena. We explain how we can engineer and test software that handle the complex conditions of the Car2X case. The talk starts by describing the engineering challenges that developers of smart car telecommunications and computing infrastructures face. We concentrate on the development of software defined networks (SDN) that support smart cars optimally and securely over a heterogeneous, dynamic and developing ICT infrastructure. The goal here is to enable smart cars to profit maximally from any bit of information available from fixed and moving objects as well as persons. For example, in a low tech situation, the lack of other options makes that one has to fall back to GPRS to download only traffic jam locations. As a contrast, in an advanced telecom environment video streams from multiple cars are transmitted via 5G pico cells to computers a few millisecond nearby. These computers fuse the video information to generate a local traffic model. In an ultimate situation, cars use all the communications infrastructures that are available including the numerous WiFi hotspots, all generations of mobile telecommunications, the developing car-to-car communications technologies and even the smart phone of a passing person. SDN technologies deal with the complexities of such communication environment.