e-Infrastructure for Remote Instrumentation

Computer Standards & Interfaces 34 (2012) 476–484

Contents lists available at SciVerse ScienceDirect

Computer Standards & Interfaces journal homepage: www.elsevier.com/locate/csi

e-Infrastructure for Remote Instrumentation Alexey Cheptsov a,⁎, Bastian Koller a, Davide Adami b, Franco Davoli b, Szymon Mueller c, Norbert Meyer c, Paolo Lazzari d, Stefano Salon d, Johannes Watzl e, Michael Schiffers e, Dieter Kranzlmueller e a

High Performance Computing Center Stuttgart (HLRS), D-70550 Stuttgart, Germany CNIT-University of Genoa/University of Pisa Research Units, Via Opera Pia 13, I-16145 Genova, Italy Poznań Supercomputing and Networking Center, 61-704 Poznań, ul. Noskowskiego 10, Poland d Dept. of Oceanography, Istituto Nazionale di Oceanografia e di Geofisica Sperimentale (OGS), B.go Grotta Gigante 42/c, 34010 Sgonico (TS), Italy e MNM-Team, Ludwig-Maximilians-Universität München, Oettingenstr. 67, D-80538 München, Germany b c

a r t i c l e

i n f o

Available online 30 October 2011 Keywords: Remote instrumentation e-Science Grid e-Infrastructure Quality of service

a b s t r a c t Despite the tremendous growth of the capacity of computation and storage IT solutions over the last years there is still a deep mismatch between the e-Infrastructures and the e-Science applications that use instruments, sensors, and laboratory equipment. The efficiency of using instruments in a remote way, i.e. Remote Instrumentation, might be largely improved by integration with the existing distributed computing and storage infrastructures, like Grids. The paper discusses major activities towards the e-Infrastructure for Remote Instrumentation – a Grid-based Information and Communication Technology environment capable of covering all the issues arising around enabling Remote Instrumentation for e-Science applications. © 2011 Elsevier B.V. All rights reserved.

1. Introduction In recent years the progress of high-performance computing and networking has enabled the deployment of large-scale infrastructures, like the ones promoted by the OSG 1 (the Open Science Grid) project in the USA, NAREGI (the Japanese National Research Grid Initiative) in Japan, or EGEE 2 (Enabling Grids for E-sciencE) and DEISA 3 (Distributed European Infrastructure for Supercomputing Applications) in the European Research Area. These foundations set up infrastructures which provide powerful distributed computing

⁎ Corresponding author. E-mail addresses: [email protected] (A. Cheptsov), [email protected] (B. Koller), [email protected] (D. Adami), [email protected] (F. Davoli), [email protected] (S. Mueller), [email protected] (N. Meyer), [email protected] (P. Lazzari), [email protected] (S. Salon), [email protected] (J. Watzl), [email protected]fi.lmu.de (M. Schiffers), kranzlmueller@ifi.lmu.de (D. Kranzlmueller). URL's: http://www.hlrs.de/ (A. Cheptsov, B. Koller), http://www.dist.unige.it/ (F. Davoli, D. Adami), http://www.man.poznan.pl/ (S. Mueller, N. Meyer), http:// www.inogs.it/ (S. Salon, P. Lazzari), http://www.mnm-team.org/ (J. Watzl, M. Schiffers, D. Kranzmueller). 1 Open Science Grid (OSG) project website, http://opensciencegrid.org/. 2 Enabling Grids for E-Science (EGEE) project website, http://www.eu-egee.org/. 3 Distributed European Infrastructure for Supercomputing Applications (DEISA) project website, http://www.deisa.eu/. 0920-5489/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.csi.2011.10.012

environments with thousands of CPU cores and petabytes of storage space available for complex applications. However, as the traditional Grid matured, the main interest is being shifted towards the real sources of the processed data — instruments and sensors. Instruments which produce, collect, and acquire large data volumes are widely used in science and technology (Fig. 1). Prominent examples of such instruments are Glider — the remotely controlled mobile device, which performs long-distance oceanographic measurement missions (Fig. 1a), synchrotron — a large-scale particle accelerator and data facility (Fig. 1c), or even any micro device that is part of a greater sensor network, such as a camera network for observation of large coastal areas (Fig. 1b). Development and spreading of the Remote Instrumentation concept — a solution for accessing and controlling distributed scientific instruments from within scientific applications — opens new opportunities for many scientific communities [1]. In particular, environmental science, earthquake and experimental physics applications will benefit from it. Remote Instrumentation is also gaining significant popularity in education. Being provided within a Grid supercomputing environment, Remote Instrumentation offers a solution not only for getting access to instruments (often including unique and expensive laboratory equipment), but also for sharing, federation and exploitation of the collective power of high-performance (computing and storage resources) facilities for academic and industrial research virtual communities.

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

477

Fig. 1. Examples of remote instruments: a) oceanographic device “Glider” b) car sensor network c) synchrotron of the ELETTRA lab.4

The recent attempts (performed in the projects RINGrid 5, GRIDCC 6, and CIMA 7) which strived to design a service-oriented Grid infrastructure for management, maintenance, and exploitation of heterogeneous instrumentation and acquisition devices, in conjunction with computing and storage resources of the traditional Grid, resulted in an e-Infrastructure for Remote Instrumentation, which has been set up in the frame of the EU-funded DORII 8 project. Thus, the e-Infrastructure facilitates full-service operational support of the experimental and laboratory equipment plugged in to the Grid. The paper addresses the main objectives of the e-Infrastructure for Remote Instrumentation's deployment and introduces the state-of-the-art solutions for hardware, middleware and networking enhancements of the traditional Grid towards the integration of instruments and sensors. Section 1 presents the conception of the e-Infrastructure and its basic deployment aspects. Section 2 introduces some pilot Remote Instrumentation application scenarios that require Grid support. Section 3 provides some information about the DORII project's infrastructure. Section 4 describes the middleware architecture of the Remote Instrumentation eInfrastructure, which allows users and their applications to get an easy and secure access to various Remote Instrumentation resources and benefit from the high-performance computing and storage facilities of the traditional Grid. Section 5 addresses the networking aspect of the e-Infrastructure. Section 6 collects the main innovations done in DORII. Section 7 contains the conclusion and final discussion points. 2. Concept of Remote Instrumentation Infrastructure Issues of remote access and operation of diverse remote instruments and experimental equipment have been particularly studied and developed in the framework of several research projects, 4

Synchrotron Light Laboratory ELETTRA, http://www.elettra.trieste.it/ Remote Instrumentation in Next-Generation Grids (RINGrid) — www.ringrid.eu. 6 Grid enabled Remote Instrumentation with Distributed Control and Computation (GRIDCC) — www.gridcc.org. 7 Common Instrument Middleware Architecture (CIMA) — http://plone.jcu.edu.au/ dart/software/cima. 8 Deployment of the R.emote Instrumentation Infrastructure (DORII) — www. dorii.eu 5

mentioned above. However, the efficient use of Remote Instrumentation goes far beyond facilitating networked access to remote instruments, which the most those projects have focused on. Grid services for Remote Instrumentation should offer a solution to fully integrate instruments (including laboratory equipment, large-scale experimental facilities, and sensor networks) in a Service Oriented Architecture (SOA), where users can operate instruments in the same fashion as the computing and storage resources offered by the traditional Grid (Fig. 2). Practical attempts to close numerous gaps between the Grid and scientific domains which provide and utilize Remote Instrumentation Services were performed within the DORII project [2]. Among the main goals addressed by the project the following are of special interest for the e-Infrastructure for Remote Instrumentation: • to provide a set of standard capabilities to support Remote Instrumentation in a Grid environment, including a suitable abstraction of Remote Instrumentation, in order to make it visible as a manageable Grid resource; • to adopt Remote Instrumentation across e-Science domains; • to design a service-oriented Grid architecture that enables the integration of instruments as services; • to set up a flexible problem-oriented middleware architecture, which not only provides users with services for Remote Instrumentation, but also enables full-fledged user-level tools for promoting the e-Infrastructure to the end-user and application developers, hiding the complexity of the underlying Grid technology; • to generalize and deploy a framework environment that can be used for fast prototyping. The architecture proposed by DORII provides an easy and secure access to Remote Instrumentation resources, supported by highperformance computing and storage facilities of the traditional Grid as well as by underlying networking technologies. The following sections address the main elements of DORII e-Infrastructure — applications, infrastructure resources, middleware, and networking facilities. 3. Pilot application scenarios Below we describe the e-Science domains that are of special interest for the e-Infrastructure, and present some pilot applications adopted by

478

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

Fig. 2. Conceptual view of the Remote Instrumentation e-Infrastructure. The instrument element (IE) becomes the standard element abstraction of the Grid, in addition to computing (CE) and storage (SE) resources.

DORII. The scenarios present only some of many examples of scientific instruments which are utilized in e-Science and might greatly benefit from Remote Instrumentation. 3.1. Environmental science Environmental domain spans a significant range of e-Science and offers a number of instruments, such as Glider (Fig. 1a). Glider is an oceanographic device performing long-term and long-distance measurement missions, used in operational oceanography (environmental science domain) [7]. Measurements of physical, chemical, and biological parameters are the basis for research and development in oceanography. Decades ago it was still necessary to venture with ships to the location of interest and perform the measurements manually. This was costly in time, personnel, and funds. The first instruments which automate registration of the necessary information were deployed in ocean moorings since the 1950s. They can conduct measurements for periods up to three years. The measured data become available to the scientists only after the recovery of the instruments. In the 1970s satellites caused another jump in observation systems. Large parts of the world's oceans could be covered in a short time and the data became available to oceanographers only minutes or hours after the measurement. Unfortunately not all interesting parameters can be observed from a satellite. Especially data from the interior of the oceans is inaccessible to satellites. In the past decade satellite phones, energy saving micro electronics, and new materials lead to a new generation of instruments. These instruments are basically autonomous, some of them drift with ocean currents or move along predetermined courses, and transmit their data in

regular intervals via satellite to the scientist on land. Glider is the youngest member in the family of oceanographic instruments. This is an autonomously diving and slowly moving platform for measurements. Glider can travel for several months over distances of several thousand kilometers and continuously perform measurements along their characteristic zigzag dive-path [3], see Fig. 3. A whole network of oceanographic scientific instruments, including a set of Gliders, is currently deployed in the Mediterranean Basin in the frame of MERSEA project [4]. In particular, they are used for near real-time observations at the sea surface and in the water column, producing such important characteristics, as temperature, salinity, and pressure profiles etc. This tasks is facilitated by means of special numerical simulation models, such as OPATM-BFM, developed by an oceanographic institute from Trieste (Istituto Nazionale di Oceanografia e di Geofisica Sperimentale, OGS) and described in [5]. OPATM-BFM is a parallel, coupled physical–biogeochemical model that that produces shortterm forecasts of some key biogeochemical variables (e.g. chlorophyll) for the Mediterranean Sea. Physical forcing fields (i.e. current velocity), used by OPATM-BFM, are indirectly coming from Float measurements. Each of the described parts of the oceanographic scenario (collecting the data by instruments and performing complex simulation on those data by applications) poses several challenging scenarios for integration with European as well as worldwide Grid infrastructures. Moreover, a great challenge is to compose the abovementioned analysis phases in a common operation chain, a workflow which links the remote measurement with the computational outputs, including all the steps of the data downloading, treatment and post-processing, as well as running simulation model and results evaluation.

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

479

Fig. 3. Remotely instrumented Glider missions.

3.2. Experimental physics Experimental stations in facilities like Synchrotrons and Free Electron Lasers of ELETTRA 9 produce huge quantities of data [6]. These data need to be analyzed on-line, which requires considerable computing power and often teamwork. The problem is even more difficult considering the increased efficiency of the light sources and detectors. Complex calculations are required to take diffraction images and convert them into a 3D protein structure. Similarly, complex calculations are required to produce tomograms and then perform an analysis of the results. The results of these analyses often need to be visualized by a distributed team and used to modify interactively the data collection strategy. Data from instruments and sensors are saved in distributed repositories, computational models are executed, and an interactive data mining process is eventually used to extract useful knowledge. This kind of application requires both the support of a standard Grid computing environment, i.e. a Virtual Organization (VO), a set of distributed storage and computing resources and some resource brokering mechanism, a workflow definition and execution environment, and the capability to integrate instruments (the detectors) and interactively collaborate in the data analysis process. A QoS handling mechanism is necessary to use the available network structure effectively. This poses a great challenge to efficient use of the application using facilitates provided by the traditional Grid.

laboratory and equipped with actuators (to apply forces or displacements) and sensors (to measure reactions). The simulation server collects the data provided by the sensors and the calculated response of the virtual building components, putting all together in order to represent this set as a unique structure. All the applications require both the support of a standard Grid computing environment, which is a virtual organization, a set of distributed storage and computing resources and some resource brokering mechanism, a workflow definition and execution environment and the capability to integrate instruments (the detectors) and interactively collaborate in the data analysis process. Such systems also take advantage of Remote Instrumentation in terms of access to computational capabilities for speeding up the calculation of shake maps. In particular, fast shake maps are very useful for damage assessment in a post-seismic scenario, when it is necessary to coordinate in a safe and quick way rescue team operations. A network of seismic sensors should be deployed and connected by means of a wireless connection to a Grid infrastructure. In the presence of an earthquake, all the typical seismic parameters (epicentre, magnitude, ground acceleration, etc.) are estimated and then used to build fragility curves. In the easiest implementation, the application has only to perform an interpolation accessing a database of use cases already calculated (with a non-trivial computing effort, simplified by the Grid), in order to fit the current situation. In the other case, the map is calculated immediately after the earthquake parameters are recorded.

3.3. Earthquake simulation Sensor networks are widely utilized in many industrial and civilian application areas, including industrial process monitoring and control, detection and response to natural disasters and many others. For example, EUCENTRE10 has developed a number of applications which perform pseudo-dynamic simulations using sub-structuring [6]. This means that a part of the building being simulated is a “virtual” structure, while another part is a physical specimen placed in a

9

Elettra Synchrotron Light Source — www.elettra.trieste.it. European Centre for Training and Research in Earthquake Engineering (EUCENTRE) — www.eucentre.it. 10

4. Infrastructure In order to facilitate access to scientific instruments for the goals of Remote Instrumentation, heterogeneous and geographically dispersed computing, storage, and instrument resources need to be loosely coupled in a common infrastructure. Along with the computing resources used for post-processing of the data acquired from instruments, as well as for performing simulations based on them, the storage resources are an essential part of the infrastructure; they are used for storing massive amounts of data, both acquired from the instruments and produced on the computing resources. Besides, a security and management infrastructure is required for all of the resource types.

480

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

For what concerns computing and storage capabilities, the eInfrastructure can greatly utilize resource sites of the already existing Grid system, ensuring their sustainable utilization. The same approach was also followed in DORII, whose infrastructure is comprised of the resource centers included in the largest European Grid infrastructure — EGEE (Fig. 4). Along with computing and storage resources, instruments constitute an essential part of the infrastructure. As stated above, the eInfrastructure is not limited to any specific kind of instruments. As the experience of DORII reveals, diverse instruments can be interconnected with the e-Infrastructure [3]. In particular, there have been pilot applications from environmental science, earthquake prediction, and experimental physics domains integrated with the e-Infrastructure. The applications span over a significant range of instruments, including mobile remotely controlled devices as Floats/Gliders or big stationary facilities as Synchrotron or Free Electron Laser. The technical characteristics of those instruments greatly differ from Actuators and Seismic Sensors operated by means of Wireless Sensor Networks (WSN), or Conductivity, Temperature, and Depth (CTD) as well as optical sensors with strict Quality of Service (QoS) constraints. 5. Middleware architecture The tremendously large number and the heterogeneous potential of Grid resources requires a significant development effort in the middleware providing the e-Infrastructure to the various users in research and production. The basic elements of the Grid middleware are low-level web services for accessing and operating the infrastructure, as for example provided by gLite [4]— a middleware stack offered by the EGEE-Grid. gLite provides a set of Information (BDII), Job Management (CE, WMS), Data Management (SE, LFC), and Security (VOMS, MyProxy) Services. Those services are complemented in

DORII with a new group — Instrument Services, which facilitates interconnection of instruments/sensors with the infrastructure. Unlike a standard Grid middleware, the e-Infrastructure middleware aims at the end-users — experts of the application domains that the e-Infrastructure is serving. Hence, in addition to the basic middleware services, the e-Infrastructure must additionally provide: • a full-featured GUI toolkit(analog of the control room used in the scientific application domains) that enables easy access to and discovery of Grid resources and services, and serves a rich end-user collaborative suite; • a workflow management and monitoring system for setup and management of complex scientific experiments performed on the infrastructure; • an integrated application development environment with enhanced “on-Grid” debugging and deployment features; • a visualization and analysis back-end for the data produced by applications in the infrastructure; • parallel applications implementation libraries, with integrated performance analysis possibilities and back-ends for visualization of performance characteristics. By setting up the middleware architecture in conformance with the requests of Remote Instrumentation, the e-Infrastructure can beneficially reuse sustainable outcomes of the projects dealing with the above-listed topics. The middleware architecture for the Remote Instrumentation Infrastructure, as elaborated in DORII, is presented in Fig. 5. The core of the Instrument Services is formed by the Instrument Element (IE) [5]— a middleware which represents a virtualization of diverse data sources and provides the traditional Grid with an abstraction of a real instrument/sensor, and a Grid user with an interactive interface to remotely control an instrument. The IE greatly extends the

Fig. 4. Geographical location of the resource centers included to the DORII infrastructure. In total, there are 9 sites offering computational and storage resources of diverse capacities and architectural design (6 sites in Greece, 1 site in Poland, 1 site in Spain, and 1 site in Italy).

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

Fig. 5. Middleware architecture for the Remote Instrumentation Infrastructure. The core Grid services (CE, SE etc.) are extended by the Instrument Element (IE) — a middleware abstraction for interconnecting an instrument with the infrastructure. Moreover, a whole bunch of user-level tools is set up on the core Grid services, including a user and developer front-end Virtual Control Room (VCR), a Grid-enabled application development environment (g-Eclipse), a workflow management system (WfMS), solutions for visualization and interactivity for scientific applications (GVid and i2glogin), etc.

functionality of the e-Infrastructure to interactively control and monitor the remote instruments both on the part of the users and from within the applications. The user-level middleware makes up a considerable part of the architecture. It targets at providing the e-Infrastructure services to the users and ensuring the efficient development, deployment and use of the applications in Grid environments. The Virtual Control Room (VCR) [6] is the central front-end for the e-Infrastructure. The VCR is a richly-featured web portal that provides an intuitive and userfriendly interface. The VCR serves a complete desktop environment for the e-Infrastructure's end-users (Fig. 6a).

481

Some of the e-Infrastructure's applications perform complex experiments described by means of workflows. In order to simplify the users' task of designing the workflows for their applications, as well as submitting the workflow to the execution and monitoring during the execution, the Workflow Management System (WfMS) [7] is included in the architecture (Fig. 6b). The WfMS-launcher is provided by the VCR, as well. To support application development activities, the architecture also includes g-Eclipse [8]— an integrated development framework for the Grid. g-Eclipse offers sustainable support for the full development cycle of the Grid applications, including code deployment, remote compiling and debugging, etc. Access from the user-level middleware components to the Grid services deployed on the infrastructure (gLite and Instrument Services) is facilitated by means of a Common Library (CL), which offers Application Program Interfaces (APIs) for unified access to the infrastructure's Grid services (gLite and IE). The CL is designed in a way that ensures the component interoperability within the architecture and simplifies the installation of the user-level middleware. To support the specific application requirements on interactivity, visualization, and to improve running parallel (e.g. utilizing the Message Passing Interface — MPI) applications on the e-Infrastructure, the following solutions are additionally included in the middleware architecture: GLogin [9] (transport layer service provider used for opening an interactive session between the infrastructure and users), GVid [10] (video streaming service intended for Grid-based visualizations in scientific applications), Open MPI [11] (communication library for MPI applications), MPI-Start (tool for improved running of the parallel applications on the infrastructure). 6. Networking aspects The design of the telecommunications network aims at improving the performance of applications from an end-user's point of view. Basically, IP-based networks only provide a best-effort packet delivery service without any performance guarantees. On the contrary, the DORII project involved a large set of applications that generate and exchange heterogeneous traffic with different requirements in terms of QoS. DORII applications have been classified into two major categories: a) applications that process pre-collected data with loose QoS requirements, since their performance could benefit from bandwidth guaran-

Fig. 6. The e-Infrastructure's front-ends: a) Virtual Control Room (VCR), b) Workflow Management System (WfMS).

482

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

tees, but QoS support is not mandatory also for in-production deployment; b) applications working on data acquired in real-time with more stringent requirements in terms of network reliability and performance guarantees, for which QoS support is highly recommended for an effective in-production deployment. The DORII network infrastructure (see Fig. 7) interconnects heterogeneous networks, characterized by various communications technologies, which include wired and wireless connections as well as a plethora of sensor networks collecting data for specific analysis and monitoring actions. More specifically, the DORII network contains: a) the Local Area Networks of the participating institutions, which include highly heterogeneous data collection parts (sensor networks, satellite links, ADSL, high-speed data transfer), b) the corresponding National Research and Education Networks (NRENs) providing access to each national research network and to the Internet, and c) GÉANT2, as the backbone interconnecting NRENs. A further issue considered in the design of the network infrastructure was represented by the intrinsic dynamic and distributed nature of DORII applications as far as the selection of Grid resources is concerned. The deployment of Remote Instrumentation applications over an e-Infrastructure typically involves a sensor network, several DORII sites (where the selected Grid resources — CE, SE — and the IE are located) and the core network that assures inter-site connectivity. Taking into account all these complexities, first, DORII applications have been deployed over a best-effort network, so as to identify their performance issues and to characterize their behavior by correlating QoE (Quality of Experience) at application level and network statistics at network level [15]. Then, an experimental test bed between

GRNET and PSNC, providing QoS support, has been deployed so as to analyze and improve the performance of a selected set of applications that require enhanced network services. An advanced monitoring platform has been deployed for the DORII network infrastructure [14]. The design and deployment of the DORII network monitoring infrastructure has been carried out so as to identify end-to-end connectivity problems, service interruptions and bottlenecks. The network monitoring platform deployed for the DORII project [12] consists of the following tools: • Smokeping, for network latency measurement • Pathload, for the estimation of the available bandwidth along a network path • SNMP-based Web applications, for monitoring network interface utilization Since each site is connected to its NREN and interconnected with other sites through the GÉANT2 network, end-to-end traffic is monitored and network statistics are collected. Moreover, part of the DORII network, including applications' sites, is IPv6-enabled. Thus, end-toend native IPv6 connectivity is available for some DORII applications and DORII Grid sites. Results of measurement on the applications' performance over the DORII networking platform are reported in [13]. 7. Main innovations in Remote Instrumentation Infrastructure The pilot version of the Remote Instrumentation Infrastructure has been successfully set up during the DORII project. The DORII research and developments made a step forward for leveraging the

Fig. 7. DORII network infrastructure and monitoring tools.

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484

traditional e-Infrastructure's resources to fully support the Remote Instrumentation Services, as conceptualized in the RINGrid project. For the majority of the existing instruments, a huge effort is required to adapt the existing systems to new users and community requirements. Based on current experiences in DORII and its predecessors, a new definition might be proposed for commonly available Grid objects, like computing and storage elements. This will be done by extending the Distributed Computing Environments, in the same matter as in case as was made for the EGEE-based eInfrastructure, with an innovative additional abstract layer, based on the Instrument Element middleware introduced in INGRID. Refinement and empowering of Instrument Element functionalities towards integration of heterogeneous instrumentation, aiming at the complete definition of an architectural element that can become a standardized middleware component, is an important research direction in further extending the Remote Instrumentation Infrastructure's level of service. This implies the DORII Common Library design heavily, as it is supposed to be the main infrastructure service provider for a wide range of the user-level middleware components. On the other hand, improvement of the currently available limitation in possibilities to trouble-free incorporate new instruments is also an important improvement direction for the Remote Instrumentation Infrastructure in the future. Usually, this process needs significant effort from system developers. The current limitation in the domain-specific integration of the Remote Instrumentation should be overcome by offering new, flexible and innovative way of providing service-oriented instrumentation, such as advanced instrument reservation for example. It will increase better e-Infrastructure resource utilisation, especially when interactive access is required. This also implied the remote instruments. Human interaction introduces new limitations which must be taken into consideration (in particular addressed in the workflow solutions). Accounting of scientific instruments is also a very important functionality which the current edition of the Remote Instrumentation Infrastructure is currently lacking. It should provide information about resource utilization (again, including the Remote Instrumentation), in particular allowing for the phases of the application life-cycle (e.g. preparation, parameters tuning, and actual execution), which next could be accounted with different rates. Another innovation concerns the advanced monitoring which will be used to provide up-to-date information about instruments and running experiments. It will extend standard mechanisms and incorporate added value, like proactive monitoring in order to facilitate fast problem resolution. This strategy is in line with a complex of activities performed currently in the European Research Area toward creating a Virtual Research Environment — a dynamic and temporally isolated environment which allows users to proceed with the complex experiment, whereby the Remote Instrumentation tasks might be greatly supported as well. 8. Conclusions and final discussion There are a number of scientific and technologic domains that require broad international collaboration for their success. A number of problems may be addressed by using complex equipment and toplevel expertise, which is often locally unavailable for many institutions. Remote Instrumentation is a new branch of the ICT domain oriented at the integration of scientific instruments into the e-Infrastructure and empowering possibilities in conducting experiments. The possibility of having shared access to complex scientific and even industrial equipment independently of their physical location, in a way similar to the one Grid technology has enabled for computational and storage facilities worldwide, is a great step forward towards designing next generation Remote Instrumentation Services (RIS). An e-Infrastructure capable of providing RIS creates an absolutely new spectrum of opportunities for a number of research projects.

483

Combining RIS with the traditional Grid computation and storage services, enabled by large-scale sustainable e-Infrastructures within a single Virtual Organization (VO), the e-Infrastructure presented in this paper allows scientist to expose, share, and federate within a common application workflow various instrumental resources, by means of universal abstractions that apply to diverse e-Science domains. The e-Infrastructure for Remote Instrumentation, which was successfully developed in the framework of the DORII project, has already facilitated the adoption of Grid technology for numerous e-Science applications. With this paper, the authors would like to encourage the direct involvement of a wider set of user communities to bring their operational experience in the experiments to be carried out on the Remote Instrumentation Infrastructure. Acknowledgments We kindly acknowledge the consortium of the EU project “DORII” for the provided material and the European Commission, which funded the DORII project (contract no. 213110). The DORII project was successfully finished in June 2010. References [1] A. Cheptsov, B. Koller, D. Kranzmueller, T. Koeckerbauer, S. Mueller, N. Meyer, F. Davoli, D. Adami, S. Salon, P. Lazzari, Remote Instrumentation Infrastructure for eScience. Approach of the DORII project, IEEE International Workshop on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS 2009), IEEE Catalog Number: CFP09803-CDR, ISBN: 978-1-4244-4882-1, 2009, pp. 231–236. [2] A. Cheptsov, R. Keller, R. Pugliese, M. Prica, A. Del Linz, M. Plociennik, M. Lawenda, N. Meyer, Towards deployment of the remote instrumentation e-Infrastructure (in the frame of the DORII project), Computational Methods in Science and Technology 15 (1) (2009) 65–74. [3] IFM-GEOMAR, Glider cruising the oceanRetrieved from, http://www.ifm-geomar. de/index.php?id=1241&L=12008. [4] N. Pinardi, et al., Operational Oceanography Observing system developments in the Mediterranean Sea in support of the MFSD implementationRetrieved from, http://www.moon-oceanforecasting.eu/files/medgoos_moonreportupdatejune2009.v9.pdf2009. [5] Lazzari, P. (2008). Coupling of transport and biogeochemical models in the Mediterranean Sea, PhD Thesis in Environmental Science, University of Trieste, Trieste, Italy. [6] D. Adami, A. Cheptsov, F. Davoli, I. Liabotis, R. Pugliese, A. Zafeiropoulos, The DORII project test bed: distributed eScience applications at work, 5th International Conference on Testbeds and Research Infrastructures for the Development of Networks & Communities and Workshops TridentCom 2009. 6–8 April 2009, Washington, DC, USA, ISBN: 978-1-4244-2846-5, 2009, Digital Object Identifier 10.1109/TRIDENTCOM.2009. 4976247. [7] S. Salon, P.-M. Poulain, E. Mauri, R. Gerin, D. Adami, F. Davoli, Remote oceanographic instrumentation integrated in a GRID environment, Computational Methods in Science and Technology 15 (1) (2009) 49–55. [8] J.-W. Huang, Overview of the gLite Middleware and Site Architecture. EUAsiaGrid Summer School (EAGSS09), , 2009 Retrieved from, http://www.euasiagrid.org/ training/Overview_of_the_gLite_middleware_and_Site_Architecture_v1.2.pdf. [9] E. Frizziero, M. Gulmini, F. Lelli, G. Maron, A. Oh, S. Orlando, A. Petrucci, S. Squizzato, S. Traldi, Instrument element: a new grid component that enables the control of remote instrumentation, Proc. 6th IEEE Internat. Symp. on Cluster Computing and the Grid Workshops (CCGRIDW'06), 2006, Retrieved from, http://ieeexplore.ieee.org/ iel5/10857/34198/01630943.pdf. [10] R. Ranon, L. De Marco, A. Senerchia, S. Gabrielli, L. Chittaro, R. Pugliese, L. Del Cano, F. Asnicar, M. Prica, Web-based tool for collaborative access to scientific instruments in cyberinfrastructures, in: F. Davoli, N. Meyer, R. Pugliese, S. Zappatore (Eds.), Grid Enabled Remote Instrumentation, Springer, New York, NY, ISBN: 9780-387-09662-9, 2008, pp. 237–251. [11] M. Okon, D. Kaliszan, M. Lawenda, D. Stokłosa, T. Rajtar, N. Meyer, M. Stroinski, Virtual laboratory as a remote and interactive access to the scientific instrumentation embedded in grid environment, Proc. 2nd IEEE International Conference on eScience and Grid Computing (e-Science'06), 2006, Retrieved from, http:// ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=04031097. [12] H. Kornmayer, M. Stümpert, M. Knauer, P. Wolniewicz, g-Eclipse — an integrated workbench tool for grid application users, grid operators and grid application developers, Cracow Grid Workshop '06, Cracow, Poland, Oct. 2006, 2006. [13] H. Rosmanith, D. Kranzlmueller, glogin — a multifunctional, interactive tunnel into the grid, Proc. 5th IEEE/ACM International Workshop on Grid Computing, Pittsburgh, PA, USA, 2009. [14] The DORII monitoring platform, http://monitor2.cnit.it/. [15] D. Adami, A. Chepstov, F. Davoli, B. Koller, M. Lanati, I. Liabotis, S. Vignola, A. Zafeiropoulos, S. Zappatore, The DORII project e-Infrastructure: deployment, applications, and measurements, Proc. TridentCom 2010, Berlin, Germany, May 2010, 2010.

484

A. Cheptsov et al. / Computer Standards & Interfaces 34 (2012) 476–484 Dr. Alexey Cheptsov is a Scientific Researcher at the High Performance Computing Center Stuttgart (HLRS). He obtained a PhD degree in Progressive Information Technology in 2007. He has been a technical leader and R&D coordinator of the EU projects DORII (2008-2010) and LarKC (2008-2011). His research interest focuses currently on large scale Semantic Web Reasoning and Big Data Clouds in the domain of High Performance Computing.

Dr.-Ing. Bastian Koller received in 2011 a PhD in Mechanical Engineering from the University of Stuttgart and a diploma in Computer Science from the University of Würzburg in 2004. He worked in many national and international research activities realizing Grid and Cloud Computing frameworks as well as improving the usage of HPC Resources. From 2006 to 2010 he was a Technical Manager of the IST BREIN project. Since 2007 he leads the Service Management and Business Processes Group of the High Performance Computing Centre Stuttgart.

Davide Adami received a degree in Electronic Engineering from the Department of Information Engineering at the University of Pisa, Italy, in 1992. From 1993 to August 1997 he worked at Consorzio Pisa Ricerche in the field of computer and telecommunication networks, with particular focus on dial-up internet working of TCP/IP networks, taking part to a lot of research projects funded by the Regione Toscana and the European Community. Within the project Internet, he was the Network Administrator of Towernet, one of the earliest ISP in Italy, where he was concerned with policy routing and intradomain routing protocols (RIP, OSPF). In the framework of the MAESTRO ACTS project he has been dealing with Quality of Service Networks in ATM networks and Resource Allocation Techniques (RSVP). In September1997 he joined the CNIT (National Consortium for Telecommunications), where he is a senior researcher in the field of telecommunication networks. His research interests mainly concern the provisioning of Quality of Service in IP networks and the TCP/IP enhancements for satellite networks. He is one of the technical responsible of the CNIT satellite network. Currently, he is an Assistant Professor at the Faculty of Engineering of the University of Pisa in the courses of “Planning and Simulation of telecommunication networks” and “Systems and Services of Telecommunication Networks”. Franco Davoli received the “laurea” degree in Electronic Engineering in 1975 from the University of Genoa, Italy. Since 1990 he has been a Full Professor of Telecommunication Networks at the University of Genoa, at the Department of Communications, Computer and Systems Science (DIST). His current research interests are in dynamic resource allocation in multiservice networks, wireless mobile and satellite networks, multimedia communications and services, and energy-efficient networking. On these and other aspects he has co-authored over 300 scientific publications in international journals, book chapters and conference proceedings. In 2004 and 2011 he was a Visiting Erskine Fellow at the University of Canterbury, Christchurch, New Zealand. He has been a Principal Investigator in a large number of projects and has served in several positions in the Italian National Consortium for Telecommunications (CNIT), an independent organization joining 37 universities all over Italy. He was the Head of the CNIT National Laboratory for Multimedia Communications in Naples, Italy, for the term 2003–2004, and Vice-President of the CNIT Management Board for the term 2005–2007. He is a Senior Member of the IEEE.

Szymon Mueller graduated from Adam Mickiewicz University in Poznań and received his M.Sc. degree in Mathematics in 2008 and in Computer Science in 2009. Currently he is working in Supercomputing Department at PSNC. Major projects that he has participated in include g-Eclipse, DORII and NEXPReS. His research interests are: distributed computing systems, Grid infrastructures, remote instrumentation, Java and web technologies.

Dr. Norbert Meyer is currently the head of the Supercomputing Department in Poznan Supercomputing and Networking Center (http://www.man.poznan.pl). His research interests concern resource management in GRID environment, GRID accounting, data management, technology of development graphical user interfaces and network security, mainly in the aspects of connecting independent, geographically distant Grid domains. The concept of remote operation, controlling and monitoring instrumentation is one of key research topics he is currently performing on national and international levels. Norbert Meyer conceived the idea of connecting Polish supercomputing centres, vision of dedicated application servers and distributed storage infrastructure. He is the author and co-author of 60+ conference papers and articles in international journals, and member of programme committees of international IT conferences. Member of the eIRG group in EC (www.eirg.org). Coordinator of EU projects (DORII, RINGRID) and national projects (KMD, KMD2, Platon-U4). Paolo Lazzari got a PhD in Environmental Sciences in 2008 at the University of Trieste. He is currently involved in numerical modeling at OGS as a researcher. His activity regards the setup of 3D biogeochemical model at the Mediterranean scale. The aim of this system is to simulate biogeochemical fluxes characterizing the ecosystem of the Mediterranean Sea, with particular focus to the lower part of the trophic web.

Stefano Salon got a PhD in Applied Geophysics and Hydraulics (2004) at the University of Trieste. He has been a researcher at OGS since 2008. His research contributions include turbulence in tidally-driven boundary layers, operational numerical forecasts of the Mediterranean Sea biogeochemistry and dynamical downscaling of climate scenarios for the Lagoon of Venice. He is a component of the Academic Board of the School of Doctorate in Environmental and Industrial Fluid Mechanics (Univ. of Trieste) for which he has been responsible of the course in Geophysical Fluid Dynamics since 2008.

Johannes Watzl is a member of the Munich Network Management Team at the Ludwig-Maximilians-Universität München, Germany. He has joined the team in 2008 after finishing his Master in Tech. Mathematics at the Johannes Kepler Universität Linz, Austria. Johannes' research interests are in the area of distributed computing. He is working for European Commission funded projects in the area of grid and cloud computing. Currently, Johannes is doing a PhD in the field of resource trading.

Michael Schiffers is a member of the Munich Network Management (MNM) team and a research scientist at the Ludwig-Maximilians Universität of Munich, Germany. His current research interests cover Grid Computing with special emphasis on the management of Virtual Organizations, and the survivability of large scale systems. Before joining the MNM Team he hold several positions in the international IT industry. His home page is located under http://www.nm.ifi.lmu.de/~schiffer. He is a member of the German Society of Informatics (GI) and ACM.

Univ.-Prof. Dr. Dieter Kranzlmüller is a full professor of computer science at the Ludwig-Maximilians-Universität München (LMU), member of the board of the Leibniz Supercomputing Centre (LRZ) of the Bavarian Academy of Sciences and Humanities, and Scientific Director of the Center for Digital Technology & Management (CDTM). He is a member of the Executive Board of the EGI.eu Organisation and German represenative on the European Grid Initiative (EGI) Council. He chairs the MNM-Team (Munich Network Management Team), which is engaged in networks and distributed systems in general, and networks, grids, clouds and HPC in particular.