CUC 2004 / New Frontiers / New Techhnologies for New Needs
CARNet logo
Workflow with Dynamic Measurement Scenarios in the Virtual Laboratory / B3
Authors: Marcin Lawenda, Norbert Meyer, Tomasz Rajtar, Marcin Okon, Dominik Stoklosa, Damian Kaliszan, Miroslaw Kupczyk, Maciej Stroiski, Poznan Supercomputing and Networking Center, Poland


In the paper Borne theoretical considerations about dynamic measurement scenarios aspects are discussed. The Dynamic Measure- ment Scenarios are used for creating and managing job workflow in the Virtual Laboratory system. It facilitate defining and monitoring mea- surement process which quite often consists of many tasks connected between themselves.

Virtual laboratories (VL) are now in the scope of activities of research groups and companies which are working on new solutions for the Grid computational environment [1,3]. The main reason for creating such laboratories is the remote access to the scientific equipment, the facility and acceleration of the education process, exchanging opinions in research groups and the management of the ex- periment results. This idea is especially attractive for the experimental sciences and technologies, particularly physics, chemistry, structural biology, experimen- tal medicine, radio astronomy and also for widely taken engineering [2].
Virtual Laboratory is a heterogeneous, distributed environment, which al- lows scientists all over the world to work on a common group of projects. This . environment should allow conducting experiments with the usage of physical devices, doing simulation using the computational application, and communica- tion between users working on the same topic. Similarly to typical laboratory tools and research techniques depending on the specific field of science, virtual laboratories can benefit from same collaboration techniques as tele-immersion, but they are not mandatory.

The experiment process execution in many types of laboratories consists of very similar stages. The given experimental process is often recurrent and must be executed many times by same parameters modification. In this situation the measurement scenarios conception seems to be very useful.
The conception of the dynamic measurement scenarios allows defining the process of an experiment in any way, from pre-processing, through executing the experiment, to the post-processing and visualization tasks. It is very important that thanks to the DMS we can connect experimental and computational jobs together in a one workflow. Users are also allowed to add their own module as a part of the scenario. Defining the measurement scenario allows to spare a lat of time during computation. The user does not have to wait for the end of a given process stage to subrnit another one. It is made automatically.
This simple task's execution chain we call a static measurement scenario (SMS). It allows connecting particular stages only between themselves. Here, the execution path is specific and the user can not manipulate it.
To increase the possibility of the jobs scenario the dynamic measurement scenario (DMS) was defined. In the DMS model, besides the definition of the tasks execution sequence, we can also define same extra connections (e.g. loops, parallel connections), conditions on the connections and different lengths of the execution paths. In fact, DMS allows to define many SMS models in one scenario. Thanks to the possibility of the conditions defining on the connections paths the real path is determined during execution and can depend on computational results.
To properly define the dynamic measurement scenario the system has to have knowledge about connections which are enabled. In case of the laboratory type where the processing software is well known, it seems to be easy. The case will be more complex when we want to define the DMS model for a wider range of virtuallaboratories.
An expertise from a particular discipline is necessary to define rules for DMS. It can be dane by a laboratory administrator together with a domain expert.

The Virtual Laboratory system is developed in Pozna Supercomputing and Net- working System [4] in collaboration with the Institute of Bioorganic Chemistry [5] and Radioastronomy Department of Nicolaus Copernicus University [6].
The Dynamic Measurement Scenarios will apply in two of our exemplary implementation of virtual laboratories: the Virtual Laboratory of NMR Spec- troscopy (devices: Bruker Avance 600, Varian Unity 300) and the Virtual Lab- oratory of Radiotelescopy (device: radiotelescope -diameter 32m) available to the Grid users [7].
As we mentioned before the connection diagram has to be defined in the initial stage of the DMS creating. In Fig. 1 we demonstrate an exemplary diagram for the Virtual Laboratory of NMR Spectroscopy. There are presented major processing stages in NMR experiment and applications assigned to each stage.
On the figure below (see Fig. 2) we present an exemplarymeasurement sce- nario diagram for the same laboratory type. This diagram can be created using special software called Scenario Submission Application (SSA). Thanks to the SSA the user can easily define, submit and monitor the progress of the DMS realization.,
Basing on the DMS we would like also entirely automating the process of defining and executing the VLBI experiment in the Virtual Laboratory of Ra- diotelescopy. Virtual laboratory user only needs specify necessary observation parameters. Parameters distribution to radio telescopes, recalculation them to appropriate radio telescope values, launching observation on time, data trans- porting and correlating and post-processing execution can be performed auto- matically.

Virtual Laboratories are undoubtedly the technology of the future. They are perceived as a remedy for problems connected with access or buying costs of expensive and rare devices. Current technologies connected with VL are in their first stadium of development. New created services allow to fun simple experi- ments, which can be managed by changes in a limited parameter paolo
The Dynamic Measurement Scenario is a flexible and very convenient way to define workflow in many types of virtuallaboratories. It allows to spare a lat of user's time and speed up of the project realization. Thanks to the Submission Scenario Application the user can easily define, submit and monitor the progress of the DMS realization. In a case of an interactive task realization, it is possible to reserve a time slot on a laboratory device or computational server and notify the user about that.

1. Afsarmanesh, H., Benabdelkader, A., Kaletas, E.C., Garita, C., and Hertzberger, L.O.: Towards a Multi-layer Architecture for Scientific Virtual Laboratories.
Bubak, M., Afsarmanesh, H., Williams, R., Hertberger, B., (Eds.), Proc. Int. Conf. High Performance Computing and Networking, Amsterdam, May 8-10, 2000, Lecture Notes in Computer Science 1823, 162-176, Springer, 2000
2. Nicholas. P., Fushman. D., Ruchinsky. V., Cowburn. D.: The Virtual NMR Spec- trometer: a computer program for efficient simulation of NMR experiments involving pulsed field gradients. (2000)
3. Lawenda. M., Meyer. N., Rajtar. T.: General framework for Virtual Laboratory. The 2nd Cracow Grid Workshop, Cracow, Poland, December 11 -14, 2002 4. Poznan Supercomputing and Networking Center
5. Institute of Bioorganic Chemistry Polish Academy ofSciences
6. Radioastronomy Department of Nicolaus Copernicus University
7. Virtual Laboratory project http: I Iv lab. psnc. pIl


Copyright © 1991- 2004. CARNet. All rights reserved. / Mail to / Legal notes / Impressum