Chiozzi, G.

Paper	Title	Page
MOAB03	Trends in Software for Large Astronomy Projects	13
	K. K. Gillies Gemini Observatory, Southern Operations Center, Tucson, AZ B. D. Goodrich, S. B. Wampler Advanced Technology Solar Telescope, National Solar Observatory, Tucson J. M. Johnson, K. McCann W. M. Keck Observatory, Kamuela S. Schumacher National Optical Astronomy Observatories, La Serena, Chile D. R. Silva AURA/Thirty Meter Telescope, Pasadena/CA A. Wallander, G. Chiozzi ESO, Garching bei Muenchen
	The current 8-10M ground-based telescopes require complex real-time control systems that are large, distributed, fault-tolerant, integrated, and heterogeneous. New challenges are on the horizon with new instruments, AO, laser guide stars, and the next generation of even larger telescopes. These projects are characterized by increasing complexity, where requirements cannot be met in isolation due to the high coupling between the components in the control and acquisition chain. Additionally, the high cost for the observing time imposes very challenging requirements in terms of system reliability and observing efficiency. The challenges presented by the next generation of telescopes go beyond a matter of scale and may even require a change in paradigm. Although our focus is on control systems, it is essential to keep in mind that this is just one of the several subsystems integrated in the whole observatory end-to-end operation. In this paper we show how the astronomical community is responding to these challenges in the software arena. We analyze the evolution in control system architecture and software infrastructure, looking into the future for these two generations of projects.
	Slides
WPPA10	Study of Portability of VLT Instrumentation Software to ACS	337
	G. Chiozzi, A. Longinotti ESO, Garching bei Muenchen P. Santin, R. Cirami INAF-OAT, Trieste
	The Very Large Telescope (VLT) will remain in operation most probably for at least two more decades. Being the software technology currently used at the VLT and more than one decade old, the maintanability of such a complex system might become a critical issue. The ALMA Common Software (ACS) is based on newer technology. Following this consideration, one of the obvious options to improve the maintainability of the VLT Software would be to port it, or parts of it, to ACS. This would allow optimizing maintenance resources for both VLT and ALMA Software, eventually making available resources for new ESO projects, such as E-ELT. Because of operational constraints, this can only be achieved gradually, possibly starting with new VLT sub-systems. In the year 2004 a pilot project has been started to study the effort needed to replace standard components of the VLT Instrumentation Software with ACS based ones. Starting from a simple instrument created from the VLT Template Instrument and entirely based on the VLTSW, we have replaced the core of the Observation Software (OS) with an ACS based equivalent. The purpose of this paper is to summarize the work done and draw some conclusions.