This PDF file contains the front matter associated with SPIE Proceedings Volume 8336, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

There has been an amazing development of modeling and simulation from its beginning in the 1920s, when the technology was available only at a handful of University groups who had access to a mechanical differential analyzer. Today, tools for modeling and simulation are available for every student and engineer. his paper gives a perspective on the development with particular emphasis on technology and paradigm shifts. Modeling is increasingly important for design and operation of complex natural and man-made systems. Because of the increased use of model based control such as Kalman filters and model predictive control, models are also appearing as components of feedback systems. Modeling and simulation are multidisciplinary, it is used in a wide variety of fields and their development have been strongly influenced by mathematics, numerics, computer science and computer technology.

The James Web Space Telescope (JWST) is a large, infrared-optimized space telescope. System-level verification of critical performance requirements will rely on integrated observatory models that predict optical response accurately enough to verify that the allocated top-level wavefront error of 150 nm root-mean-squared (rms) through to wavefront sensor focal plane is met. The assembled models themselves are complex and require the insight of technical experts working as a team across multiple disciplines and organizations. Furthermore, responses in several key disciplines are strongly cross-coupled. This paper describes the breadth of the systems engineering and modeling approach used on the JWST including technical performance metrics management. The scope of the systems-level modeling includes stowed dynamics, deployed dynamics, thermal, thermal distortion, straylight, optics and attitude control systems.

In this paper the opto-mechanical modeling of the Herschel infrared space telescope at ESA/ESTEC is presented. The aim of the paper is to give an overview of all modeling activities that took place between 2006 and 2010. In 2006 ESA commissioned a Tiger Team to review the discrepancy between the prediction and measurement of the change in telescope back focal length of the Herschel infrared space telescope. The understanding of the discrepancy was essential since the telescope did not have a refocusing mechanism and hence had to be shimmed to the focus position at cryogenic operational temperature. A team of 16 engineers and scientists collocated at ESA/ESTEC to review the finite element models, optical models and test data used for the prediction of the telescope back focal length. The methodology of prediction, the uncertainties and the obtained results were critically assessed. The team used various modeling techniques including paraxial optical models, first order linear thermal expansion models, full system and metrology ray tracing, deterministic and stochastic thermo-elastic finite element analyses. The opto-mechanical analysis techniques, assumptions and results are discussed. In addition the impact of new measurements of coefficients of thermal expansion, performed after shimming of the telescope flight model, are addressed.

Modeling is an integral part of systems engineering. It is utilized in requirement validation, system verification, as well as for supporting design trade studies. Modeling highly complex systems poses particular challenges, including the definition and interpretation of system performance, and the combined evaluation of physical processes spanning a wide range of time frames. Our solution is based on statistical interpretation of system performance and a unique image quality metric developed by TMT. The Stochastic Framework and Point Source Sensitivity allow us to properly estimate and combine the optical effects of various disturbances and telescope imperfections.

To predict the performance of the E-ELT three sets of toolkits are developed at ESO: i) The main structure and associated optical unit dynamical and feedback control toolkit, ii) Active optics and phasing toolkit, and iii) adaptive optics simulation toolkit. There was a deliberate policy not to integrate all of the systems into a massive model and tool. The dynamical and control time scale differences are used to separate the simulation environments and tools. Therefore, each toolkit contains an appropriate detail of the problem and holds sufficient overlap with the others to ensure the consistency of the results. In this paper, these toolkits together with some examples are presented.

Ring aperture telescope is a new kind of telescope, it has complete U-V coverage and the limiting resolution equivalent to that of a full aperture telescope with the same diameter. There are two ring telescope proposals in China. One is 30m Ring Interferometric Telescope (RIT), its primary mirror is expected to be composed by 90 segments; another is 8m Chinese Giant Solar Telescope (CGST), its primary mirror is expected to be composed by 24 segments if segmented mirror scheme is adopted . In this paper, an overview of ring telescope and modeling for it are introduced. The main difference and characteristic in active control strategy and adaptive optics implementation between ring segmented telescope and full segmented telescope are pointed out. Detailed analysis about these two aspects are presented. The results of simulation and analysis show that the current strategies on active control and adaptive optics could meet the demand of most science cases not only in infrared waveband but also in near infrared even in visible waveband.

The Stratospheric Observatory for Infrared Astronomy (SOFIA) is an airborne observatory for astronomical observations at wavelengths ranging from 0.3-1600 µm. It consists of a telescope with an effective aperture of 2.5 m, which is mounted in a heavily modified Boeing 747SP. The aircraft features an open port cavity that gives the telescope an unobstructed view of the sky. Hence the optical system is subject to both aerodynamic loads from airflow entering the cavity, and to inertial loads introduced by motion of the airborne platform. A complex suspension assembly was designed to stabilize the telescope. Detailed end-to-end simulations were performed to estimate image stability based on the mechatronic design, the expected loads, and optical influence parameters. In December 2010 SOFIA entered its operational phase with a series of Early Science flights, which have relaxed image quality requirements compared to the full operations capability. At the same time, those flights are used to characterize image quality and image stability in order to validate models and to optimize systems. Optimization of systems is not based on analytical models, but on models derived from system identification measurements that are performed on the actual hardware both under controlled conditions and operational conditions. This paper discusses recent results from system identification measurements, improvements to image stability, and plans for the further enhancement of the system.

Chinese Giant Solar Telescope is the next generation ground based solar telescope of china. Due to the characteristics of ring aperture in polarization detection and thermal control, the current design of CGST is an 8 meter ring solar telescope. The simulation of science cases and the current observations from 1m class solar telescopes indicated the necessity of such a big aperture especially in infrared bands. The integrated simulation of optical system and imaging ability such as optical design, MCAO, active maintenance of primary mirror were carried out in this paper. The results of simulation and analysis showed that the current design could meet the demand of most science cases not only in infrared band but also in near infrared band and visible band.

The National Research Council Canada Integrated Modeling (NRCIM) toolset is based in the MATLAB numerical computing environment. It links ANSYS finite element structural models with ZEMAX optical models to provide a powerful integrated opto-mechanical engineering tool. This toolset is utilized by the Thirty Meter Telescope Project to assess the telescope active optics system performance for gravity and thermal loads, as well as the opto-mechanical performance of the NFIRAOS first light adaptive optics system and the laser guide star beam transfer optics path. This paper describes the NRCIM toolset, including the interfaces to ANSYS, ZEMAX, the method of calculation of the results, and the internal data structures used.

The Thirty Meter Telescope (TMT) is a Ritchey-Chritien optical telescope with a 30-meter diameter primary mirror made up of 492 hexagonal segments. Such a large and complex optical system requires detailed modeling of the optical performance during the design phase. An optical modeling computational framework has been developed to support activities related to wavefront & image performance prediction. The model includes effects related to mirror shape sensing & control, mirror alignment & phasing, M1 segment control, low order wavefront correction, adaptive optics simulation for high order wavefront correction, and high contrast imaging. Here we give an overview of this optical simulation framework, the modeling tools and algorithms that are used, and a set of sample analyses. These tools have been used in many aspects of the system design process from mirror specification to instrument & sensor design to algorithm development and beyond.

The Performance Error Budget of the Thirty Meter Telescope (TMT) suggests that nearly one third of the total image degradation is due to aero-thermal disturbances (mirror and dome seeing, dynamic wind loading and thermal deformations of the optics, telescope structure and enclosure). An update of the current status of aero-thermal modeling and Computational Fluid-Solid Dynamics (CFSD) simulations for TMT is presented. A fast three-dimensional transient conduction-convection-radiation bulk-air-volume model has also been developed for the enclosure and selected telescope components in order to track the temperature variations of the surfaces, structure and interstitial air over a period of three years using measured environmental conditions. It is used for Observatory Heat Budget analysis and also provides estimates of thermal boundary conditions required by the CFD/FEA models and guidance to the design. Detailed transient CFSD conjugate heat transfer simulations of the mirror support assemblies determine the direction of heat flow from important heat sources and provide guidance to the design. Finally, improved CFD modeling is used to calculate wind forces and temperature fields. Wind loading simulations are demonstrated through TMT aperture deflector forcing. Temperature fields are transformed into refractive index ones and the resulting Optical Path Differences (OPDs) are fed into an updated thermal seeing model to estimate seeing performance metrics. Keck II simulations are the demonstrator for the latter type of modeling.

The dynamical behavior of the primary mirror (M1) has an important impact on the control of the segments and the performance of the telescope. Control of large segmented mirrors with a large number of actuators and sensors and multiple control loops in real life is a challenging problem. In virtual life, modeling, simulation and analysis of the M1 bears similar difficulties and challenges. In order to capture the dynamics of the segment subunits (high frequency modes) and the telescope back structure (low frequency modes), high order dynamical models with a very large number of inputs and outputs need to be simulated. In this paper, different approaches for dynamical modeling and simulation of the M1 segmented mirror subject to various perturbations, e.g. sensor noise, wind load, vibrations, earthquake are presented.

This paper summarizes the optical integrated modeling efforts for the James Webb Space Telescope reported from 2003 to 2009. The topics include: 1) development of the linear optical model (or LOM), 2) the extension of the LOM to a field of view, 3) tolerance analysis of figure and alignment modes of the telescope, 4) introduction of math software toolkits, 5) system level modeling and closed loop alignment updates of the observatory on orbit, and 6) primary mirror figure compensation of large figure aberration.

HINODE, Japanese for "sunrise", is a spacecraft dedicated for observations of the Sun, and was launched in 2006 to study the Sun's magnetic fields and how their explosive energies propagate through the different atmospheric layers. The spacecraft carries the Solar Optical Telescope (SOT), which has a 50 cm diameter clear aperture and provides a continuous series of diffraction-limited visible light images from space. The telescope was developed through international collaboration between Japan and US. In order to achieve the diffraction-limited performance, thermal and structural modeling of the telescope was extensively used in its development phase to predict how the optical performance changes dependent on the thermal condition in orbit. Not only the modeling, we devoted many efforts to verify the optical performance in ground tests before the launch. The verification in the ground tests helped us to find many issues, such as temperature dependent focus shifts, which were not identified only through the thermal-structural modeling. Another critical issue was micro-vibrations induced by internal disturbances of mechanical gyroscopes and momentum wheels for attitude control of the spacecraft. Because the structural modeling was not accurate enough to predict how much the image quality was degraded by the micro-vibrations, we measured their transmission in a spacecraft-level test.

The European Solar Telescope (EST) is a European collaborative project promoted by the European Association for Solar Telescopes (EAST) to build a high resolution 4m class solar telescope in the Canary Islands. Since 2008 to 2011 a Design Study has been developed to define the Conceptual Design of EST. This paper presents a description of the conceptual design of EST and the results of the preliminary performance simulations, mainly focused to obtain the best telescope image quality.

ESO is preparing to build the E-ELT which will be the largest optical telescope in the world. Its primary mirror will be (see manuscript)39.3 m and will consist of nearly 1000 hexagonal segments of 1.2 meter width. Each segment is to be supported by the same structure although the segments vary in size and shape, hence this will cause variable surface deflection. Its support structure must guarantee a maximum surface form error of 30 nm rms independent of the segment orientation and environmental temperature. Measures, such as mass balancing and active surface form compensation are needed to compensate for these effects. To meet the main optical and mechanical performance requirements, numerous FEM analysis runs have been made. Dedicated software was used to calculate mirror deformations under mechanical loads and to express these in Zernike modes. Especially this tool proved to be very powerful in proving that the optical surface form requirements could be met for all different sized mirror segments.

For a ring segmented telescope and some others dilute aperture segmented telescope, active alignment is essential for closing the control loop. We choose an internal optical metrology to sense the tip and tilt of each segment for the active alignment. The local seeing is an important factor for the performance of this sensing. The effect of temperature gradient and turbulence near ground on the accuracy and frequency of the tip and tilt sensing is modeled. We conclude that the effect of local seeing must be considered for a low surface error of the primary mirror. Thermal control of the enclosure and telescope should be done for achieving a high accuracy and frequency active alignment.

The Carlina hypertelescope is a planned sparse aperture 100 m telescope with pupil densification. The telescope has a spherical primary with segments located in a valley between mountains, and additional optical elements in a gondola suspended in eight cables some 100 m above the primary mirror. The resolution is about 1.2×10^-3 arcsec. It is imperative that the position and attitude of the gondola be maintained within tight tolerances during observation and star tracking. The present design has servo-controlled winches on the ground for control of the gondola via the cables. An integrated model of the system, including optics, cables, gondola, position and attitude control system, and wind disturbances has been set up. The structural and control models are linear. Calculations in the frequency domain and simulations in the time domain show that the performance of the telescope with the present design seems adequate for short exposures. However, for long-exposure operation, the gondola stability should be improved by about two orders of magnitude. Recommendations are given on possible approaches for performance improvement.

The Advanced Technology Solar Telescope (ATST) is a 4-m class solar telescope to be built in Haleakala Observatory in Maui (HI). It will be the largest solar telescope in the world, with unprecedented abilities to view details of the Sun. Using adaptive optics technology, ATST will be able to provide the sharpest views ever taken of the solar surface. It is expected that the 4-meter class telescope will have a significant impact on the study of stellar magnetic fields, plasma physics and astronomy, allowing scientists to learn even more about the Sun and solar-terrestrial interactions. Apart from the traditional weather protection functions, ATST Enclosure supports the telescope Aperture Stop which must be accurately positioned to fully illuminate the primary mirror, while preventing insolation of any other system components. This positioning is performed by the enclosure shutter (altitude) and azimuth mechanisms. In order to evaluate the positioning capabilities of both mechanisms and verify the structure-mechanism-control interaction of the involved systems a dynamic coupled model has been created. On the other hand, the skin of the enclosure is thermally controlled to avoid the "enclosure seeing" by rejecting solar radiation and keeping its surface temperature as close as possible below the ambient temperature. The thermal control is achieved by a set of water cooled plate coils. The cooling fluid flow is controlled to achieve the desired reference temperature. The performance of the system is being analyzed by means of mathematical model with couples the thermal radiation analysis and the control system performances.

The present paper presents the concept and integrated modeling analysis of a novel mechanism for supporting segmented dishes for radio telescopes and eventually segmented mirrors of optical telescopes. The concept comprises a novel type of hexapod with an original organization of actuators hence degrees of freedom, with a swaying arm based design concept for the connecting joints between panels/segments, thus achieving an iso-static master-slave active surface concept for any triangular and/or hexagonal panel/segment pattern. The integrated modeling comprises all the multifold sizing and performance aspects which must be evaluated concurrently in order to optimize and validate the design and the configuration: in particular kinematic behavior, dynamic analysis, wavefront error and sensitivity analysis. Some experimental verifications already performed validating single aspects of the integrated concept are also presented with the results obtained.

This document will discuss the efforts and progress made in understanding, modeling, quantifying, and mitigating the stresses and behavior induced by creep motion in the 100 meter Green Bank Telescope's (GBT) azimuth track. From its commissioning the GBT azimuth track has been plagued by the side effects of creep motion induced by the 1,000,000 lb rolling load of each of its wheels. This paper will discuss the degrading effects of creep motion, similarities seen in other track systems, and the mitigation efforts employed to maintain track performance. It will also discuss efforts to understand and numerically model the track behavior, efforts to bound stresses using classical calculations, and suggest the direction for further work.

We study a concept for a low-cost, large deformable mirror for an Extremely Large Telescope. The use of inexpensive voice-coil actuators leads to a poorly damped faceplate, with many modes within the desired control bandwidth. A control architecture, including rate and position feedback to add damping and stiness, for the faceplate has been presented in our previous papers. An innovative local control scheme which decouples adjacent actuators and suppresses low-order eigenmodes is a key feature in our controller. Here, we present an integrated model of a partially illuminated large deformable mirror in an experimental laboratory setup with a limited amount of actuators. From the model, conclusions are drawn regarding the number of actuators needed to identify the key features, such as local control performance, dynamic range, and controllability and robustness of the deformable mirror.

In this paper two case studies, highlighting the role of the integrated modeling in the performance prediction and requirement verification of the E-ELT are presented. First example discusses the effect of the local wind load on the secondary mirror and the global wavefront error. Analysis and simulations showed that such local effect is one of the main contributors to the telescope error budget. A detailed modeling and simulation approach with the possibility to modify some important mechanical and control parameters led to a better understanding of the problem and searching for appropriate mitigation strategies. In the second example the effect of friction devices of the lateral support of the azimuth structure is presented. Such devices were originally proposed by the telescope structure contractor to reduce the induced stresses and force due to the temperature variations. A detailed modeling of the friction devices and its inclusion into the telescope dynamical model revealed that some performance and interface requirements are violated.

In order to evaluate the telescope performance and to derive error budget and stroke allocations a ray-tracing and performance analysis toolkit was developed at the European Southern Observatory (ESO) during project Phase B. Performance estimates and individual error and stroke budget allocations are derived by analyzing the impact of a defined set of perturbations at sub - system level after propagation through a model of the wavefront control. The ray-tracing and performance analysis toolkit is used in parallel to other modeling activities such as structural, control and detailed adaptive optics modeling and provides interfaces to any of those. The ray-tracing and performance analysis toolkit consists of a set of models each suitable for analyzing sub-systems at a specific temporal and spatial frequency. The impact of quasi - static and dynamic loads is computed by implementing finite element model (FEM) and control model analysis results in optical models which either use linear optical sensitivities or ray - tracing at different levels of resolution.

The 25-m aperture CCAT submillimeter-wave telescope will have a primary mirror that is divided into 162 individual segments, each of which is provided with 3 positioning actuators. CCAT will be equipped with innovative Imaging Displacement Sensors (IDS) - inexpensive optical edge sensors - capable of accurately measuring all segment relative motions. These measurements are used in a Kalman-filter-based Optical State Estimator to estimate wavefront errors, permitting use of a minimum-wavefront controller without direct wavefront measurement. This controller corrects the optical impact of errors in 6 degrees of freedom per segment, including lateral translations of the segments, using only the 3 actuated degrees of freedom per segment. The edge sensors do not measure the global motions of the Primary and Secondary Mirrors. These are controlled using a gravity-sag look-up table. Predicted performance is illustrated by simulated response to errors such as gravity sag.

The Thirty Meter Telescope primary mirror is composed of 492 segments that are controlled to high precision in the presence of wind and vibration disturbances, despite the interaction with structural dynamics. The higher bandwidth and larger number of segments compared with the Keck telescopes requires greater attention to modeling to ensure success. We focus here on the development and validation of a suite of quasi-static and dynamic modeling tools required to support the design process, including robustness verification, performance estimation, and requirements flowdown. Models are used to predict the dynamic response due to wind and vibration disturbances, estimate achievable bandwidth in the presence of control-structure-interaction (CSI) and uncertainty in the interaction matrix, and simulate and analyze control algorithms and strategies, e.g. for control of focus-mode, and sensor calibration. Representative results illustrate TMT performance scaling with parameters, but the emphasis is on the modeling framework itself.

The Stratospheric Observatory for Infrared Astronomy (SOFIA) is a 2.5m infrared telescope built into a Boeing 747SP. During observations the telescope will not only be subject to aircraft vibrations and maneuver loads - by opening a large door to give the observatory an unhindered view of the sky, there will also be aerodynamic and aeroacoustic disturbances. A critical factor in the overall telescope performance is the SOFIA Secondary Mirror Assembly (SMA). The 35cm silicon carbide mirror is mounted on the Secondary Mirror Mechanism (SMM), which has five degrees-offreedom and consists of two parts: The slow moving base for focusing and centering, and on top of that the Tilt Chop Mechanism (TCM) for chopping with a frequency of up to 20Hz and a chop throw of up to 10arcmin. The development of the controller that is used to meet the stringent performance requirements relys heavily on a state space model of the system. A pole-placement controller is compared to an optimal LQG control approach which makes use of the model to calculate all required system states in real-time. The paper explains the modeling of the TCM with linear differential equations and the optimization via a grey-box model approach with system identification data. Simulated data is then compared to measurements taken on ground and in flight.

The James Webb Space Telescope (JWST) is a three-mirror anastigmatic telescope. For a "blind" telescope alignment estimation and control study at the Goddard Space Flight Center (GSFC), influence functions for each of the primary mirror's 18 segments (6 degrees of freedom each) and for the secondary mirror (5 degrees of freedom) were used to construct a 113-element influence matrix. The singular value vector obtained from pseudoinverse of this influence matrix indicates that the most dominant modes in the control of the telescope are rotation, piston, clocking and translation, in order of significance. The application of an iterative-control algorithm to ten cases of random misalignments, involving both the primary segmented mirror and secondary mirror, shows strong convergence after 5 iterations in most of the cases.

In earlier work we have described the thermal modelling for design and control of a fully insulated, and sometimes ventilated, high precision radio telescope. For such an insulated telescope the modelling of the time-variable dynamic influence of the thermal environment (air, sky and ground radiation, insolation) is relatively simple. The modelling becomes however quite complex for an open-air radio telescope where each individual member of the reflector backup structure (BUS) and the support structure (fork or yoke) is exposed under a different and time-dependent aspect angle to the thermal environment, which applies in particular to solar radiation. We present a time-dependent 800-element thermal model of an open-air telescope. Using the IRAM 30-m radio telescope as the basic mechanical structure, we explain how the temperature induced, real-time pointing and reflector surface deformations can be derived when using as input the day of the year, the thermal environment, and the geographic position of the telescope and its changing pointing direction. Thermal modelling and results similar to those reported here can be used for radio telescope design and real-time control of pointing and surface adjustment of a telescope with active panels.

Thermal modeling of the Thirty Meter Telescope (TMT) was conducted for evaluations of thermal performances by finite element (FE) and optical analysis tools. The thermal FE models consist of the telescope optical assembly systems, instruments, laser facility, control and electronic equipments, and telescope structural members. A three-consecutive-day thermal environment data was implemented for the thermal boundary created by Computational Fluid Dynamics (CFD) based on the environment conditions of the TMT site. Temporal and spatial temperature distributions of the optical assembly systems and the telescope structure were calculated under the environmental thermal conditions including air convections, conductions, heat flux loadings, and radiations. With the calculated temperature distributions, the thermo-elastic analysis was performed to predict thermal deformations of the telescope structure and the optical systems. The line of sight calculation was made using the thermally induced deformations of the optics and structures. Merit function routines (MFR) were utilized to calculate the Optical Path Difference (OPD) maps after repositioning the optics based on a best fit of M1 segment deformations. The goal of this thermal modeling is to integrate the mechanical and optical deformations in order to simulate the thermal effects with the TMT site environment data from CFD.

The two spectropolarimeters for PEPSI (Potsdam Echelle Polarimetric and Spectroscopic Instrument) have been de¬signed in order to reconstruct the full Stokes vector measuring linear and circular polarization simultaneously with a re¬solving power of 120,000. The polarimeters will be attached to the Gregorian focus of the so far largest LBT 2x8.4m telescope and will feed together with permanent focus stations the spectrograph via 44m long fibers connection. The spectrograph will be located in a pressure-temperature controlled chamber within the telescope pier. We present hereafter the last results from combined structural and CFD analyses in order to fulfill the optical requirements.

TNO is developing the Optical Tube Assemblies (OTAs) for the ESO VLT Four Laser Guide Star Facility. The OTAs are Galilean 20x beam expanders, expanding a (see manuscript)15 mm input beam to a steerable (see manuscript)300 mm output beam with a wavefront quality requirement of 50 nm rms. The allowed defocus under the influence of the changing environmental air temperature (0-15°C, -0.7°C/hr gradient) is only 0.2 waves. An integrated model of the thermal behavior has been made, combining optical, thermal lumped mass and FE analyses. This paper describes the athermal design and testing of the system.

The James Webb Space Telescope (JWST) is the successor to the Hubble Space Telescope and will be NASA's premier observatory of the next decade. Image-based wavefront sensing (phase retrieval) is the primary method for ground testing and on-orbit commissioning. For ground tests at NASA's Goddard Space Flight Center (GSFC) and Johnson Space Center (JSC), near-real-time analysis is critical for ensuring that pass/fail criteria are met before completion of a specific test. To address this need we have developed a computational architecture for image processing and phase retrieval. Using commercially available off-the-shelf hardware and software, we have designed, implemented, and tested a solution for high-speed parallel computing. The architecture is a hybrid solution utilizing both CPUs and GPUs and exploiting the unique advantages of each. Discussions are presented of the architecture, performance, and current limitations.

To meet the ever increasing demand for computational speed and use of ever larger datasets, multi GPU instal- lations look very tempting. Lunarc and the Theoretical Astrophysics group at Lund Observatory collaborate on a pilot project to evaluate and utilize multi-GPU architectures for scientific calculations. Starting with a small workshop in 2009, continued investigations eventually lead to the procurement of the GPU-resource Timaeus, which is a four-node eight-GPU cluster with two Nvidia m2050 GPU-cards per node. The resource is housed within the larger cluster Platon and share disk-, network- and system resources with that cluster. The inaugu- ration of Timaeus coincided with the meeting "Computational Physics with GPUs" in November 2010, hosted by the Theoretical Astrophysics group at Lund Observatory. The meeting comprised of a two-day workshop on GPU-computing and a two-day science meeting on using GPUs as a tool for computational physics research, with a particular focus on astrophysics and computational biology. Today Timaeus is used by research groups from Lund, Stockholm and Lule in fields ranging from Astrophysics to Molecular Chemistry. We are investigating the use of GPUs with commercial software packages and user supplied MPI-enabled codes. Looking ahead, Lunarc will be installing a new cluster during the summer of 2011 which will have a small number of GPU-enabled nodes that will enable us to continue working with the combination of parallel codes and GPU-computing. It is clear that the combination of GPUs/CPUs is becoming an important part of high performance computing and here we will describe what has been done at Lunarc regarding GPU-computations and how we will continue to investigate the new and coming multi-GPU servers and how they can be utilized in our environment.

Previous work produced a parallel and moderately scalable wavefront sensor model as part of a larger integrated telescope model. This relied on traditional high performance computing (HPC) techniques using optimised C and MPI based parallelism to marry maximum performance with the productive high-level modelling environment of MATLAB. In the intervening period the computational power and flexibility offered by graphics processors (GPUs) has increased dramatically. This presents both new options in terms of the level of hardware required to perform simulations and also new capabilities in terms of the scope of such simulations. We present a discussion of the currently available approaches and test case performance results based on a port to a GPU platform.