A Virtual Observatory for Pulsar Astronomy

Size: px
Start display at page:

Download "A Virtual Observatory for Pulsar Astronomy"

Transcription

1 A Virtual Observatory for Pulsar Astronomy Supervisor: Dr. John Brooke MASTERS PROJECT INITIAL REPORT Akshay M. Sangodkar ID:

2 Table of Contents Abstract...3 Introduction...3 Background...4 Pulsars...4 Workflow for processing pulsar radio signals...5 Architecture...6 Model for work distribution...6 High-level architectural view...7 Workflows...8 Risk analysis...10 Tools...11 g-eclipse...11 Authentication and authorisation...11 Grid project...13 Job submission and monitoring...14 Workflow builder...16 Grid application deployment...17 SIGPROC...17 Project Plan...18 Conclusion...20 References...21 Appendix I

3 Abstract Electromagnetic signals received by a radio telescope need to be processed in order to detect existence of pulsars. The processing is highly resource intensive. This report discusses the feasibility of running parallel workflows on the Grid. It describes an architecture which can invoke workflows that are resource-aware and can efficiently utilise the power of the Grid resources. Workflow models are compared and evaluated. In a diverse distributed Grid environment, a software application should be robust and fault-tolerant. Therefore, the risk factors have been examined. g-eclipse [1] is described in detail because it is an integrated Grid environment which will be used to develop and manage workflows. Introduction Radio telescopes observing or searching for pulsars receive electromagnetic signals. These signals are weak and spread across a wide frequency spectrum. In order to detect a pulsar, these signals need to be combined in different frequency bands. Complex algorithms need to be applied to observe pulsar signal characteristics. These algorithms are computationally expensive. Earlier, algorithms were applied on a set of distributed supercomputers using Internet-aware Message Passing Interface (MPI) [2]. Workflows were implemented using real-time parallelism on dedicated clusters. These were developed using PACX-MPI [3] as an unconventional workflow language. Parallelism is necessary because the exploration of new pulsars is in a large parameter space spanning multiple frequency bands. Computational efficiency can be achieved if the parameter space is divided and processed in parallel. The current work on Service Oriented Architecture (SOA) based Grid middlewares has leveraged the possibility of creating workflows based on services. Parallelism in a workflow can be achieved by instantiating multiple independent worker services. These services can specify resource requirements such as a dedicated number of processors, software requirements, environment variables, estimated time for execution etc. Hence, a workflow processing radio signals on a Grid to detect pulsars, characterises a virtual observatory. The report describes the architecture and tools to be used to invoke workflows in a Grid. It also describes and evaluates workflow approaches which can efficiently utilise the Grid resources. The tools include g-eclipse which is an integrated Grid environment. It provides a user-friendly interface and tools, to connect and manage jobs submitted to the Grid. It has a collection of editors and wizards which facilitate development of Grid applications and workflows. The architecture uses g-eclipse to connect to glite [4] middleware to submit workflows for pulsar processing. 3

4 Background Pulsars Pulsars are neutron stars emitting concentrated beams of electromagnetic radiation from its radioemitting region. As shown in the Figure 1, the centre of the neutron star has a powerful magnetic field, O(1012)Gauss. These rotate rapidly and if the direction of the radiation beam is not aligned with the rotational axis then it characterises the working of a lighthouse. This radiation beam is picked up as a radio signal by a telescope on the earth, if it coincides periodically with the line of sight between the earth and the pulsar. These periodic signals can act as highly accurate clocks because of the frictionless rotation of massive and condensed star mass. Pulsar astronomy is described comprehensively in [5]. Figure 1: Illustration of a pulsar. When a pulsar radio signal is received at a radio telescope, a periodic increase in the broadband radio noise is observed. The radio signal travels through charged interstellar medium which has a frequency-dependent refractive index. This means that the higher frequency radio waves travel faster through the medium than those at a lower frequency. As a result, in the frequency spectrum that the radio telescope is tuned into, it receives the radio signal earlier in the higher frequency than the lower frequency. This is called dispersion. In order to observe the pulse profile (periodic detection of radio waves), the dispersion from the received signal needs to be cleaned. Several dispersion measures (DMs) are applied to the received signal depending on the distance between the earth and the pulsar, and the charge density of the interstellar medium along the path of the signal. Terrestrial noise is also a part of the received signal, but it is not affected by the interstellar dispersion and hence can be removed. Applying multiple DM to a received radio signal, real-time computing, storing and re-analysing requires high computing power and storage space. This problem is important in both astronomical as well computational science perspective. In [6], the aspects of pulsar observation are described in detail. 4

5 Workflow for processing pulsar radio signals The chapter, Workflows in Pulsar Astronomy by Brooke et al. in the book [7], describes a workflow to analyse stored radio telescope signals received from pulsars and how parallelism can be introduced during the de-dispersion stage of the workflow. Figure 2: Workflow to analyse stored radio telescope signals from pulsars. [Taken from the chapter, Workflows in Pulsar Astronomy by Brooke et al. in the book [7]] The antenna of a radio telescope gathers electromagnetic radiation by focussing at a particular region in the sky. This is called a pointing. It receives a signal in a range of frequencies from the pointing which is then stored as one or more beams. Each beam has a range of frequencies with an upper limit, depending on the smoothness of the received signal. If the received signal surface is rough with respect to the wavelength, then it will be scattered across multiple beams. The received radio signal contains interference due to terrestrial noise and the noise generated from the antenna. Such interferences are not affected by interstellar dispersion and are cleaned. Digital sampling is applied to the analog signal and stored. As shown in Figure 2, beams are read from the repository and passed over to the de-dispersion stage of the workflow. The radio signal contains dispersion which is caused due to the charged interstellar medium along its path towards the earth. The radio waves at a higher frequency travel faster and reach the earth earlier than those at a lower frequency. DMs are applied to each beam. If unknown pulsars are being searched then a specific DM is not known before hand, hence a range of DMs are applied on an trial and error basis. Fourier transforms are applied to each de-dispersed beam in the next stage. Fourier transforms correct Doppler shifting of frequency due to orbital acceleration. For example, some exotic pulsars belong to a binary star system, such as a pulsar and a normal star, a pulsar and a white dwarf etc., which rotate along an orbit. There may be different types of orbits depending on the pulsar period and the orbit period. Fourier transforms need to be applied to correct the radio signals received from such binary star systems in order to observe the pulse profile. Again, for unknown pulsars, a range of fast Fourier transforms (FFTs) need to be applied. We observe that there is ample opportunity to introduce parallelism during the de-dispersion stage of the workflow which involves application of multiple independent DMs and FFTs. After FFTs are applied to the beam, raw files are generated for each trial DM which are then passed to the postanalysis and visualization stage to determine the presence of periodic signals. These raw files may 5

6 represent pulsars, terrestrial interference or the noise due to the antenna and other equipment. Further observations determine whether they are confirmed or rejected. Architecture Model for work distribution The aim for a good model is to extract as much computational power out of the resources as possible, by applying basic concepts of high performance computing such as load-balancing, locality of data and reduced overhead. The data can be divided into small chunks and then distributed to each processor for execution. The chunks can be evenly distributed among the processors and can effectively help in load-balancing. Figure 3: Work distribution model As shown in Figure 3, the original data set of the pulsar signal is divided into chunks of contiguous data. A finite number of worker nodes are instantiated to process these chunks. Each worker node is a collection of dedicated processors. After a short time taken for instantiation, each node requests for a chunk of data. Since, these chunks are separate files, they can be simultaneously loaded into the requesting worker nodes. When the processing is complete, the node requests for the next chunk of data for processing. In the figure, we can see four worker nodes generating a pipeline effect horizontally. Also, there are two post-de-dispersion worker nodes which collect the output data from the worker nodes and send it for post-processing. The chunks are cyclically distributed among the worker nodes for de-dispersion so that the post-de-dispersion nodes collect de-dispersed contiguous set of chunks for post-processing. Cyclic distribution means that a set of contiguous chunks are distributed in a cyclic order. Another option is the blocked strategy where the entire set of chunks are divided into dedicated blocks for 6

7 each worker node. A worker node gets a chunk of data from its own dedicated block. This strategy is not advantageous because then, the post-de-dispersion nodes cannot collect a contiguous set of chunks for post-processing. If we just post-process one output chunk of a block at a time then we will not use the full processing power of the post-de-dispersion node in a processing cycle. Post-processing requires lesser processing power therefore we instantiate fewer post-de-dispersion nodes and process more chunks at a time. This is a work distribution model which is similar to the client-server model in [7] but, it eliminates the dead time observed due to sequential loading of data into the clients via the servers. This model can be extended to support varying number of worker nodes, processors, memory etc. High-level architectural view Figure 4:High-level architectural view The program which is going to be used to process pulsar signals is called SIGPROC [8]. It is a collection of CLI tools. The data chunks will be loaded into the worker nodes as depicted in Figure 3 and then processed using SIGPROC in various stages. The dedisperse tool in SIGPROC applies a number of DMs to the input chunk of data. It is the stage where parallelism can be introduced. In Figure 4, the g-eclipse plugin (explained in detail in the Tools section) to the Eclipse IDE1 is used to launch workflows into a glite Virtual Organisation(VO). The g-eclipse provides a rich user 1 Eclipse platform, 7

8 interface and a Grid middleware independent framework to connect to various existing Grid infrastructures. Jobs are described in the form of Job Service Description Language (JSDL) [9] which is Grid middleware independent and defined by the OGF2. Its user interface supports creation of workflows as a diagrammatic representation, direct acyclic graph (DAG), consisting of jobs and directed connectives. Dependency between a job pair is described using a directed connective. glite is a Grid middleware which provides SOA-based Grid services. Researchers can access the vast geographically distributed computing and storage resources using these services. Jobs can be submitted using the Job Description Language (JDL) [10] to the glite infrastructure through a glite Workload Management System (WMS). Since glite does not accept jobs described in JSDL, geclipse automatically transforms it to JDL before submitting. The g-eclipse framework can be extended to transform jobs described in JSDL in the UI to any existing Grid middleware dependent language. This ability makes g-eclipse an integrated Grid environment for connecting to Grid resources and developing applications. When a job is submitted to the glite WMS, it will try to choose the best suiting Computing Element (CE) which matches the resource requirements in the job description. A CE is a collection of computing resources such as a cluster, which is local to a site. Matching is done using the cached information at the Information Supermarket. Information Supermarket contains the information about the current status of the computing and storage resources. The WMS then prepares the job for submission to the matched CE. The CE then receives the job, gathers all the required files, sets up the environment and then submits the job for execution. More detailed information on the working of glite can be found in [11]. Workflows In Figure 3, we saw how the chunks of pulsar signal data set can be distributed among worker nodes. We will now discuss and evaluate some possible model of workflows to execute the jobs. Figure 5: Workflow using data collector mechanism 2 Open Grid Forum, 8

9 Figure 5 shows that, the entire work of de-dispersion of the pulsar signal data set, can be subdivided into a collection of worker nodes. Each worker node can specify the resource requirements such as the processor count, the operating system, the main memory size, the execution time required etc. Each worker node instantiated will be independent of other worker nodes. A data collector node will be instantiated, in order to manage the chunks of data being sent to the worker nodes. When a worker node is instantiated, it will request the data collector node for a chunk of data. The data collector will then lookup for the next chunk to be processed and respond to the worker node with a specific URI for the chunk. The worker node then will download the chunk and process it. Some post-de-dispersion worker nodes will also be instantiated to collect the output data from the worker nodes and start post-processing. The post-de-dispersion worker nodes request the data collector for a set of de-dispersed chunks generated by the worker nodes after a processing cycle. Figure 6: Workflow which assigns jobs to worker nodes prior to submission In the workflow shown in Figure 6, the collector mechanism is not needed. The pulsar signal data set can be divided into multiple files (chunks) of a desired size (size determined by the duration of the pulsar signal in seconds) using the filterbank tool in SIGPROC. Each file will have a specific URI known prior to job submission. Hence, a workflow can be generated using g-eclipse. For example, lets say the data set is divided into nine chunks named as File 1 to File 9 as shown in the above figure. Then we can generate a workflow which instantiates three worker nodes. Each worker node will have three jobs to be instantiated in sequence. Each job will process (apply DM) to a unique file (chunk) and generate output. The number of worker nodes can be decided depending on the number of processors reserved at the CE to run the entire workflow. The number of jobs in each node can be decided depending upon the number of processors dedicated to a worker node. In g-eclipse, a workflow can describe a dependency between a pair of jobs. A dependent job will start only after its parent job has completed execution. Hence, we can generate a pipeline of jobs 9

10 within a worker node. Also, dependency can be asserted on the post-de-dispersion nodes to start execution only after the relevant de-dispersion jobs have completed execution. This workflow model is advantageous than the workflow in Figure 5 because we do not need additional collector mechanism to manage the sending of chunks to each worker node and to send the output data for post-processing. Hence, it will be used during the development phase. Risk analysis The Grid infrastructure can be unreliable, especially due to the vast geographically distributed computing resources, storage resources, instrumentation and connectivity. A running application may experience problems due to factors such as unreliable communication bandwidth, crashing of requisite software, changes made by administrator, processor malfunctioning etc. In Figure 6, a failure of a worker node may leave it in an inconsistent state. In such a case, the workflow should be able to re-instantiate the worker node and continue processing. A mechanism is needed which can make the system robust and fault-tolerant. Figure 7: Queuing structure A queuing structure is used in AQuA[12], which is an object-based storage device, to control and maintain consistent data-transfer rates to clients. A similar queuing structure can be used for faulttolerance in a workflow as shown in Figure 7. A queue hierarchy is maintained having a root queue and a set of second level queues. The root queue will represent the entire workflow and the second level queues will represent the individual worker nodes. In Figure 6, each worker node has a set of jobs which process a unique file. Correspondingly, the worker node queue will have entries of the files to be processed. When a worker node completes a job, it will remove the corresponding entry in its dedicated queue. All post-processing worker nodes will share one post-processing worker node queue. Each entry in the queue will be a set of files that will be processed by a post-processing worker node at a time. When a post-processing worker node has completed its work, it will remove the corresponding entry in the queue. The root queue will have entries of all the files and the set of files (for post-processing). Whenever any entry is removed from the second level queues, the corresponding entry will also be removed from the root queue. 10

11 If a job in a worker node fails, then the corresponding entry in the second level queue and the root queue will not be removed. A mechanism at the root queue can detect that a job has failed and other dependent jobs have starved. This way we can detect a worker node malfunction and re-instantiate it. Tools g-eclipse The need for computational power and storage requirements have increased tremendously which has given way to the development of Grid infrastructure. In order to coordinate and manage such large distributed resources, several Grid middlewares have been developed. Earlier, the primary focus was to develop the core services required to work with the underlying Grid infrastructure. Basically, they were command-line interfaces which were used mostly by the people who developed them. GUIs have been developed to connect to these CLIs, but mostly they are middleware dependent. The popularity and usefulness of the Grid has rapidly increased the number of potential users. A user-friendly interface is needed for the inexperienced users to connect to the Grid. The g-eclipse is a user-friendly integrated Grid environment which has a Grid middleware independent framework. It can be used to develop Grid applications and connect to Grid resources. It has been built as a plugin to the Eclipse platform, which is an IDE for software development. Currently, it provides connectives to middlewares such as glite, Globus[14], GRIA[15] and AWS[16] [13]. The framework provided by g-eclipse can be extended, to incorporate the functionality to connect to new middlewares. The g-eclipse plugin was used to submit jobs and workflows using the tutorials provided by the geclipse team[17, 18, 19]. The following describes the important features of g-eclipse and the experiences while working with it. Authentication and authorisation In order to connect to a glite infrastructure, the g-eclipse needs to be configured using a Certification Authority (CA) certificate, the VO settings and a signed personal certificate. The CA certificate is mostly available at the CA website. This certificate is imported into the internet browser and the g-eclipse. A personal certificate is generated using the same browser. This new personal certificate is sent to the CA for signing. The CA then signs the certificate using its private key, hence anybody who has the CA's public key can authenticate the signed certificate. A personal certificate signed by a CA is required for authentication at the glite WMS. 11

12 Figure 8: VOMS VO connection settings Connection settings can be taken from a VO manager or can be imported from the VOMS (Virtual Organisation Membership Service) [11] repository. A VOMS repository maintains the information of roles and privileges of users, of VOMS-aware VOs. Figure 8 shows the settings to connect to the geclipsetutor VO. Also, the user needs to register at the VO by sending the public key to the VO manager. The manager then assigns the privileges to the user to use the VO. Figure 9: Proxy certificates When a job is submitted to the glite VO, the user needs to prove his identity by using the signed personal certificate and also a VOMS proxy certificate. This is because a remote service may need to communicate with another service on the user's behalf. A proxy certificate will enable the remote service to prove its entitlement to work on the user's behalf. First a public/private key pair is generated locally and then a new proxy certificate is built having the public key and a Subject Name (SN) having the user's information. When the proxy certificate is activated, the one of the VOMS servers is contacted to get an Attribute Certificate (AC) containing the information about the user's role and privileges at the glite VO. The AC is then embedded into the proxy certificate. It is then signed by the user's personal private key. When a job needs to be submitted from geclipse to the glite VO then, the proxy certificate, the private key for the proxy certificate and the user's personal certificate are sent along with it. The user's private key is never shared. Figure 9 shows two proxy certificates generated in the Authentication Tokens tab. It is a security risk to send the private key with the proxy certificate, but these are time limited. Therefore, long term harm can be avoided. 12

13 Grid project A g-eclipse Grid project is an entity which has a structure suited to manage jobs, workflows, files, resources etc. A project is created for a single VO. In Figure 10, the project is created for the geclipsetutor VO. A virtual folder representing the VO is created in the project structure having the name of the VO. This folder is a virtual representation of the VO infrastructure and gives detailed information about Computing Elements (CE), Storage Elements (SE), installed applications and other services. A CE is a collection of computing resources such as a cluster, which is local to a site. A SE provides virtual links to the storage resources available at the VO. Researchers require certain applications to be deployed at every CE in a VO, so that experiments can be run using all the computational power. These applications are listed in the Applications folder. Other services include information services, job services etc. Information services provide access to information servers, which in turn provide detailed information about the VO. Job services are the available glite WMS to which a job can be submitted for execution. A user can create a connection to any SEs available to allocate and access a personal user space. When an attempt was made to create a personal folder at an SE, only iwrse.fzk.de allowed access. All other SEs rejected the request. When a user is granted access to a VO, the access should be consistent with all the resources the VO is collaborating with. When a connection is established at an SE, it appears in the Connections folder in the Grid project. This connection is like an ordinary folder on the local system. Files can be created, edited, moved in, moved out and deleted. Job Descriptions folder contains a list of jobs created in JSDL/JDL in the local workspace. When these jobs are submitted to the glite, a new entry is created in the Jobs folder. This entry contains the submitted JSDL/JDL jobs, the files staged-in and the files staged out. Workflows folder contains a list of workflow jobs in the local workspace. When a workflow is submitted, a new entry is created in the jobs folder. Figure 10: Grid project 13

14 Job submission and monitoring When a new job is created, an entry is created in the Job Descriptions folder in the Grid project. The job is described in JSDL by default. As shown in the Figure 11, the job EnvJob1.jsdl is opened in a multi-page JSDL editor. Each page is a collection of settings which can be specified in the job. The editor is very user-friendly and provides an easier way to create and describe a job. The tabs to each page can be found in the bottom-left corner. The Overview page gives a brief introduction of the other pages in the editor. The job can be transformed from the JSDL format to the JDL format as shown in Figure 12. This job runs the command /usr/bin/env at the CE and returns the output to a file wf_env_out in the user space at the SE. The staging-in and staging-out of files is described in the InputSandbox and OutputSandbox constructs respectively in JDL. Multiple CEs may be defined under a single VO. The JDL file in Figure 12 specifies that the job requirement is that it should be submitted to the CE having the name iwrce.fzk.de. When this job was submitted without the requirement of a specific CE, it was by default redirected to a CE having the name gilda-01.pd.infn.it. The job would fail consistently due to some authorisation issues. When a specific CE requirement was asserted in the job, it would then get submitted to that CE. It was observed that although all the CEs belong to the geclipsetutor VO, each CE has a dedicated administrator. Hence, a user's roles and permissions may not be applied to all the resources. Moreover, each CE has its own dedicated support team. Each team had to be approached separately; inter-team (inter-ce) support was lacking. It was noted that a single manager for the geclipsetutor VO was a necessity. Figure 11: Multi-page JSDL editor 14

15 Figure 12: Job described in JDL When a job is submitted to glite, the JSDL job gets automatically transformed into JDL, which glite understands. An entry is created in the Jobs folder in the Grid project. Also, the status of the submitted jobs can be viewed in the Jobs tab shown in Figure 13. Figure 13: Jobs tab showing submitted job information The statuses of submitted jobs are mentioned in detail in the Table 1 [11]. If a submitted job fails then it is resubmitted automatically until allowed resubmission limit has reached. The resubmission count can also be mentioned in the job description. Job submission status SUBMITTED WAITING Description A job is submitted to the glite WMS and the files mentioned in the InputSandbox are copied initially to WMS. The WMS queries the Information Services to find a CE which matches the job requirements. READY The WMS readies the job by creating a job-wrapper with required instructions and sending it to the selected CE. SCHEDULED The CE submits the job to a local Worker Node after receiving the instructions from the WMS. RUNNING DONE The files mentioned in the InputSandbox are copied from the WMS to the Worker node and execution starts. Indicates that the job has executed successfully. The files mentioned in the OutputSandbox are copied back to the glite WMS. CLEARED The output can now be retrieved. ABORTED Some error has occurred and the job has failed. Table 1: Submitted job statuses 15

16 The Logging and Bookkeeping service in glite is always active in the background. It keeps track of the events and logs the status of the submitted job. The logs maintained by the service can be retrieved during job execution using g-eclipse UI. Appendix I is a log for the job EnvJob1.jsdl after its successful completion. Workflow builder The workflow builder UI in g-eclipse provides an easy way to create workflows. In Figure 14, a simple workflow is shown where two JSDL jobs interact with each other. The EnvJob1 job runs the command /usr/bin/env at the submitted CE and retrieves all the environment variables at the CE in the output file. The GrepJob1 job runs the command /bin/grep with an argument HOME on the output file generated by EnvJob1 job and generates the output. The workflow palette on the right is used to generate workflows. A job can be added in the workflow using the Workflow Job icon. Input Port and Output Port icons describe the staging-in and staging-out of files in a workflow job respectively. The Link icon is used to connect an output port of a workflow job to an input port of another workflow job. This introduces dependency between the linked jobs. Once the output port generates a file, the dependent job can then retrieve the file at the input port and start running. When a workflow is submitted to glite, an entry is created at the Jobs folder and the Jobs tab. The entry contains the workflow file and the workflow jobs as shown in Figure 14. The JDL version of the workflow is shown in Figure 15. It describes both the workflow jobs which is similar to that of a standalone job description. It also has a Dependencies construct which specifies the dependencies among the included workflow jobs. In JDL, a workflow is represented as a direct acyclic graphs (DAG). Hence, the type of the workflow is defined as Type dag. Figure 14: Workflow builder 16

17 Figure 15: Workflow job in JDL Grid application deployment Sometimes, Grid users have to install a software at a specific CE before they can run a job. In glite, deployment of requisite software on a specific CE is done using job submission. The g-eclipse provides deployment wizards to assist users in the deployment process. In glite, the users have to write a script which describes how to install/uninstall a software on a specific CE. The deployment wizard accepts this script and proceeds for deployment. This may be an option to deploy SIGPROC at a CE before submitting a job. SIGPROC SIGPROC is a set of tools designed to process electromagnetic signals received from a radio telescope and facilitate the search for pulsar profiles. The tool used for chunking of the raw pulsar signal dataset is called the filterbank program. The pulsar signal received by the radio telescope for a particular duration is stored as a raw dataset. This dataset can be divided into smaller chunks using filterbank by specifying the chunk size (duration in seconds). For example, the following command takes the rawdata as the input. The filterbank program then skips 10 seconds of data and then reads for the next 20 seconds indicated by -s and -r options respectively. The resultant dataset is stored in chunk1. % filterbank rawdata -s r 20.0 > chunk1 Multiple chunks of the raw dataset can be generated and then assigned to individual jobs for processing as shown in Figure 6. Each processing job will use tools such as dedisperse to apply a specific DM and FFTs, to generate a de-dispersed output dataset. The other post-processing tools in SIGPROC can be applied to the output dataset in the post-de-dispersion worker nodes. The postprocessing output can then be used for off-line analyses to detect and visualize pulsars. 17

18 Project Plan The project is based on a software development life cycle model which is derived from the waterfall model and the incremental model. As shown in Figure 16, the requirement gathering, design, development, integration, testing and acceptance phases belong primarily to the waterfall model. But the development, testing and integration phases are performed incrementally. At the time of writing this report the requirement gathering and the design phases have been completed. One component of the workflow at a time will be developed, tested and integrated, before the next component is developed. This incremental approach will help in achieving the milestones more effectively. Any shortfall in meeting a milestone can be detected early and addressed appropriately. Figure 16: SDLC for the project The Gantt chart describing the project plan is shown in Figure 17. Figure 17: Gantt chart describing the project plan 18

19 Requirement gathering and design phases have been completed. This includes the background study, learning g-eclipse and glite, developing a suitable architecture and workflow models to process pulsar signals et al. Milestones are indicated as diamonds in the chart. Development and testing phase will take around two months to complete. Setting up of an account in a Grid such as the National Grid Service (NGS)3 may take some days due to creation and signing of certificates and becoming a member of a VO. Therefore, 11 days have been marked for setting up of account and obtaining the required permissions. A break of 12 days from 11th May 2009 to 23rd May 2009 is considered because of the second semester examinations. There are two major milestones during this phase. First, the implementation of the workflow model in Figure 6 must be ready and tested by 29th June This will mark the end of one iteration of the development and testing cycle. In the second iteration, the queuing structure as shown in Figure 7 will enter development phase. The second milestone on 20th July 2009 is achieved when the queuing structure is implemented and tested successfully. This will mark the end of the second iteration of the development and testing cycle. The entire workflow implementation will then have to be approved through acceptance testing. The buffer time contains a few days meant for catching up if the project is lagging behind. Around one month is allocated to write up the dissertation before the submission deadline on 7th September National Grid Service (NGS), 19

20 Conclusion The dataset extracted from the output of the pulsar signal received from a radio telescope, contains a weak pulse profile signal embedded within the noise. Intensive computing is required to search for a new pulsar from such a dataset. The described architecture includes g-eclipse, SIGPROC and glite middleware and it is feasible. It will be used to instantiate workflows for processing the pulsar data. Parallelism will be achieved using the workflow models described. Each workflow will be resource-aware such that, the Grid resources which it uses, satisfy the job's requirements. An application running on a diverse distributed Grid infrastructure should be robust and fault-tolerant. The queuing structure will be used so that a failed node is re-instantiated and pending work is completed. The g-eclipse is an integrated Grid environment and will be used to submit jobs to glite and manage them. Tools and wizards in the g-eclipse to create and deploy Grid applications, will be used. The workflow builder in the g-eclipse is a user-friendly tool and will be used to quickly generate workflows. 20

21 References [1] Wolniewicz, P., Meyer, N., Stroiński, M., Stümpert, M., Kornmayer, H., Polak, M., et al. (2007). Accessing Grid computing resources with g-eclipse platform. Computational Methods in Science and Technologies, 13 (2), [2] Pickles, S. M., Brooke, J. M., Costen, F. M., Gabriel, E., Müller, M., Resch, M., et al. (2001). Metacomputing across intercontinental networks. Future Generation Computer Systems, 17, [3] Brune, M. A., Fagg, E. G., & Resch, M. M. (1999). Message-passing environments for metacomputing. Future Generation Computer Systems, 15, [4] (n.d.). Retrieved April 25, 2009, from glite: [5] Lyne, A. G., & Smith, G. (2005). Pulsar Astronomy (3rd edition ed.). Cambridge: Cambridge University Press. [6] Lorimer, D., & Kramer, M. (2005). A Handbook of Pulsar Astronomy. Cambridge: Cambridge University Press. [7] Taylor, I. J., Deelman, E., Gannon, D. B., & Shields, M. (2007). Workflows for e-science: Scientific Workflows for Grids. London: Springer-Verlag London Limited. [8] Lorimer, D. (2006, March 25). SIGPROC v3.7: (Pulsar) Signal Processing Programs. Retrieved April 25, 2009, from [9] Job Submission Description Language (JSDL) Specification, Version 1.0. (n.d.). Retrieved April 25, 2009, from [10] Job Description Language Attributes Specification. (n.d.). Retrieved April 25, 2009, from [11] glite 3.1 User Guide. (n.d.). Retrieved April 25, 2009, from [12] Wu, J. C., & Brandt, S. A. (2006). The design and implementation of AQuA: an adaptive quality of service aware object-based storage device. In Proceedings of the 23rd IEEE / 14th NASA Goddard Conference on Mass Storage Systems and Technologies, [13] g-eclipse User Guide. (n.d.). Retrieved April 25, 2009, from [14] Globus. (n.d.). Retrieved April 25, 2009, from [15] GRIA. (n.d.). Retrieved April 25, 2009, from [16] Amazon web services. (n.d.). Retrieved April 25, 2009, from [17] g-eclipse GridKA School 2008 Handout. (n.d.). Retrieved April 25, 2009, from [18] Köckerbauer, T. (n.d.). Access the power of the Grid: Tutorial - Dornbirn. Retrieved April 25, 2009, from [19] Kornmayer, H. (n.d.). g-eclipse Hands On. Retrieved from 21

22 Appendix I Logging and Bookkeeping service report for EnvJob1.jsdl ********************************************************************** LOGGING INFORMATION: Printing info for the Job : --Event: RegJob - ns - nsubjobs - seed --Event: RegJob - ns - nsubjobs - seed --Event: Accepted - from - from_host --Event: EnQueued - queue - result --Event: EnQueued - queue - result --Event: DeQueued - local_jobid - queue Sangodkar/CNproxy/CNproxy --Event: Match - dest_id Sangodkar/CNproxy/CNproxy --Event: EnQueued - queue - result Sangodkar/CNproxy/CNproxy --Event: EnQueued 25-Apr :05: NetworkServer 25-Apr :05:37 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :05: NetworkServer 25-Apr :05:37 /CDE/OGermanGrid/OUFZK/CNhost/ 25-Apr :05:39 NetworkServer host range btcentralplus.com NetworkServer 25-Apr :05:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :05:39 /var/glite/workload_manager/input.fl START NetworkServer 25-Apr :05:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :05:39 /var/glite/workload_manager/input.fl OK NetworkServer 25-Apr :05:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :05:40 /var/glite/workload_manager/input.fl WorkloadManager Apr :05:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:40 iwrce.fzk.de:2119/jobmanager-lcgpbs-dgiseq WorkloadManager Apr :05:40 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:40 /var/glite/jobcontrol/queue.fl START WorkloadManager Apr :05:40 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:40 22

23 - queue - result Sangodkar/CNproxy/CNproxy --Event: DeQueued - local_jobid - queue Sangodkar/CNproxy/CNproxy --Event: Transfer - dest_host - dest_instance - dest_jobid - destination - reason - result Sangodkar/CNproxy/CNproxy --Event: Transfer - dest_host - dest_instance - dest_jobid - destination - reason - result Sangodkar/CNproxy/CNproxy --Event: Running - node --Event: Done - exit_code - reason - status_code --Event: Accepted - from - from_host Sangodkar/CNproxy/CNproxy --Event: Transfer - dest_host - dest_instance - dest_jobid - destination - reason - result Sangodkar/CNproxy/CNproxy --Event: Running - node /var/glite/jobcontrol/queue.fl OK WorkloadManager Apr :05:40 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:41 unavailable /var/glite/jobcontrol/queue.fl JobSubmission unique 25-Apr :05:41 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:41 localhost /var/glite/logmonitor/condorg.log/condorg log unavailable LogMonitor unavailable START JobSubmission unique 25-Apr :05:41 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :05:42 localhost /var/glite/logmonitor/condorg.log/condorg log LogMonitor unavailable OK JobSubmission unique 25-Apr :05:42 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :09:05 iwrwn02.fzk.de iwrwn02.fzk.de LRMS 25-Apr :07:22 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :09:05 0 iwrwn02.fzk.de (nil) LRMS OK 25-Apr :07:32 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay Sangodkar 25-Apr :05:45 JobSubmission localhost LogMonitor unique 25-Apr :05:45 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :06:03 iwrce.fzk.de:2119/jobmanager-lcgpbs /var/glite/logmonitor/condorg.log/condorg log unavailable LRMS Job successfully submitted to Globus OK LogMonitor unique 25-Apr :06:03 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :09:32 gt2 iwrce.fzk.de:2119/jobmanager-lcgpbs LogMonitor 23

24 Sangodkar/CNproxy/CNproxy --Event: ReallyRunning Sangodkar/CNproxy/CNproxy --Event: Done - exit_code - reason - status_code Sangodkar/CNproxy/CNproxy 25-Apr :09:28 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :11:39 LogMonitor unique 25-Apr :11:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 25-Apr :11:39 0 Job terminated successfully LogMonitor unique OK 25-Apr :11:39 /CIT/OGILDA/OUPersonal Certificate/LThe University of Manchester/CNAkshay 24

Eclipse Technology Project: g-eclipse

Eclipse Technology Project: g-eclipse (Incubation) Document classification: Made available under the Eclipse Public License v1.0. Date: September 11, 2007 Abstract: This document contains the Release Review Documentation for the Eclipse Technology

More information

g-eclipse A Framework for Accessing Grid Infrastructures Nicholas Loulloudes Trainer, University of Cyprus (loulloudes.n_at_cs.ucy.ac.

g-eclipse A Framework for Accessing Grid Infrastructures Nicholas Loulloudes Trainer, University of Cyprus (loulloudes.n_at_cs.ucy.ac. g-eclipse A Framework for Accessing Grid Infrastructures Trainer, University of Cyprus (loulloudes.n_at_cs.ucy.ac.cy) EGEE Training the Trainers May 6 th, 2009 Outline Grid Reality The Problem g-eclipse

More information

glite Grid Services Overview

glite Grid Services Overview The EPIKH Project (Exchange Programme to advance e-infrastructure Know-How) glite Grid Services Overview Antonio Calanducci INFN Catania Joint GISELA/EPIKH School for Grid Site Administrators Valparaiso,

More information

Advanced Job Submission on the Grid

Advanced Job Submission on the Grid Advanced Job Submission on the Grid Antun Balaz Scientific Computing Laboratory Institute of Physics Belgrade http://www.scl.rs/ 30 Nov 11 Dec 2009 www.eu-egee.org Scope User Interface Submit job Workload

More information

Gergely Sipos MTA SZTAKI

Gergely Sipos MTA SZTAKI Application development on EGEE with P-GRADE Portal Gergely Sipos MTA SZTAKI sipos@sztaki.hu EGEE Training and Induction EGEE Application Porting Support www.lpds.sztaki.hu/gasuc www.portal.p-grade.hu

More information

30 Nov Dec Advanced School in High Performance and GRID Computing Concepts and Applications, ICTP, Trieste, Italy

30 Nov Dec Advanced School in High Performance and GRID Computing Concepts and Applications, ICTP, Trieste, Italy Advanced School in High Performance and GRID Computing Concepts and Applications, ICTP, Trieste, Italy Why the Grid? Science is becoming increasingly digital and needs to deal with increasing amounts of

More information

Easy Access to Grid Infrastructures

Easy Access to Grid Infrastructures Easy Access to Grid Infrastructures Dr. Harald Kornmayer (NEC Laboratories Europe) On behalf of the g-eclipse consortium WP11 Grid Workshop Grenoble, France 09 th of December 2008 Background in astro particle

More information

Architecture of the WMS

Architecture of the WMS Architecture of the WMS Dr. Giuliano Taffoni INFORMATION SYSTEMS UNIT Outline This presentation will cover the following arguments: Overview of WMS Architecture Job Description Language Overview WMProxy

More information

Introduction to Grid Computing

Introduction to Grid Computing Milestone 2 Include the names of the papers You only have a page be selective about what you include Be specific; summarize the authors contributions, not just what the paper is about. You might be able

More information

Eclipse Technology Project: g-eclipse

Eclipse Technology Project: g-eclipse Document classification: Made available under the Eclipse Public License v1.0. Date: November 13, 2009 Abstract: This document contains the Release Review Documentation for the Eclipse Technology g- Eclipse

More information

g-eclipse - An Integrated Framework to Access and Maintain Grid Resources

g-eclipse - An Integrated Framework to Access and Maintain Grid Resources g-eclipse - An Integrated Framework to Access and Maintain Grid Resources Harald Gjermundrød #1, Marios D. Dikaiakos #2, Mathias Stümpert *3, Paweł Wolniewicz^4, and Harald Kornmayer &5 # University of

More information

WORKFLOW ENGINE FOR CLOUDS

WORKFLOW ENGINE FOR CLOUDS WORKFLOW ENGINE FOR CLOUDS By SURAJ PANDEY, DILEBAN KARUNAMOORTHY, and RAJKUMAR BUYYA Prepared by: Dr. Faramarz Safi Islamic Azad University, Najafabad Branch, Esfahan, Iran. Task Computing Task computing

More information

Access the power of Grid with Eclipse

Access the power of Grid with Eclipse Access the power of Grid with Eclipse Harald Kornmayer (Forschungszentrum Karlsruhe GmbH) Markus Knauer (Innoopract GmbH) October 11th, 2006, Eclipse Summit, Esslingen 2006 by H. Kornmayer, M. Knauer;

More information

Grid Scheduling Architectures with Globus

Grid Scheduling Architectures with Globus Grid Scheduling Architectures with Workshop on Scheduling WS 07 Cetraro, Italy July 28, 2007 Ignacio Martin Llorente Distributed Systems Architecture Group Universidad Complutense de Madrid 1/38 Contents

More information

From: Sudarshan N Raghavan (770)

From: Sudarshan N Raghavan (770) Spectrum Software, Inc. 11445 Johns Creek Pkwy. Suite 300 Duluth, GA 30097 www.spectrumscm.com Subject: SpectrumSCM Plugin for the Eclipse Platform Original Issue Date: February 2 nd, 2005 Latest Update

More information

G-ECLIPSE: A MIDDLEWARE-INDEPENDENT FRAMEWORK TO ACCESS AND MAINTAIN GRID RESOURCES

G-ECLIPSE: A MIDDLEWARE-INDEPENDENT FRAMEWORK TO ACCESS AND MAINTAIN GRID RESOURCES G-ECLIPSE: A MIDDLEWARE-INDEPENDENT FRAMEWORK TO ACCESS AND MAINTAIN GRID RESOURCES Harald Gjermundrod, Nicholas Loulloudes, and Marios D. Dikaiakos University of Cyprus PO Box 20537, 75 Kallipoleos Str.

More information

A VO-friendly, Community-based Authorization Framework

A VO-friendly, Community-based Authorization Framework A VO-friendly, Community-based Authorization Framework Part 1: Use Cases, Requirements, and Approach Ray Plante and Bruce Loftis NCSA Version 0.1 (February 11, 2005) Abstract The era of massive surveys

More information

Interconnect EGEE and CNGRID e-infrastructures

Interconnect EGEE and CNGRID e-infrastructures Interconnect EGEE and CNGRID e-infrastructures Giuseppe Andronico Interoperability and Interoperation between Europe, India and Asia Workshop Barcelona - Spain, June 2 2007 FP6 2004 Infrastructures 6-SSA-026634

More information

Kentico CMS 6.0 Intranet Administrator's Guide

Kentico CMS 6.0 Intranet Administrator's Guide Kentico CMS 6.0 Intranet Administrator's Guide 2 Kentico CMS 6.0 Intranet Administrator's Guide Table of Contents Introduction 5... 5 About this guide Getting started 7... 7 Installation... 11 Accessing

More information

Release Notes Release (December 4, 2017)... 4 Release (November 27, 2017)... 5 Release

Release Notes Release (December 4, 2017)... 4 Release (November 27, 2017)... 5 Release Release Notes Release 2.1.4. 201712031143 (December 4, 2017)... 4 Release 2.1.4. 201711260843 (November 27, 2017)... 5 Release 2.1.4. 201711190811 (November 20, 2017)... 6 Release 2.1.4. 201711121228 (November

More information

International Collaboration to Extend and Advance Grid Education. glite WMS Workload Management System

International Collaboration to Extend and Advance Grid Education. glite WMS Workload Management System International Collaboration to Extend and Advance Grid Education glite WMS Workload Management System Marco Pappalardo Consorzio COMETA & INFN Catania, Italy ITIS Ferraris, Acireale, Tutorial GRID per

More information

Grid services. Enabling Grids for E-sciencE. Dusan Vudragovic Scientific Computing Laboratory Institute of Physics Belgrade, Serbia

Grid services. Enabling Grids for E-sciencE. Dusan Vudragovic Scientific Computing Laboratory Institute of Physics Belgrade, Serbia Grid services Dusan Vudragovic dusan@phy.bg.ac.yu Scientific Computing Laboratory Institute of Physics Belgrade, Serbia Sep. 19, 2008 www.eu-egee.org Set of basic Grid services Job submission/management

More information

Problemi di schedulazione distribuita su Grid

Problemi di schedulazione distribuita su Grid Problemi di schedulazione distribuita su Grid Ivan Porro Università degli Studi di Genova, DIST, Laboratorio BioLab pivan@unige.it 010-3532789 Riadattato da materiale realizzato da INFN Catania per il

More information

IBM Best Practices Working With Multiple CCM Applications Draft

IBM Best Practices Working With Multiple CCM Applications Draft Best Practices Working With Multiple CCM Applications. This document collects best practices to work with Multiple CCM applications in large size enterprise deployment topologies. Please see Best Practices

More information

OnCommand Cloud Manager 3.2 Deploying and Managing ONTAP Cloud Systems

OnCommand Cloud Manager 3.2 Deploying and Managing ONTAP Cloud Systems OnCommand Cloud Manager 3.2 Deploying and Managing ONTAP Cloud Systems April 2017 215-12035_C0 doccomments@netapp.com Table of Contents 3 Contents Before you create ONTAP Cloud systems... 5 Logging in

More information

WP3 Final Activity Report

WP3 Final Activity Report WP3 Final Activity Report Nicholas Loulloudes WP3 Representative On behalf of the g-eclipse Consortium Outline Work Package 3 Final Status Achievements Work Package 3 Goals and Benefits WP3.1 Grid Infrastructure

More information

Chapter 2 Introduction to the WS-PGRADE/gUSE Science Gateway Framework

Chapter 2 Introduction to the WS-PGRADE/gUSE Science Gateway Framework Chapter 2 Introduction to the WS-PGRADE/gUSE Science Gateway Framework Tibor Gottdank Abstract WS-PGRADE/gUSE is a gateway framework that offers a set of highlevel grid and cloud services by which interoperation

More information

THEBES: THE GRID MIDDLEWARE PROJECT Project Overview, Status Report and Roadmap

THEBES: THE GRID MIDDLEWARE PROJECT Project Overview, Status Report and Roadmap THEBES: THE GRID MIDDLEWARE PROJECT Project Overview, Status Report and Roadmap Arnie Miles Georgetown University adm35@georgetown.edu http://thebes.arc.georgetown.edu The Thebes middleware project was

More information

DreamTeam Suite User Guide

DreamTeam Suite User Guide DreamTeam Suite User Guide Product Usage and Best Practices Guide By Adam Buxbaum September 20, 2011 Table of Contents Introduction... 4 Project Manager... 6 Creating Projects... 9 Project Import/Export...

More information

Research Collection. WebParFE A web interface for the high performance parallel finite element solver ParFE. Report. ETH Library

Research Collection. WebParFE A web interface for the high performance parallel finite element solver ParFE. Report. ETH Library Research Collection Report WebParFE A web interface for the high performance parallel finite element solver ParFE Author(s): Paranjape, Sumit; Kaufmann, Martin; Arbenz, Peter Publication Date: 2009 Permanent

More information

Import and preprocessing of raw spectrum data

Import and preprocessing of raw spectrum data BioNumerics Tutorial: Import and preprocessing of raw spectrum data 1 Aim Comprehensive tools for the import of spectrum data, both raw spectrum data as processed spectrum data are incorporated into BioNumerics.

More information

The glite middleware. Ariel Garcia KIT

The glite middleware. Ariel Garcia KIT The glite middleware Ariel Garcia KIT Overview Background The glite subsystems overview Security Information system Job management Data management Some (my) answers to your questions and random rumblings

More information

vcenter Server Installation and Setup Modified on 11 MAY 2018 VMware vsphere 6.7 vcenter Server 6.7

vcenter Server Installation and Setup Modified on 11 MAY 2018 VMware vsphere 6.7 vcenter Server 6.7 vcenter Server Installation and Setup Modified on 11 MAY 2018 VMware vsphere 6.7 vcenter Server 6.7 You can find the most up-to-date technical documentation on the VMware website at: https://docs.vmware.com/

More information

Advanced School in High Performance and GRID Computing November Introduction to Grid computing.

Advanced School in High Performance and GRID Computing November Introduction to Grid computing. 1967-14 Advanced School in High Performance and GRID Computing 3-14 November 2008 Introduction to Grid computing. TAFFONI Giuliano Osservatorio Astronomico di Trieste/INAF Via G.B. Tiepolo 11 34131 Trieste

More information

Grid Computing in SAS 9.4

Grid Computing in SAS 9.4 Grid Computing in SAS 9.4 SAS Documentation The correct bibliographic citation for this manual is as follows: SAS Institute Inc. 2013. Grid Computing in SAS 9.4. Cary, NC: SAS Institute Inc. Grid Computing

More information

g-eclipse A Contextualised Framework for Grid Users, Grid Resource Providers and Grid Application Developers

g-eclipse A Contextualised Framework for Grid Users, Grid Resource Providers and Grid Application Developers g-eclipse A Contextualised Framework for Grid Users, Grid Resource Providers and Grid Application Developers Harald Kornmayer 1, Mathias Stümpert 2, Harald Gjermundrød 3, and Pawe l Wolniewicz 4 1 NEC

More information

Scalable Computing: Practice and Experience Volume 10, Number 4, pp

Scalable Computing: Practice and Experience Volume 10, Number 4, pp Scalable Computing: Practice and Experience Volume 10, Number 4, pp. 413 418. http://www.scpe.org ISSN 1895-1767 c 2009 SCPE MULTI-APPLICATION BAG OF JOBS FOR INTERACTIVE AND ON-DEMAND COMPUTING BRANKO

More information

Introduction to Grid Infrastructures

Introduction to Grid Infrastructures Introduction to Grid Infrastructures Stefano Cozzini 1 and Alessandro Costantini 2 1 CNR-INFM DEMOCRITOS National Simulation Center, Trieste, Italy 2 Department of Chemistry, Università di Perugia, Perugia,

More information

Expense: Process Reports

Expense: Process Reports Expense: Process Reports User Guide for Standard Edition Applies to these SAP Concur solutions: Expense Professional/Premium edition Standard edition Travel Professional/Premium edition Standard edition

More information

PRISM - FHF The Fred Hollows Foundation

PRISM - FHF The Fred Hollows Foundation PRISM - FHF The Fred Hollows Foundation MY WORKSPACE USER MANUAL Version 1.2 TABLE OF CONTENTS INTRODUCTION... 4 OVERVIEW... 4 THE FHF-PRISM LOGIN SCREEN... 6 LOGGING INTO THE FHF-PRISM... 6 RECOVERING

More information

vcenter Server Installation and Setup Update 1 Modified on 30 OCT 2018 VMware vsphere 6.7 vcenter Server 6.7

vcenter Server Installation and Setup Update 1 Modified on 30 OCT 2018 VMware vsphere 6.7 vcenter Server 6.7 vcenter Server Installation and Setup Update 1 Modified on 30 OCT 2018 VMware vsphere 6.7 vcenter Server 6.7 You can find the most up-to-date technical documentation on the VMware website at: https://docs.vmware.com/

More information

GRID COMPANION GUIDE

GRID COMPANION GUIDE Companion Subject: GRID COMPANION Author(s): Miguel Cárdenas Montes, Antonio Gómez Iglesias, Francisco Castejón, Adrian Jackson, Joachim Hein Distribution: Public 1.Introduction Here you will find the

More information

EPM Live 2.2 Configuration and Administration Guide v.os1

EPM Live 2.2 Configuration and Administration Guide v.os1 Installation Configuration Guide EPM Live v2.2 Version.01 April 30, 2009 EPM Live 2.2 Configuration and Administration Guide v.os1 Table of Contents 1 Getting Started... 5 1.1 Document Overview... 5 1.2

More information

SwinDeW-G (Swinburne Decentralised Workflow for Grid) System Architecture. Authors: SwinDeW-G Team Contact: {yyang,

SwinDeW-G (Swinburne Decentralised Workflow for Grid) System Architecture. Authors: SwinDeW-G Team Contact: {yyang, SwinDeW-G (Swinburne Decentralised Workflow for Grid) System Architecture Authors: SwinDeW-G Team Contact: {yyang, jchen}@swin.edu.au Date: 05/08/08 1. Introduction SwinDeW-G is a scientific workflow management

More information

ACET s e-research Activities

ACET s e-research Activities 18 June 2008 1 Computing Resources 2 Computing Resources Scientific discovery and advancement of science through advanced computing Main Research Areas Computational Science Middleware Technologies for

More information

DIRAC pilot framework and the DIRAC Workload Management System

DIRAC pilot framework and the DIRAC Workload Management System Journal of Physics: Conference Series DIRAC pilot framework and the DIRAC Workload Management System To cite this article: Adrian Casajus et al 2010 J. Phys.: Conf. Ser. 219 062049 View the article online

More information

AutoPyFactory: A Scalable Flexible Pilot Factory Implementation

AutoPyFactory: A Scalable Flexible Pilot Factory Implementation ATL-SOFT-PROC-2012-045 22 May 2012 Not reviewed, for internal circulation only AutoPyFactory: A Scalable Flexible Pilot Factory Implementation J. Caballero 1, J. Hover 1, P. Love 2, G. A. Stewart 3 on

More information

Grid Experiment and Job Management

Grid Experiment and Job Management Grid Experiment and Job Management Week #6 Basics of Grid and Cloud computing University of Tartu March 20th 2013 Hardi Teder hardi@eenet.ee Overview Grid Jobs Simple Jobs Pilot Jobs Workflows Job management

More information

SZDG, ecom4com technology, EDGeS-EDGI in large P. Kacsuk MTA SZTAKI

SZDG, ecom4com technology, EDGeS-EDGI in large P. Kacsuk MTA SZTAKI SZDG, ecom4com technology, EDGeS-EDGI in large P. Kacsuk MTA SZTAKI The EDGI/EDGeS projects receive(d) Community research funding 1 Outline of the talk SZTAKI Desktop Grid (SZDG) SZDG technology: ecom4com

More information

The Virtual Observatory and the IVOA

The Virtual Observatory and the IVOA The Virtual Observatory and the IVOA The Virtual Observatory Emergence of the Virtual Observatory concept by 2000 Concerns about the data avalanche, with in mind in particular very large surveys such as

More information

DataGrid. Document identifier: Date: 24/11/2003. Work package: Partner: Document status. Deliverable identifier:

DataGrid. Document identifier: Date: 24/11/2003. Work package: Partner: Document status. Deliverable identifier: DataGrid WMS GUI USER G UIDE Document identifier: Work package: Partner: WP1 Datamat SpA Document status Deliverable identifier: Abstract: This document provides a description of all functionalities provided

More information

Grid Infrastructure For Collaborative High Performance Scientific Computing

Grid Infrastructure For Collaborative High Performance Scientific Computing Computing For Nation Development, February 08 09, 2008 Bharati Vidyapeeth s Institute of Computer Applications and Management, New Delhi Grid Infrastructure For Collaborative High Performance Scientific

More information

Enterprise Architect. User Guide Series. Portals. Author: Sparx Systems. Date: 19/03/2018. Version: 1.0 CREATED WITH

Enterprise Architect. User Guide Series. Portals. Author: Sparx Systems. Date: 19/03/2018. Version: 1.0 CREATED WITH Enterprise Architect User Guide Series Portals Author: Sparx Systems Date: 19/03/2018 Version: 1.0 CREATED WITH Table of Contents Portals 3 Perspective Portal 6 Workspace Portal 7 Window Portal 9 Status

More information

How to build Scientific Gateways with Vine Toolkit and Liferay/GridSphere framework

How to build Scientific Gateways with Vine Toolkit and Liferay/GridSphere framework How to build Scientific Gateways with Vine Toolkit and Liferay/GridSphere framework Piotr Dziubecki, Piotr Grabowski, Michał Krysiński, Tomasz Kuczyński, Dawid Szejnfeld, Dominik Tarnawczyk, Gosia Wolniewicz

More information

Deploying virtualisation in a production grid

Deploying virtualisation in a production grid Deploying virtualisation in a production grid Stephen Childs Trinity College Dublin & Grid-Ireland TERENA NRENs and Grids workshop 2 nd September 2008 www.eu-egee.org EGEE and glite are registered trademarks

More information

Setting Up the Server

Setting Up the Server Managing Licenses, page 1 Cross-launch from Prime Collaboration Provisioning, page 5 Integrating Prime Collaboration Servers, page 6 Single Sign-On for Prime Collaboration, page 7 Changing the SSL Port,

More information

EnterpriseLink Benefits

EnterpriseLink Benefits EnterpriseLink Benefits GGY a Moody s Analytics Company 5001 Yonge Street Suite 1300 Toronto, ON M2N 6P6 Phone: 416-250-6777 Toll free: 1-877-GGY-AXIS Fax: 416-250-6776 Email: axis@ggy.com Web: www.ggy.com

More information

ms-help://ms.technet.2004apr.1033/ad/tnoffline/prodtechnol/ad/windows2000/howto/mapcerts.htm

ms-help://ms.technet.2004apr.1033/ad/tnoffline/prodtechnol/ad/windows2000/howto/mapcerts.htm Page 1 of 8 Active Directory Step-by-Step Guide to Mapping Certificates to User Accounts Introduction The Windows 2000 operating system provides a rich administrative model for managing user accounts.

More information

OpenIAM Identity and Access Manager Technical Architecture Overview

OpenIAM Identity and Access Manager Technical Architecture Overview OpenIAM Identity and Access Manager Technical Architecture Overview Overview... 3 Architecture... 3 Common Use Case Description... 3 Identity and Access Middleware... 5 Enterprise Service Bus (ESB)...

More information

AGATA Analysis on the GRID

AGATA Analysis on the GRID AGATA Analysis on the GRID R.M. Pérez-Vidal IFIC-CSIC For the e682 collaboration What is GRID? Grid technologies allow that computers share trough Internet or other telecommunication networks not only

More information

DreamTeam Suite User Guide

DreamTeam Suite User Guide DreamTeam Suite User Guide Product Usage and Best Practices Guide By Adam Buxbaum August 22, 2011 Table of Contents Introduction... 4 Project Manager... 6 Creating Projects... 9 Managing Resources: Addition

More information

Grid Computing Middleware. Definitions & functions Middleware components Globus glite

Grid Computing Middleware. Definitions & functions Middleware components Globus glite Seminar Review 1 Topics Grid Computing Middleware Grid Resource Management Grid Computing Security Applications of SOA and Web Services Semantic Grid Grid & E-Science Grid Economics Cloud Computing 2 Grid

More information

Assistant Professor, School of Computer Applications,Career Point University,Kota, Rajasthan, India Id

Assistant Professor, School of Computer Applications,Career Point University,Kota, Rajasthan, India  Id International Journal of Scientific Research in Computer Science, Engineering and Information Technology 2018 IJSRCSEIT Volume 3 Issue 7 ISSN : 2456-3307 An Architectural Framework of Cloud Computing behind

More information

ECE 7650 Scalable and Secure Internet Services and Architecture ---- A Systems Perspective

ECE 7650 Scalable and Secure Internet Services and Architecture ---- A Systems Perspective ECE 7650 Scalable and Secure Internet Services and Architecture ---- A Systems Perspective Part II: Data Center Software Architecture: Topic 3: Programming Models CIEL: A Universal Execution Engine for

More information

Monitoring the Usage of the ZEUS Analysis Grid

Monitoring the Usage of the ZEUS Analysis Grid Monitoring the Usage of the ZEUS Analysis Grid Stefanos Leontsinis September 9, 2006 Summer Student Programme 2006 DESY Hamburg Supervisor Dr. Hartmut Stadie National Technical

More information

Setup Desktop Grids and Bridges. Tutorial. Robert Lovas, MTA SZTAKI

Setup Desktop Grids and Bridges. Tutorial. Robert Lovas, MTA SZTAKI Setup Desktop Grids and Bridges Tutorial Robert Lovas, MTA SZTAKI Outline of the SZDG installation process 1. Installing the base operating system 2. Basic configuration of the operating system 3. Installing

More information

Troubleshooting Grid authentication from the client side

Troubleshooting Grid authentication from the client side Troubleshooting Grid authentication from the client side By Adriaan van der Zee RP1 presentation 2009-02-04 Contents The Grid @NIKHEF The project Grid components and interactions X.509 certificates, proxies

More information

Deliverable 7.3. Workflow Manager. Poznao Supercomputing and Networking Center

Deliverable 7.3. Workflow Manager. Poznao Supercomputing and Networking Center Deliverable 7.3 Workflow Manager Poznao Supercomputing and Networking Center Table of contents 1 Introduction... 3 2 E-VLBI Correlation Platform... 3 2.1 Liferay introduction... 3 2.2 NEXPReS model...

More information

For Trace and Log Central to work, you must resolve DNS lookup for all nodes in the cluster on the client machine.

For Trace and Log Central to work, you must resolve DNS lookup for all nodes in the cluster on the client machine. Trace and Log Central, page 1 Log Viewers, page 42 Plugins, page 45 Trace and Log Central For Trace and Log Central to work, you must resolve DNS lookup for all nodes in the cluster on the client machine.

More information

Harnessing Grid Resources to Enable the Dynamic Analysis of Large Astronomy Datasets

Harnessing Grid Resources to Enable the Dynamic Analysis of Large Astronomy Datasets Page 1 of 5 1 Year 1 Proposal Harnessing Grid Resources to Enable the Dynamic Analysis of Large Astronomy Datasets Year 1 Progress Report & Year 2 Proposal In order to setup the context for this progress

More information

A data dependency based strategy for intermediate data storage in scientific cloud workflow systems

A data dependency based strategy for intermediate data storage in scientific cloud workflow systems CONCURRENCY AND COMPUTATION: PRACTICE AND EXPERIENCE Published online 27 August 2010 in Wiley Online Library (wileyonlinelibrary.com)..1636 A data dependency based strategy for intermediate data storage

More information

The LHC Computing Grid

The LHC Computing Grid The LHC Computing Grid Gergely Debreczeni (CERN IT/Grid Deployment Group) The data factory of LHC 40 million collisions in each second After on-line triggers and selections, only 100 3-4 MB/event requires

More information

AliEn Resource Brokers

AliEn Resource Brokers AliEn Resource Brokers Pablo Saiz University of the West of England, Frenchay Campus Coldharbour Lane, Bristol BS16 1QY, U.K. Predrag Buncic Institut für Kernphysik, August-Euler-Strasse 6, 60486 Frankfurt

More information

Contents. Properties: Field Area Fields Add a Table to a Form... 23

Contents. Properties: Field Area Fields Add a Table to a Form... 23 Workflow Design Guide Version 18 February 2018 Contents About This Guide... 7 Workflows and Forms Overview... 7 Security Permissions for Workflows and Forms... 8 Search for a Workflow Design, Workflow

More information

Assignment 5. Georgia Koloniari

Assignment 5. Georgia Koloniari Assignment 5 Georgia Koloniari 2. "Peer-to-Peer Computing" 1. What is the definition of a p2p system given by the authors in sec 1? Compare it with at least one of the definitions surveyed in the last

More information

Troubleshooting Grid authentication from the client side

Troubleshooting Grid authentication from the client side System and Network Engineering RP1 Troubleshooting Grid authentication from the client side Adriaan van der Zee 2009-02-05 Abstract This report, the result of a four-week research project, discusses the

More information

UNICORE Globus: Interoperability of Grid Infrastructures

UNICORE Globus: Interoperability of Grid Infrastructures UNICORE : Interoperability of Grid Infrastructures Michael Rambadt Philipp Wieder Central Institute for Applied Mathematics (ZAM) Research Centre Juelich D 52425 Juelich, Germany Phone: +49 2461 612057

More information

Workload Management. Stefano Lacaprara. CMS Physics Week, FNAL, 12/16 April Department of Physics INFN and University of Padova

Workload Management. Stefano Lacaprara. CMS Physics Week, FNAL, 12/16 April Department of Physics INFN and University of Padova Workload Management Stefano Lacaprara Department of Physics INFN and University of Padova CMS Physics Week, FNAL, 12/16 April 2005 Outline 1 Workload Management: the CMS way General Architecture Present

More information

Clouds: An Opportunity for Scientific Applications?

Clouds: An Opportunity for Scientific Applications? Clouds: An Opportunity for Scientific Applications? Ewa Deelman USC Information Sciences Institute Acknowledgements Yang-Suk Ki (former PostDoc, USC) Gurmeet Singh (former Ph.D. student, USC) Gideon Juve

More information

vsphere Installation and Setup Update 2 Modified on 10 JULY 2018 VMware vsphere 6.5 VMware ESXi 6.5 vcenter Server 6.5

vsphere Installation and Setup Update 2 Modified on 10 JULY 2018 VMware vsphere 6.5 VMware ESXi 6.5 vcenter Server 6.5 vsphere Installation and Setup Update 2 Modified on 10 JULY 2018 VMware vsphere 6.5 VMware ESXi 6.5 vcenter Server 6.5 You can find the most up-to-date technical documentation on the VMware website at:

More information

Parallel Computing in EGI

Parallel Computing in EGI Parallel Computing in EGI V. Šipková, M. Dobrucký, and P. Slížik Ústav informatiky, Slovenská akadémia vied 845 07 Bratislava, Dúbravská cesta 9 http://www.ui.sav.sk/ {Viera.Sipkova, Miroslav.Dobrucky,

More information

Using the vrealize Orchestrator Operations Client. vrealize Orchestrator 7.5

Using the vrealize Orchestrator Operations Client. vrealize Orchestrator 7.5 Using the vrealize Orchestrator Operations Client vrealize Orchestrator 7.5 You can find the most up-to-date technical documentation on the VMware website at: https://docs.vmware.com/ If you have comments

More information

QosCosGrid Middleware

QosCosGrid Middleware Domain-oriented services and resources of Polish Infrastructure for Supporting Computational Science in the European Research Space PLGrid Plus QosCosGrid Middleware Domain-oriented services and resources

More information

Boundary control : Access Controls: An access control mechanism processes users request for resources in three steps: Identification:

Boundary control : Access Controls: An access control mechanism processes users request for resources in three steps: Identification: Application control : Boundary control : Access Controls: These controls restrict use of computer system resources to authorized users, limit the actions authorized users can taker with these resources,

More information

Account Plan Pro Set Up Guide

Account Plan Pro Set Up Guide Account Plan Pro Set Up Guide Version 7 PLAN2WIN SOFTWARE TABLE OF CONTENTS INSTALLING THE APPLICATION 4 SET UP CUSTOM PROFILE 4 SET PERMISSIONS 5 ENABLE VISUALFORCE PAGE ACCESS 7 MANAGE LICENSES/ SET

More information

LCG-2 and glite Architecture and components

LCG-2 and glite Architecture and components LCG-2 and glite Architecture and components Author E.Slabospitskaya www.eu-egee.org Outline Enabling Grids for E-sciencE What are LCG-2 and glite? glite Architecture Release 1.0 review What is glite?.

More information

OGCE User Guide for OGCE Release 1

OGCE User Guide for OGCE Release 1 OGCE User Guide for OGCE Release 1 1 Publisher s Note Release 2 begins the migration to open standards portlets. The following has been published by the Open Grids Computing Environment: OGCE Release 2

More information

On the employment of LCG GRID middleware

On the employment of LCG GRID middleware On the employment of LCG GRID middleware Luben Boyanov, Plamena Nenkova Abstract: This paper describes the functionalities and operation of the LCG GRID middleware. An overview of the development of GRID

More information

Using NetShow Commands

Using NetShow Commands CHAPTER 10 Network show (NetShow) commands represent a set of read-only commands. These are primarily, show commands that you can run on devices that are managed in LMS. You can specify the commands that

More information

A Simplified Access to Grid Resources for Virtual Research Communities

A Simplified Access to Grid Resources for Virtual Research Communities Consorzio COMETA - Progetto PI2S2 UNIONE EUROPEA A Simplified Access to Grid Resources for Virtual Research Communities Roberto BARBERA (1-3), Marco FARGETTA (3,*) and Riccardo ROTONDO (2) (1) Department

More information

SharePoint User Manual

SharePoint User Manual SharePoint User Manual Developed By The CCAP SharePoint Team Revision: 10/2009 TABLE OF CONTENTS SECTION 1... 5 ABOUT SHAREPOINT... 5 1. WHAT IS MICROSOFT OFFICE SHAREPOINT SERVER (MOSS OR SHAREPOINT)?...

More information

Monitoring System for the GRID Monte Carlo Mass Production in the H1 Experiment at DESY

Monitoring System for the GRID Monte Carlo Mass Production in the H1 Experiment at DESY Journal of Physics: Conference Series OPEN ACCESS Monitoring System for the GRID Monte Carlo Mass Production in the H1 Experiment at DESY To cite this article: Elena Bystritskaya et al 2014 J. Phys.: Conf.

More information

Scientific data processing at global scale The LHC Computing Grid. fabio hernandez

Scientific data processing at global scale The LHC Computing Grid. fabio hernandez Scientific data processing at global scale The LHC Computing Grid Chengdu (China), July 5th 2011 Who I am 2 Computing science background Working in the field of computing for high-energy physics since

More information

Contents. Add a Form Element to a Group Box Add a Field to a Form... 22

Contents. Add a Form Element to a Group Box Add a Field to a Form... 22 Workflow Design Guide Version 17 November 2017 Contents About This Guide... 7 Workflows and Forms Overview... 7 Security Permissions for Workflows and Forms... 8 Search for a Workflow Design, Workflow

More information

Portfolios Creating and Editing Portfolios... 38

Portfolios Creating and Editing Portfolios... 38 Portfolio Management User Guide 16 R1 March 2017 Contents Preface: Using Online Help... 25 Primavera Portfolio Management Overview... 27 Portfolio Management Software for Technology Leaders... 27 Solution

More information

3 Connecting to Applications

3 Connecting to Applications 3 Connecting to Applications 3 Connecting to Applications...1 3.1 Prerequisites...1 3.2 Introduction...1 3.2.1 Pega, the Widget Supplier...2 3.2.2 Mega, the Widget Procurer...2 3.3 Create Requisition...3

More information

Real-time grid computing for financial applications

Real-time grid computing for financial applications CNR-INFM Democritos and EGRID project E-mail: cozzini@democritos.it Riccardo di Meo, Ezio Corso EGRID project ICTP E-mail: {dimeo,ecorso}@egrid.it We describe the porting of a test case financial application

More information

Administration 1. DLM Administration. Date of Publish:

Administration 1. DLM Administration. Date of Publish: 1 DLM Administration Date of Publish: 2018-07-03 http://docs.hortonworks.com Contents ii Contents Replication Concepts... 4 HDFS cloud replication...4 Hive cloud replication... 4 Cloud replication guidelines

More information

SPARC 2 Consultations January-February 2016

SPARC 2 Consultations January-February 2016 SPARC 2 Consultations January-February 2016 1 Outline Introduction to Compute Canada SPARC 2 Consultation Context Capital Deployment Plan Services Plan Access and Allocation Policies (RAC, etc.) Discussion

More information

DOWNLOAD PDF SQL SERVER 2012 STEP BY STEP

DOWNLOAD PDF SQL SERVER 2012 STEP BY STEP Chapter 1 : Microsoft SQL Server Step by Step - PDF Free Download - Fox ebook Your hands-on, step-by-step guide to building applications with Microsoft SQL Server Teach yourself the programming fundamentals

More information