AoD

From EGIWiki
Jump to: navigation, search


Applications on Demand (AoD) Service Information pages

Applications on Demand Service menu: Home Documentation for providers Documentation for developers Architecture




The EGI Applications on Demand service (AoD) is the EGI’s response to the requirements of researchers who are interested in using applications in a on-demand fashion together with the compute and storage environment needed to compute and store data.

The Service can be accessed through the EGI Marketplace

Description of service

The service allows user-friendly access to a portfolio of applications and application hosting frameworks (e.g.: Science Gateways, VREs) that are configured to use the dedicated pool of cloud computing clusters from EGI. The service also allows users to run their own simulation/analysis models with custom input data.

The service operates as an open and extensible ‘hub’ for providers and e-infrastructure user support teams who wish to federated and share applications and services with individual researchers, or small, fragmented communities, typically referred to as ‘the long tail of science’.

The portfolio of applications is currently composed by a readily available set of applications relevant to different scientific and research areas. This portfolio is open to be extended thanks to the contributions of users of the service. If you are interested in this, please get in touch with us: support (at) egi.eu


Intended user groups

  • Researchers and scientists.
  • Application developers who want to share their applications available to the community through AoD.
  • National e-Infrastructures who want to offer the EGI applications library to their national users by "plugging" compute and storage resources.

Requirements and user registration

The service is open for any researchers, scientists and engineers who need a simple and user-friendly access to compute, storage and applications services in order to carry out data/compute intensive science and innovation. The user needs to be affiliated with, or at least have a partner (for example a referee), a research institution in Europe to qualify for access.

Access requires acceptance of Acceptable Use Policy (AUP) and Conditions of Use of the 'EGI Applications on Demand Service'.

Users of the service are asked to provide appropriate acknowledgement of the use in scientific publications. The following acknowledgement text can be used for this purpose:

This work used the EGI Applications on Demand service, which is co-funded by the EOSC-hub project (grant number 777536). The HNSciCloud project (grant number 687614) is also sponsoring the service, allowing users to access the HNSciCloud services pilot for limited scale usage using the voucher schemes provided by the Exoscale contractor.

When requesting access to AoD users are guided through a lightweight registration process. Through a lightweight vetting process, members of the EGI User Support team will perform the needed checks to validate the users' requests before to grant him/her the access to the large scale computing resources.

Service components

  • Scientific applications that are offered ‘as services’ through graphical environments.
  • Science gateways and Virtual Research Environments that offer integrated development environments to port custom applications with cloud resources.
  • Cloud compute resources suited for both compute/data intensive applications and for the hosting of scientific services.
  • A network of Consultants and supporters who can provide guidance on the use of the service.

Service Grant

Each authenticated and authorized User is offered a grant, providing a pre-defined quota of resources, which can be used to run an application of choice.

The EGI grant includes up to 4 vCPU cores, 8GB of RAM and 100GB of block storage.

The grant to run applications is initially valid for 6 months and can be extended/renewed upon request.

Benefits

Through this service:

  • Researchers can access applications underpinned by high-capacity compute and storage servers, to carry out compute and data-intensive simulations. Use of the service does not require advanced experience with distributed and/or cloud computing.
  • Developers can integrate custom applications into the service and offer them as 'scalable, online services' for researchers and scientists worldwide.
  • Providers of compute and storage services can share their clusters and clouds to serve 'the long tail of science'. For more information see AoD for infrastructure providers

Application library

The service operates as an open modular environment where any application developer can integrate and share applications, and providers can plug their own data centre share. The following components are already integrated into the service and are available for users to access:


Computer Science & Mathematics

  • RStudio (v3.2.2) s a free and open-source integrated development environment (IDE) for R, a programming language for statistical computing and graphics..
    • The application is accessible through the FutureGateway Science Gateway. The user manual is available here.

Life Sciences

  • Chispter is a user-friendly analysis software for high-throughput data. It contains over 300 analysis tools for next generation sequencing (NGS), microarray, proteomics and sequence data. Users can save and share automatic analysis workflows, and visualize data interactively using a built-in genome browser and many other visualizations.
    • The application is accessible through the FutureCatania Science Gateway. The user manual is available here.
  • Galaxy is an open, web-based platform for data intensive biomedical research.
    • The application is accessible through the EC3 portal. Instructions on how to use Galaxy workflows with EC3 are available here.
  • NAMD is a parallel molecular dynamics code designed for high-performance simulation of large bio-molecular systems.
    • The application is available through the EC3 portal.

Tools

  • Docker is an open-source project that automates the deployment of Linux applications inside software containers.
    • The application is accessible through the EC3 portal.
  • Hadoop is a framework that allows for the distributed processing of large data sets across clusters of computers using simple programming models.
    • The application is accessible through the EC3 portal.
  • Marathon a production-grade container orchestration platform for Mesosphere's Data Center/Operating System (DC/OS) and Apache Mesos.
    • The container orchestrator is accessible through the EC3 portal.
  • Chronos is a distributed and fault-tolerant scheduler that runs on top of Apache Mesos that can be used for job orchestration.
    • The scheduler is accessible through the EC3 portal.
  • OSCAR is a framework for to efficiently support on-premises FaaS (Functions as a Service) for general-purpose fileprocessing computing applications.
    • The framework is accessible through the EC3 portal.
  • Kubernetes is an open-source container-orchestration system for automating application deployment, scaling, and management. Through this framework users can create a containerized version of workflows and services that can be deployed in the EGI Infrastructure.
    • The container-orchestration system is accessible through the EC3 portal.

Science Gateways and Virtual Research Environments

  • The FutureGateway Science Gateway (FGSG) is a programmable interface of a RESTful API Server, compliant with CSGF APIs specifications, able to provide an easy acces to the PaaS layer by leveraging on recent Web technologies.
  1. The FGSG incorporates several applications and offers these "as services" for the user.
  2. The FGSG provides users with an intuitive web interface to execute dockerized applications on the cloud resources of the platform.
    • The FGSG take cares of starting the job in one of the available resources of the platform, of transferring the needed files (e.g. executable, input files, etc.) and downloading the output(s) in behalf of the users. FGSG is open source and released under the Apache 2.0 license.

The FGSG integrated in the AoD is available here


  1. The cluster, which is defined with a 'job wizard' interface for the user, is composed by a front node, where a batch job scheduler is running, and a number of compute nodes.
    • These compute nodes will be dynamically deployed and provisioned to fit increasing load, and undeployed when they are in idle status.
  2. The list of application libraries/tools installed in the front-node can be exported via NFS in all the compute nodes.
  3. The installation and configuration of the cluster is performed by means of the execution of Ansible receipts.
  4. An Enterprise Linux OS is used for both he front-node and the compute nodes of the cluster.
  5. The cluster provided with AoD is personal - users have root-access to the environment, and can setup and configure the system to their needs. It is meant to be shut down when no longer needed.

The EC3 platform requires that ports 8800 and 8989 have to be accessible from outside.

The EC3 platform integrated in the AoD is available here


  • The EGI VMOps dashboard, a framework allows users to perform Virtual Machine (VM) management operations on the EGI Federated Cloud. The graphical interfaces of the dashboard allows the user to choose VMs from the AppDB Catalogue, to define an execution topology for them, and then to instantiate them on infrastructure as a service clouds.

The EGI VMOps dashboard is available here

How can you access the service ?

  1. Login to the EGI Marketplace with the EGI AAI Check-In service.
  2. Setup a profile, including details about your affiliation and role within a research institute/project/team.
  3. Navigate the marketplace top-menu and click on the category: "Applications".
  4. Click on the "Applications on Demand" service and submit an order to use one of the available applications.
  5. When the request is approved, run the requested application(s) hosted in one of the gateway/portal environment.

Please check the following guide for further details.

Help-desk

For generic questions or inquiries, please contact the EGI Support team at: applications-platform-support@mailman.egi.eu.

For gateway-specific support please contact:

  • FutureGateway Science Gateway (FGSG): fgsg@ct.infn.it
  • EC3 portal: micafer1@upv.es

References

Scientific paper describing the service and its status at May 2017: https://documents.egi.eu/document/3132

Presentations about the service:

  • Overview of the EGI Infrastructure for serving the long tail (EGI Community Forum, November 2015): [1]
  • Poster and animated slides from Demo at EGI Community Forum, November 2015 (Winner of best demo prize): [2]
  • Slideset about the concept of the EGI long-tail of science platform (from Nov. 2014): [3]
  • Slideset about the authentication and authorization model adopted (from Nov. 2015): [4]
  • Slideset about the staus report of the platform at the EGI Conference 2016: [5]
  • Slideset about the staus report of the EGI platform at the DI4R Conference 2016: [6]
  • Webinar to introduce the Applicatios on Demand (AoD) service to NGIs/USTs representatives, RI architects, resource providers and researchers [7]
  • Slideset about the Applications on Demand (AoD) service introduced at IWSG 2017: [8]
  • IWSG2017 Proceeding [9]
  • The EGI applications on Demand service, Future Generation Computer Systems, Volume 98, September 2019, Pages 171-179 [10]