Difference between revisions of "Category:NCBO Virtual Appliance"

The NCBO Virtual Appliance image contains a pre-installed, pre-configured version of commonly-used open source NCBO software running on a Linux operating system. The image was created using VMWare and is available for download by contacting support@bioontology.org.

The following software is included on the image:

• NCBO Web Services (BioPortal Core)
• BioPortal Web User Interface (including ontology visualization, Flex widgets, Annotator and Resource Index UIs)
• BioPortal Admin (a UI for administering BioPortal Ontology Services)
• Annotator
• Resource Index

Please see below for how-to documentation for managing the software and running data population for Annotator and Resource Index.

Getting Started

• To obtain the Virtual Appliance, contact NCBO Support and include your BioPortal account username. You can create a BioPortal account at: http://bit.ly/bioportal-account
• The download is provided as a tar archive containing several files. One of these is an Open Virtualization Format (OVF) file that may need to be converted to work in your virtualization environment.
• You can supply the hostname (machine name) for the virtual machine during the deployment process. Documentation will refer as this hostname as 'example'.
• Change default passwords
  • Operating System
    • Username: root
    • Password: password is prompted on the first boot
  • BioPortal Admin User
    • Username: admin
    • Password: changeme
• Add an ontology using the BioPortal Admin User here: http://{YourDomainName}/ontologies/new
  • BioPortal Ontology Services can be configured to automatically process new ontologies every hour at 30 minutes past the hour (see documentation for enabling the scheduler). This processing includes:
    • Parsing any new, unparsed ontologies
    • Calculating a set of metrics for these ontologies
    • Indexing these ontologies for use with search
• Parsing, indexing, metrics calculation, and ontology deletion can all be manually triggered or re-triggered using the BioPortal Admin user interface, which is available here: http://example:8080/bioportal_admin
• REST services are available at the following locations. The URLs below are for the root of the REST service and generally will not return information. Please read the documentation for guidelines on how to interact with the services, including sample URLs:
  • NCBO Web Services - http://example:8080/bioportal (documentation)
  • Annotator - http://example:8080/obs/annotator (documentation)
  • Resource Index - http://example:8080/resource_index (documentation)

System Requirements

The following requirements are for the resources that you devote to your Appliance instance, not for the machine running your host environment. For example, if you are using a system with 4GB of RAM, then you will need to devote all of that RAM to your guest Appliance.

Note: these requirements are for basic usage. System requirements will vary greatly depending on the size of the ontologies you work with, the number of ontologies in the system, and the number of concurrent requests that the system needs to respond to. It can also vary depending on how the ontologies are used. For example, the search index can be RAM-intensive but parsing ontologies can be CPU-intensive. You will need to experiment with your Appliance resource settings to find what works for your scenario.

• Minimum
  • 1 CPU (2 GHz)
  • 4GB RAM
  • Hard disk space: 3GB

• Recommended for heavier usage
  • 2 CPU (3 GHz)
  • 6GB RAM (When using more than 4GB RAM, you should configure Tomcat to use the additional memory)
  • Hard disk space: 10GB (or more depending on number/size of ontologies)

Population Workflows

When storing an ontology using the BioPortal Ontology Services, some amount of processing is done automatically so that the ontologies can be visualized using the Web UI. However, the Annotator and Resource Index require separate, manual data population workflows in order to function with the provided ontologies. Details on running these populations can be found in the "Howto" sections located at the bottom of this page.

NOTE: Loading ontologies [1] developed using the NCI version of Protege will require an additional manual pre-processing step to remove XML tags added into the synonym and definition property values. This pre-processing step is especially important for populating the correct synonym value into the Annotator workflow. The code to pre-process ontologies developed using the NCI version of Protege is located at: https://bmir-gforge.stanford.edu/gf/project/bp_helper_tools/scmsvn/?action=browse&path=%2Ftrunk%2FNCItProcessor%2F, the code can be checked out via anonymous SVN.

[1] The NCI Thesaurus and the NanoParticle Ontology are developed using the NCI version of Protege

Image Format and Operating System Details

The NCBO Appliance image was created using the Open Virtualization Format, which should allow the machine to be used in a variety of environments.

The operating system is CentOS 5.7 64-bit running Tomcat 6.0.26, Java 6, MySQL 5.1.x, PHP 5.1.6, Rails 2.3.x, and Ruby Enterprise Edition 1.8.7, and memcached.

The following applications use these services/runtime environments:

• Tomcat, Java
  • BioPortal Ontology Services (BioPortal Core)
  • BioPortal Admin (a UI for administering BioPortal Ontology Services)
  • Annotator and Recommender
  • Resource Index
• Rails, Ruby, memcached
  • BioPortal Web User Interface (including ontology visualization, Flex widgets, Annotator and Resource Index UIs)
• PHP
  • Resource Index UI
• 4store
  • Mappings

Basic System Administration

• Most of our administration scripts and build environment assume that you will be running as the root user.
• Helper commands:
  • ncbostart: the required services are started on boot automatically, but if they need to be started automatically
  • ncbostop: manually stop services
  • ncborestart: manually start/stop services
• Start individual services:
  • /sbin/service httpd start
  • /sbin/service memcached start
  • /sbin/service tomcat6 start
  • /sbin/service mgrep start
  • /sbin/service 4store start
• Stop individual services:
  • /sbin/service httpd stop
  • /sbin/service memcached stop
  • /sbin/service tomcat6 stop
  • /sbin/service mgrep stop
  • /sbin/service 4store stop

License

All NCBO software is released with the 2-clause BSD license. Source code is included on the Virtual Appliance.