Difference between revisions of "VO installation"
From Gcube Wiki
Line 15: | Line 15: | ||
##* deploy a [[VRE_Management_Installation#VREModeler|VREModeler]] instance and configure it to join the VO scope | ##* deploy a [[VRE_Management_Installation#VREModeler|VREModeler]] instance and configure it to join the VO scope | ||
##* deploy a [[VRE_Management_Installation#Resource_Manager|Resource Manager]] instance and a [[VRE_Management_Installation#Resource_Broker|Resource Broker]] instance and configure them to join the VO scope | ##* deploy a [[VRE_Management_Installation#Resource_Manager|Resource Manager]] instance and a [[VRE_Management_Installation#Resource_Broker|Resource Broker]] instance and configure them to join the VO scope | ||
− | ##* deploy a [[VRE_Management_Installation# | + | ##* deploy a [[VRE_Management_Installation#Software_Gateway|Software Gateway]] instance and configure it to join the VO scope; this is an optional step, since also the infrastructure level Software Repository can be joined to the new VO via the Resource Management tool |
##start the 3 containers following the order of the deployments and verify that they work properly | ##start the 3 containers following the order of the deployments and verify that they work properly |
Latest revision as of 14:23, 20 November 2014
These are the steps needed to setup a new Virtual Organization:
- select the gCube infrastructure to join
- decide the scope hierarchy, i.e. the name of the VO within the new infrastructure
- identify at least 8 machines to turn on as gHNs (however, their number may vary depending on the VO needs). These machine will serve the needs for a typical exploitation scenario of the VO: (i) 3 machines for the enabling services, (ii) 3 machines for the VO services (Content Management, Indexes and Metadata Management), and (iii) then consider, 2 machines for each VRE that is expected to be deployed within the VO
- prepare a Service Map file for the VO scope
- install the VO Enabling Services:
- install gCore in 3 machines and copy the Service Map files under the $GLOBUS_LOCATION/config folder
- configure the first gHN to join to the VO scope and
- deploy the IS-Collector and configure it to join the VO scope
- configure the second gHN as ROOT and to join to the VO scope and
- deploy a IS-Registry instance and configure it to join the VO scope
- deploy a IS-Notifier instance and configure it to join the VO scope
- configure the third gHN to join it to the VO scope and
- deploy a VREModeler instance and configure it to join the VO scope
- deploy a Resource Manager instance and a Resource Broker instance and configure them to join the VO scope
- deploy a Software Gateway instance and configure it to join the VO scope; this is an optional step, since also the infrastructure level Software Repository can be joined to the new VO via the Resource Management tool
- start the 3 containers following the order of the deployments and verify that they work properly