SG SLA
From EGEE-see WIki
Contents |
SLA between Sites and the SEEGRID-2 Project
1. Purpose
This Service Level Agreement (SLA) between Grid Resource Centers (Sites) and the SEEGRID-2 project is to ensure that sites will provide the necessary quality needed in order to satisfy the requirements of the SEEGRID infrastructure and its applications.
2. SLA Structure
The SLA is structured in 5 different sections that are related to specific issues that need to be addressed in order for sites to satisfy the requirements of the SEEGRID infrastructure.
The above mentioned sections are summarized in the following:
- Hardware and connectivity criteria
- Level of support
- Level of expertise
- VO support
- Conformance to Operational Metrics
3. Hardware and connectivity criteria
1. The Site must not have less than 4 CPUs. Service nodes should be extra.
2. The Site must have enough connectivity to pass SAM and GSTAT tests.
3. The Site must have enough connectivity to support the SEEGRID VO and its applications.
4. Service nodes (CE, SE, Site BDII etc) must have sufficient computational and storage resources to support normal operation of these services.
4. Level of support
1. The Site must provide one dedicated site administrator who will also take role of security administrator.
2. The name and contact details of the site administrator has to be provided and be available in the HGSM tool.
3. The Site must provide administrators with minimum availability 9-5 on weekdays.
4. The site administrators must be available to react to major security incidents that might affect the whole infrastructure during weekends and holidays.
5. Site administrators must respond to operational tickets within next working day
6. Site administrators must produce periodic reports every 3 months to be sent to country GIM.
5. Level of expertise
1. The site must provide one experienced site admin.
2. The site must provide one experienced network support person, or a direct link to network support / network operations center
3. The site must make available one security-oriented person, for advice any time
4. Names and contact details (e-mail) of the above people should be available to SEEGRID partners via the HGSM tool.
6. VO support
1. The Sites needs to accommodate successfully one or more of the applications that are associated to the SEEGRID VO
2. Support for the SEEGRID VO should not conflict with the existing VO allocation policy of their NGI.
3. The site must the OPS role of the SEEGRID VO.
7. Conformance to Operational Metrics
1. The site must conform to the following Operational Metrics.
a) Site Availability (Quality Metric) – Sites must have 90% availability during uptime in a given quarter (a period of 3 months). The site availability is calculated as follows: If a site has degraded performance during a given day (>50% of the SAM tests fail) then the site is considered down for that day.
b) Site declared Downtime – Sites must not be in downtime for more that 10% of the time in a given quarter (a period of 3 months). Extended downtime for more that 10% for a given month is permitted when the case is well justified in written form.
8. SLA Enforcement
1. Monitoring via SEEGRID monitoring tools and helpdesk
2. Sites that fail the obligations stated in this SLA will be discussed at the next SEEGRID-2 PSC.
3. The decision might include suspension of the site from the SEEGRID infrastructure if the PSC finds it appropriate.
4. The responsible for enforcing the obligations stated in this SLA is the country GIM.
5. If for any reason the GIM does not properly enforce the SLA obligations then the SEEGRID-2 WP3 leader should take this action.
9. General Information
1. The SLA must be signed by a legal representative who is responsible for the department of the institution that hosts the site and the SEEGRID project. However this SLA is not a strict legal document.
2. The validity of the SLA is form the date of signature to the end of the SEEGRID-2 project.
3. A Site can be temporarily suspended by the WP3 leader or the SEEGRID-2 project for security reasons.
