Service Level Agreement (SLA) Policies

Introduction

When describing the OpenAIRE Infrastructure Service Level Agreements (SLAs) we refer to the levels of availability, serviceability, performance, operation, or other attributes of the Infrastructure. The level of service is specified as "target" and "minimum," which allows the user of the OpenAIRE Infrastructure to be informed what to expect (the minimum), while providing a measurable (average) target value that shows the level of organization performance.
The following OpenAIRE Infrastructure SLAs have been defined:

• Content Access SLA,
• Content Storage SLA,
• Applications as-a-service hosting SLA,
• Support Response Time SLA,
• Scheduled Infrastructure Downtime SLA.

Content Access SLA

The OpenAIRE APIs provide access to a graph of interlinked metadata records and to a corpus of article PDFs. While metadata records are accessible as CC-BY, access to PDFs may only take place on request and is subject to restrictions based on the content providers.
If Users create technology that works with OpenAIRE API Services, they must comply with the current technical documentation applicable to the used OpenAIRE Services available at http://develop.openaire.eu.
All API services are running in production on a best effort basis, targeting 24/7, within the OpenAIRE infrastructure premises deployed at the data center facilities of the Interdisciplinary Centre for Mathematical and Computational Modelling (ICM) - Data center facilities at ICM, Poland.

Content Storage SLA

OpenAIRE users

Users may submit two kinds of information to the OpenAIRE infrastructure: personal sign in information or claims, intended as assertions of associations between research products (publications, datasets, software, other products) and between research products and research projects or research communities.
The OpenAIRE Infrastructure will make reasonable efforts to ensure that data are persisted. In the event of hardware or software failures caused by failures to a hard drive or power supply, the OpenAIRE Infrastructure will make reasonable attempts to restore the user's data. No guarantee whatsoever is provided on the success of any User's data recovery.

OpenAIRE Content Providers

Content providers submit two kinds of information to the OpenARE infrastructure: personal/organizational sign in information and operational information to access their metadata and full-texts.
The OpenAIRE Infrastructure will make reasonable efforts to ensure that data are persisted. In the event of hardware or software failures caused by failures to a hard drive or power supply, the OpenAIRE Infrastructure will make reasonable attempts to restore the Content provider’s data. No guarantee whatsoever is provided on the success of any Content provider’s data recovery.

OpenAIRE applications as-a-service hosting SLA

OpenAIRE may offer applications as-a-service (e.g. aggregators and brokers), whereas external organizations operate/administrate instances of OpenAIRE services hosted by the OpenAIRE infrastructure. Once deployed, such instances will deliver a best effort uptime as monitored within the OpenAIRE Infrastructure monitoring system. The following holds:

• Only failures due to hardware and hypervisor layers delivering individual servers constitute Failures and so only those Failures are covered this SLA. Examples of Failures include power interruptions, hardware problems such as failures to a hard drive or power supply, and failures to the hypervisor environment supporting User Web Applications.
• Individual aggregators will be monitored by Prometheus/Nagios and monthly availability rate below 99.5% will require a further extraordinary approval managed via an incident ticket through the OpenAIRE Support RedMine system.
• The OpenAIRE Infrastructure will make reasonable efforts to ensure that server storage is "persistent" by performing overnight dumps of the content. In the event of hardware or software Failures as defined above, the OpenAIRE Infrastructure will restore the previous pre-Failure state including restoring the Application software as configured by the User and Data. Furthermore, these efforts will include supporting the User via chat or Email at the User's choosing. No guarantee whatsoever is provided on the success of any Users recovery.

Support Response Time SLA

The Production System Support Team (ISTI-CNR, Italy - ICM, Poland) is responsible for the resolution of all production support requests related to the OpenAIRE production system. This SLA covers incident tickets in the case of issue related to infrastructure incidents (e.g. it does not cover the resolution time of the issues coming from software bugs). In case of incident, a reply time of at most 1 business day is ensured, including the initiation of a resolution procedure. The resolution time may depend on the specific incident (e.g. on warranty terms from hardware suppliers, which are generally longer than 1 day).

Scheduled Infrastructure Downtime SLA

• Scheduled Infrastructure Downtime User notice > 24 hours.

 For more, please read D9.2 - Infrastructure Policies