In this article, we explained what an SLA should look like and how it supports the SLM process.
We also gave examples to show the key elements in an SLA:
- What the provider is promising.
- How the provider will deliver on those promises.
- Who will measure delivery, and how.
- What happens if the provider fails to deliver as promised.
- How the SLA will change over time
We have shown two templates to illustrate how these principles can be translated in real examples. Remember that every business is unique and that one should be flexible in applying these principles.
We mapped the commitments in an SLA to KPIs to help facilitate the performance metrics definition. These KPIs drive the internal goals and even influence the tool selection of the SLM system as explained in the Service Level Management in the Data Center article.
We have described the importance of SLAs to avoid the most common pitfall as quoted by nextslm.org: "The IT industry has a history of over-promising and under-delivering. This history repeats itself in the ASP model, with promises of absolute (99.99999%) reliability, global availability and rock bottom costs."
The benefits of a good SLA are many, but the most important ones are:
- Sets clear customer relationships
- Sets goals for internal organizations
- Sets a framework for continuous quality improvement
- Drives the SLM system architecture
We ended with some suggestions to assure the creation of good SLAs. The most important one is to keep the SLA simple, measurable and realistic.