The SLA signed between your organization and the cloud service provider is the key constituent in concluding the deal. Since SLAs are critical for working with cloud infrastructure, let’s first examine some of the basic points that a cloud SLA should include. We may then proceed to understand details of cloud SLAs.
SLA’s terms common to all cloud service providers:
- List of parameters and minimum levels required for each element of the service they provide. For e.g., availability, performance, location of data and so on.
- Outline the alternatives in case of a failure in meeting service committment.
- Detailed information about the system infrastructure.
- List of security standards maintained by the cloud service provider.
- Provision that allows you to audit their compliance.
- Indications about your rights and ownership of applications, data and content including intellectual property and proprietary rights.
- Specification regarding fee structure and the rights to continue or discontinue the service.
Understanding complex cloud SLAs:
To understand complex SLAs you need to decode them. Once you understand you can negotiate the SLAs so that you can extract the most out of cloud computing services. These services are flexible and predictable with an assurance of 99.9% up-time and a flawless service. What if the assurances are not met? What remedial steps do provider take in case of failure to meet the service levels? This is where the need and importance of studying the SLAs arises. Cloud services can be considered as an extension of your organization’s IT department. You should familiarize yourself with some of the SLA definitions mentioned below for a deeper understanding of cloud SLAs.
- Downtime: During this period the cloud service does not operate. The reasons may be a malfunction or a natural disaster.
- Downtime Period: This period is ten consecutive minutes of downtime for a domain. Less than ten minutes will not be counted as Downtime Periods.
- Availability – The time a system is fully functional.
- Latency – Time delay experienced in a system.
- Throughput - The information that travels between two systems in a period of time.
- Time-to-response – The time interval between the moments a person enters the request and the instant at which the first characters appear.
- Monthly Uptime Percentage: Total number of minutes in a month minus the total number of minutes of downtime in the same month, divided by the total number of minutes in that month.
- Scheduled Downtime: A maximum of twelve hours of Scheduled Downtime per calendar year.
Vital points to consider in SLAs
Your IT manager who is responsible for signing the deal should study the SLAs carefully and be very specific on the following crucial terms and conditions.
Performance, Uptime and Downtime: Most providers assure permanent uptime, but in situations like a natural disaster, you need to know your vendor’s preparedness and what the SLA says about their role in supporting your IT operations in case of a disaster. You should be aware of the response times and resolution timeframe in case of an outage. Generally weekly maintenance periods are not included in the calculated downtimes by vendors so you need to include that time too for your internal downtime calculations. Zero downtime is an ideal situation and SLA should define an outage in clear terms. Let’s take a look at the variations offered by cloud service providers.
- 99% uptime implies 7.5 hours of downtime per month
- 99.5% uptime implies 3.5 hours of downtime per month
- 99.9% uptime implies less than 45 minutes of downtime per month
Many SLAs make provisions for at least an hour’s scheduled maintenance time per week. Ensure that it is scheduled at a convenient time for your users . It is necessary to discuss worst case scenarios extensively and include details of tools that are used to clear outage, the number of staff deployed and incident reporting systems.
Security: For every business security is a top priority and when you hire a cloud service provider you entrust activities related to email, applications or data to them. The SLA should list specific details about their technical staff handling their data and the measures taken to protect their data. Though SLAs may not include such clauses, it is for you to claim the ultimate authority over the data and measures for settling security breaches and vendor responsibilities.
Cost: SLAs broadly define the basic fee structure with external fees as additional costs. It is these external fees that are hidden costs. License, upgrading, taxes and additional cost for exceeding storage limits may be a part of external fees. It is for you to dig deep and gain clarity before you strike a deal.
Why we are emphasizing so much on cloud computing SLA? Well, in past few years cloud has proved its potential and hence there are numerous vendors selling and reselling cloud services. Searching cloud solutions with the product related keywords like hosted exchange , hosted Sharepoint or VPS Hosting yields millions of results. How to find the efficient vendor? Here comes the need to evaluate SLA with cloud vendors. You can always have the basic idea of the services and reliability from the SLA . To summarize, cloud computing comes with its own challenges such as security, networks, storage, processing capabilities, software availability, and legislation changes and therefore it affects productivity, costs and growth of the organization. In such a scenario it is all the more important for to find a cloud service provider with SLAs that effectively addresses these challenges. You may always negotiate if you feel that a particular service level does meet your expectations. Use your resources in vendor management to address the risks and your organization’s relationship with the service provider.