13 November 2014 blogs Daniel Örneling 8min read
SLA – What´s that and why do I need it? by guest blogger, Daniel Örneling
These days, the cloud is on almost everyone’s lips and everyone’s talking about moving to the cloud. When speaking of moving to the cloud, there is a lot to think about. What parts of our environment can we move to the cloud? How do we do it? And who should we choose to host our environment?
Those questions are just a few that need to be answered before any decisions can be taken about where to move the data. Moving to the cloud might involve everything from moving the servers to migrating the mail boxes into Exchange Online using Office 365. When moving to the cloud is under consideration, another question that definitely needs to be considered is the level of availability for the service you are about to move to. When moving to the cloud, you´d want to increase the up-time for your environment instead of the opposite.
What is an SLA?
An SLA (Service Level Agreement) is a number that shows how much up-time the service promises to deliver during the month (or year). For you as a customer, the up-time is very important when you rely on other companies, like Microsoft, to deliver your critical services. A typical number for a SLA is 99.9% but it may vary from case to case. So what can you expect when committing to buy a service like Microsoft Azure?
Looking at Azure´s SLA for Virtual Machine gateways, the SLA says that it should be available 99.9% of the time per month. A great number of course, but what does it mean in reality?
Based on 30 days per month, the service will have to be up and running for 43 157 minutes/month for the SLA to be fulfilled. This means that the gateway can be down for a maximum of 43 minutes a month before the SLA is breached. That´s a very good number and you know what to expect from the hosting provider.
I´m a hosting provider, how can I prove to my customers that their SLA is fulfilled?
When delivering a service, take a web site for example, you need to be able to present reports to the customer showing the uptime for the service they pay you to deliver. In this post I will explain how you can do this using System Center 2012 R2 Operations Manager and Savision’s Live Maps. The first thing you need to do is to write down what components the service relies on. The usual components to monitor for a web site are:
- The actual web site, monitor that it´s actually possible to browse
- Web server instances, IIS for example
- SQL Server instances and databases
- Network devices
- The servers hosting the above components
When all of the above components are being monitored, you can start thinking about building an overview of the services that you deliver to your customers. This is where Live Maps comes into the picture and below, you will see what the service dashboard looks like after having created the website services in Live Maps.
Live Maps Services Dashboard
Below is a drill down into the service where you can find the service levels for both the service itself and also for all the components.
Live Maps Service Map drill dow
The above dashboard shows the state of the service that is being delivered and this is what you can present to your customers or your own business executives to demonstrate the fulfillment of the SLA. What you can see above is an overview of the service itself, but also three other components.
- The first component is the End User Component which represents what the user sees, the web shop start page for example.
- The second component is the Application Component which represents the applications needed for the service to be up and running. This might represent the databases and web servers in a web shop scenario.
- The third and last component is the Infrastructure Component which represents the underlying infrastructure such as the servers and the network switches.
When looking at the service, you can see the SLA level for the complete service but also per separate component. This way you can easily point out which components needs attention.
When using SCOM without Live Maps to monitor your SLA´s, you will only get the SLA levels per service and not for the underlying components as when using Live Maps.
If you would like to monitor a web site as is the case in this blog post, SCOM offers two different monitor types:
- Web Application Availability Monitoring is used to monitor the response times and whether the web page is actually delivered. This is very helpful for the web server team as it shows what HTTP codes the web page responds with.
- Web Application Transaction Monitoring can be used to record a browser session where a web site is browsed the same way all the time and based on how long it takes, a warning or a critical alert can be generated.
You can always create a fancier custom view for every customer showing their. See below for an example:
Live Maps CIO Dashboard (click on image to enlarge)
New features in the latest release of Live Maps
During TechEd Europe that took place in Barcelona in the last week of October, Savision released a new version of Live Maps. A few of the new features include:
- Out-of-box Live Maps for some of the most regular services: Exchange, SharePoint and Active Directory meaning you don´t have to build these services by yourself. As long as the services are implemented in the environment, Live Maps can discover them and include the right components to build the service so that it can be presented to the company along with the service level agreements.
- Show performance data in the context of a Diagram or List view. In the previous release of Live Maps, the maps showed the health state, alerts, service level agreements and properties of the objects contained in the service. Added to that list is now performance data for the service. This means that you will no longer need to leave the service overview to find out more about the performance, it can instead be done directly via the Live Maps.
- .Net Application Discovery can help you discover all .Net applications running in your environment. By discovering the applications running, you will get a clear picture of what´s actually out there in your environment. Not only are the applications discovered but also the connection to databases which gives you a clear picture of all the relationships between the components as well. Another great part of this feature is that Live Maps will automatically keep the service maps up to date for you without interaction.
- Dynamically updating services makes life easier as it automatically keep track of changes to the services. Often when you´ve integrated Operations Manager with third party products there are also dynamic distributed applications created. Since these are automatically updated, Live Maps can now automatically update your services as well which leads to service maps always being up to date.
- Service Manager Integration has been around for long in Live Maps but has now been evolved and your incident and change requests can automatically be updated with which services they affect. You can now also open the service map from the Service Manager console and if you´re running Cireson’s portal for Service Manager you will now get the Live Maps information directly inside the web console.
As a service provider, other than having an SLA to live up to, being able to show to the customer that you fulfill the SLA is crucial. What I have shown above is what I think is the best way to deliver the SLA and how to visualize it for your management or your customers. If you’re using Operations Manager or consider using Operations Manager you should also think about how to visualize the service status and how to present it to your management or your customers. Using Operations Manager, all of these views can also be presented on a SharePoint site for easy access to the information.
Want to know more about what Live Maps can do for your organization? Find out here.
About Daniel Örneling
Daniel Örneling is a specialist consultant working for Approved Consulting. He focuses on SCOM, OMS and those parts of Azure that come along with it. Follow his blog if you want to learn more about his tips and fixes for System Center Operations Manager, Operations Management Suite and much more.