Many customers ask if they should be multi-cloud? People who choose this option are typically looking for redundancy and reducing risk of their entire system going down if AWS, Google Cloud or Azure go down or have an outage.
I get it, we don’t want to see anyone’s system go down – the cloud is supposed to improve reliability and scalability. What you need know are some of the challenges of being multi-cloud, as well as that there are ways to architect the system so it’s effectively impermeable to an outage unless the entire public cloud goes down.
Here are some problems with multi-cloud:
1. Performance – All these clouds are not designed to work together. Moving data and compute back and forth between clouds is typically very slow and clunky. You do have some very specific workarounds, but they may introduce more risk than solve.
2. Double the Skillset – Your team will need to understand both clouds and that’s a lot to ask of any team. The team would need double the skillset at an expert level to run a multi-cloud solution 24/7, which most companies can’t sustain.
3. Interoperability – As I said in #1, AWS is designed to work with other AWS components; Azure is designed to work with on premise, hybrid and all the other Azure components, as well as the Microsoft enterprise solution. Because Azure is a better cloud for enterprises is one reason why we see Azure deployed more often than AWS.
So, what do we recommend? Choose the cloud based on the services that you’re going to get the most value from. For our customers doing a lot with data, they go with Azure because all the data services and the Azure data platform are built to work together. It’s also made to work globally across the different regions that Azure brings to the table, as well as work in the government cloud and in a hybrid mode.
There are many ways it can be deployed to ensure you’re not facing the fears of an outage or increased risk. To make the best choice, look at:
- The types of services and the value you’ll get from them.
- The ease of migration.
- What your contingency plans would look like.
- Automation – What would it look like? Being able to automate those things and automate testing of them, as well as automate faking failures and failovers and going through a full cycle test. Can you do that in your cloud?
- How is identity and security going to work.
Bottom line is before you jump into all the things that have to do with a true multi-cloud set up, make sure you truly understand what your Azure cloud can do for you to mitigate all that risk. Most people don’t have an in depth understanding of all the opportunities and options inside Azure. But remember this, Microsoft runs all their machine critical business on Azure—kind of says it all right? They’ve obviously found great patterns to ensure they don’t have outages or issues.
Need further help? Our expert team and solution offerings can help your business with any Azure product or service, including Managed Services offerings. Contact us at 888-8AZURE or [email protected].