When it comes to the hybrid cloud — a cloud computing environment that combines on-premises, private cloud, and public cloud services with orchestration between the platforms — we are living in exciting times. More CIOs are starting to realize the potential that the hybrid cloud can offer, but companies that have actually capitalized on the hybrid cloud remain in the minority.
This means that the hybrid cloud stands to be one of the most transformative forces in information technology (IT) over the next couple of years. In the near future, when you look at your IT infrastructure, what you have will depend on a lot on how you handle your hybrid cloud integration now. The following are some of the common mistakes you should avoid.
1. Using networking technology that isn’t mature enough to support integration
When you attempt to integrate a hybrid cloud, you’re really asking a lot of your networking technology. When you first begin to move to the cloud, a standard Internet connection may be good enough to get the job done. But with a hybrid cloud, it takes much more.
Make sure you have a networking infrastructure in place that is robust enough to support the needs of a production hybrid cloud infrastructure. Without the ability to connect internal and external components properly, the whole concept of a hybrid cloud breaks down.
Free eBook — Hybrid IT: Make Colocation Your Conduit to the Cloud
2. Failing to test effectively
Testing a hybrid cloud infrastructure is something that almost everyone talks about doing, but most people don’t actually do. The consequences of failing to test your cloud infrastructure appropriately can be far-reaching and can include unwelcome surprises after you deploy it. Testing should take place against a concrete set of objectives. Write these criteria out before you begin your testing so that you’ll have a clear picture of what constitutes failure and what constitutes success. This can go a long way toward avoiding the dreaded “early morning phone call” when something goes wrong.
3. Focusing too much on location
Most people spend little or no time thinking about where the websites they visit on a daily basis are hosted, so why should your hybrid cloud infrastructure be any different? It is true that locating assets in different parts of the country can lead to low latency; for instance, the round-trip latency between Virginia and California is about 90 milliseconds.To meet the latency needs of many of today’s IT strategies, reducing the distance that data travels for some businesses critical. But beyond latency, the operational benefits of a solution which can minimize communications costs while reducing failure points are equally compelling.
However, we have technologies available today that make it easy to address the issue of latency. As a result, you can now have workloads on different sides of the country that work the same as if they were located side by side. This means that location shouldn’t even enter into your consideration when you’re in the process of planning cloud services.
Free eBook: Security and Compliance in the Cloud
4. Getting the biggest cloud server you can
Have you ever noticed that the price of laptops hasn’t come down that much over the last decade? The hardware has come a long way during that time, so it would only make sense that prices would go down as a result. The reason they’ve stayed relatively static has more to do with added hardware features and expensive software. Overall the software overhead on actual computer resources has decreased over the years, while the actual hardware performance continues to improve. Think about how long it takes to boot your laptop today as opposed to five years ago.
The implications of this concept can also be applied to a hybrid cloud: sometimes, it doesn’t make sense to go for the biggest cloud server that’s available to you. Finding the cloud server that’s appropriately sized for your infrastructure can help you get a similar level of results, without having to overpay in order to do so.
5. Failing to take advantage of hybrid storage
Not all servers can be virtualized, but you can still take advantage of the next best thing: hybrid storage. This means keeping all of the data from a server on a storage array that includes multiple disks and controllers. This provides a practical and cost-effective way of ensuring redundancy, without having to virtualize the server. Now, anytime there’s a problem with one server, you can easily add another server to the array to replace it with no data loss or degradation of service.
Hybrid storage can be an effective part of your hybrid cloud infrastructure. Failing to take it into consideration is doing yourself a severe disservice.