Imagine a home, maybe your own. Some rooms are exclusive perhaps to you – your bedroom (your space) and some are common areas – like the living, dining, kitchen and so on. Such a perfect blend of privacy when you want to be left alone along with a place to hang out together in the common area and socialize when necessary. Now compare this to living in an isolated log cabin – pretty much cutoff from everyone – totally private. As opposed to a railway station waiting area or an airport lounge – eat-drink-sleep but out in the open? Perhaps not the best way to put it, but a Hybrid cloud strategy makes use of both Public Clouds such as AWS, Azure, Google Cloud, etc. for some computing needs of the organization whereas retaining the rest on Private Cloud, mostly on-premise Data Centers or Enterprise Data Centers as explained in my previous blog.

Why would an organization go for this Hybrid Strategy? There could be multiple reasons such as

1. The organization is in its early stage of cloud migration, testing the waters and does not wish to go whole hog in one shot.
2. The organization does not wish to offload certain workloads (computing resources) to the Public cloud – perhaps due to security concerns or compliance needs.
3. Simply wishes to retain more control on certain workloads – perhaps with the surety that it can be more responsive and able to react with more agility in times of emergency or if the workloads need highly specialized talent to manage and Public Cloud operators could be found wanting for this need.
4. There is definitely a possibility of some cost savings using on-premise solutions over the long term if managed well – recall my small digression when I mentioned buying a car outright (still the most popular method) as compared to leasing arrangements with fine prints.

By going with a Hybrid Strategy, it is easy to see that it looks like it is the best of both worlds. And frankly it is. For many chosen workloads, an organization can go for the Public Cloud that offers far superior scalability and flexibility with a pay for only as much as you use model. These workloads could be experimental in nature so why procure infrastructure for it or the workloads that could be non-core workloads. And perhaps for mission critical workloads, use an on-premise solution with many of the bells and whistles of Public Cloud with a robust Failover and DR strategy but with right technology talent.

But aren’t there any downsides? Certainly there are a few to consider such as

1. Implementing Hybrid Could correctly is a bit complex as the organization straddles both sides – public cloud and on-premise. This requires a more thoughtful and planned approach and anything rushed could cause serious long term issues.
2. Integration could be a challenge. If Data is the Oil of the 21st century, there is a possibility of data getting a bit siloed and getting the most out of it could prove difficult. It may start with latency caused between the two environments as well as lack of visibility due the inherent nature of the environment being in two or more different entities. But there is more.
3. There is no stopping an organization choosing multiple Public Cloud providers that then further compounds the problem and makes Hybrid more complex and difficult.

These are some of the common problem areas but as this was envisaged quite early, there are good tools/solutions to handle them along with a growing number of talent/expertise who specialize in making it easier to manage. This has made the Hybrid Strategy the winning strategy in the Cloud Adoption Lifecycle in my humble opinion.