Why A Hybrid Model Is Best


Stratifyd’s AI powered data analytics platform was designed around the cloud. By far the most reliability and flexibility to achieve valuable business insights come from fully utilizing unsupervised machine learning in the cloud. However, at Stratifyd, we understand that full cloud implementation is not a good fit for companies that have highly sensitive information that must remain secure. For those companies we do offer two other options for deployment. First is a hybrid model that employs both the cloud and on site hardware to create a network designed for flexibility and security. Second is a fully on premise model that offers the highest level of security, but fully neglects the adaptability of either a hybrid or cloud model.  Here is a breakdown of how each model works as well as the benefits and costs of each.


The cloud based model allows for the best adaptability while still providing high security encryption of data both at rest and in transit. The cloud based model allows for off hour updates that will cause no interruption in functionality. Additionally, the cloud is ready to deploy in a fraction of the time of the other models, while costing only a per user licensing feed to implement.

The hybrid model is ideal for those companies that require a slightly higher degree of security over the cloud based model. The hybrid creates a encrypted bridge between the on premise hardware that houses internal data, and the external cloud that’s helping to organize outside information and analyze it. Once the additional hardware is in place, it takes no longer to implement than a simple cloud model and provides greater security while retaining adaptability, but with greater cost.

The on premise model is by far the most expensive and time consuming to implement, in that all necessary tools must be on site. The equipment required makes this option generally cost prohibitive, yet does provide a high level of security. On premise lacks the adaptability of either previous models and takes a much greater length of time to update. This should be a model of last resort for those who are heavily regulated away from any sort of hybrid or cloud model.

To understand more about the different models and their effect on data analysis and security, download our white paper here.