It was over three years ago I first heard the term “Data Gravity”. I remember my first reaction being “Great… another buzzword… just what we need!”. However, over the years I have come around to find the concept very useful and even appreciate its self-explanatory nature. The idea is simple – data has “gravity”, which means larger datasets will attract smaller datasets (and as a result more applications) towards them. For example, if you have a data lake that contains 15 of the 20 datasets required for your new analytics application, it is easier to merge the other 5 datasets into the data lake than exporting 15 datasets into a new system. It would also be sensible to point your application to consume the data from the data lake. Sounds like common sense, right? So what’s the problem?
“We are excited about the collaboration with Robin Systems,” says Dean Paron, Director, Azure Storage at Microsoft Corp. “This solution helps us migrate applications alongside data for Azure non-native applications like Mongo DB and Casandra into the Azure Cloud.”
The Migration Dilemma
Data gravity has become a significant hurdle for many infrastructure modernization projects. CIOs have issued a clear mandate to migrate all workloads to the cloud. In order to achieve this, the IT and DevOps teams follow a two pronged approach. First, all new applications should be deployed in the cloud. Second, existing applications need to be carefully migrated to the cloud. Eventually, all your applications will be running in the cloud. This strategy would make sense – if data gravity didn’t exist. Often new applications are added on-premise due to the massive amount of relevant data being stored in on-premise data systems. This means new applications are added to the migration pipeline and infrastructure modernization becomes a never ending process. To avoid this, you can try moving the data to the cloud, but moving petabytes of data to the cloud over the network is very challenging and expensive. Moreover, existing applications may not work as-is in the cloud, leaving you to run them on-premise and causing latency issues.
Data Migration with Azure Data Box
Microsoft Azure Data Box is a device that enables the transfer of large amounts of data to Azure with limited time frames, eliminating network bandwidth issues and reducing costs. The device is ordered through the Azure portal. Customers then manually copy the on-premise application data to the device and send the device to Azure data center. Data is uploaded to Azure through a secured end-to-end process. This simplifies the migration for massive amounts of data, where moving the data over the network may take months and consume all your network bandwidth. It is also very cost effective and secure.
ROBIN Hyper-Converged Kubernetes Platform + Azure Data Box: Migrate Applications, Not Just Data
Robin is one of the first solutions that can do a “lift-and-shift” transfer of SQL, NoSQL, and Big Data applications into Azure by leveraging Azure Data Box. Robin is designed from ground-up to define, deploy, and manage (scale, move, snapshot, clone, migrate) entire applications, not just the data volumes or file systems.
Robin captures the application configuration and metadata, SLA policies, topology (network and deployment) and data in a single “application snapshot”. This “application snapshot” can be transferred anywhere and be used to provision a new instance of the application that has the exact same setup, state, and data. Complex distributed or clustered apps can then be deployed, copied, moved or migrated with 1-click. Once migrated, customers can interact with the application as if it were the original on-premise application.
Migrating existing applications to cloud can be very complex, especially if there are multiple data systems and massive amounts of data involved. ROBIN Hyper-Converged Kubernetes Platform combined with Azure Data Box goes a long way in alleviating the pain of migration for the DevOps/IT teams. Download the solution guide to learn more about how you can set up a 1-click migration process.
with Azure Data Box & ROBIN Hyper-Converged Kubernetes Platform – Solution Guide
Today migration of Enterprise Applications from on-premise to cloud is a costly, time consuming and error-prone manual effort that involves saving and transferring only the application data to the cloud, re-configuring the on-premise application setup to work with cloud compute, network and storage APIs and then launching the application in the cloud. Microsoft and Robin Systems have worked together to automate this entire process by integrating Azure Data Box and ROBIN Hyper-Converged Kubernetes Platform. Customers can now save effort, time and money with 1-click migration of on-premise applications including all the application data and its state to get an identical live running application on Azure.