Once upon a time I had a very old VW Jetta with a manual transmission. One day, the reverse gear stopped working. A friend of mine owned a VW Scirocco, which at the same time blew its engine gasket. He offered me the transmission out of his Scirocco, it being a VW. The mechanic said that he would need a special mounting bracket to make it work. How is this story akin to data mobility?
Just because it looks the same, doesn’t mean it is the same in the required ways. An application may work with both NAS and Object S3, but there might need to be some adjustments made when moving data between these protocols, without which the application won’t recognize the data.
Depending on the application and the storage system, to achieve true data mobility, there will be a need for an application connector. An application connector is code that transforms data so the application can recognize and consume it. Think of the application connector as a transmission mounting bracket.
What is the difference between data copy and true data mobility?
Anyone can do data copy; there are many tools available, both open source and proprietary, that can copy data from one system to another. These don’t account for any differences in how the application interacts with the data on a storage system. Here are some examples:
When writing data, there are two main identifiers: what is the name of the file and where in the file system and then on the drive does it reside. The application tracks the name and the location in the file system. When copying data to S3, data may not be identified by a name or even a data path. The object store may generate a unique ID that is associated with a particular file. The application would know this ID and therefore, where to find the file. Once it accesses the file, there must be metadata that will enable the application to recognize the data. Just copying data from a file system to object without addressing the required changes will result in the application not able to recognize or access the file.
Though all object storage vendors support S3, which is in turn a standard, the inner workings of each object store vary. It is necessary to understand and accommodate the differences to ensure application compliance once data is copied. This is also true when copying data across different vendors’ object stores or even across different versions of the same vendor’s object store. Think how often you received a notice when opening an older Word document that many of the formats may be changed, forcing you to reformat the whole document.
Each application works differently with various storage protocols. Applying one methodology that has worked for an application doesn’t guarantee it will work with another application. Copying data from storage system to storage system without understanding the application may result in application non-compliance. This is why many choose to migrate data via the application. This poses its own challenges, including longer time, many more infrastructure resources, and unexpected costs.
True data mobility ensures that all application requirements are addressed. This implies that at the end of the copy process, the application can discover and consume data on the new system.
Interlock Application Connector
Interlock Technology has developed application connectors for several applications* and a framework for developing new ones as need arises. These are akin to a codec that enables a file to be viewed on a different platform than it was created. Imagine taking a video with an Apple device and application. The recording then must be viewed on a Windows machine. When the file is being opened, you may need to select the application with which to open the file. Then the message will pop up saying it is not
compatible or you will see a file full or gibberish. To open and consume this file, you need to have a codec, a sort of translator.
When Interlock performs a data copy, the application connectors ensure that data is transformed in a way necessary for the application to recognize and access it on the new storage system, whether it is another NAS or S3. Key attributes of application connectors:
Copy data from source to target (any to any) while remaining application compliant.
Accommodates applications, source storage, and target storage (unstructured data protocols).
Transform all relevant metadata required for applications to recognize data.
Without application connectors, once data is moved to the target system, additional steps will be required to achieve application compliance. These actions, when performed manually, may expose data to human error, compliance breaches, and unnecessary time and labor. When moving data at scale, make sure application compliance is maintained.
Data Mobility demands that data is recognized and accessed by applications regardless of the storage system, service, or protocol it is moved to. It also requires timeliness, which can only be achieved when the intelligence required is built into the system, eliminating the need for manual intervention. Interlock Technology is paving the math to data mobility with its application connector technology, which will be available as part of the DATAFORGE mobility software solution.