Data storage is changing. What was true five years ago about the way you managed databases and storage systems may not hold true anymore.
Keep reading for a primer on data storage today.
Once upon a time, data storage was pretty simple. From the early 1990s through the 2000s, most data lived in simple relational databases, like MySQL and PostgreSQL.
They were hosted on traditional servers using magnetic disks.
Each type of relational database had its special nuances.
But in general, if you knew one relational database, you could learn to work with another one easily enough.
And when it came time to choose which type of database to use for your business or customers, there was not much to think about, because each of the databases worked in more or less the same way.
Data Storage Today
Today, however, the data storage world has grown much more complicated. It is changing radically as a result of the following trends:
- The rise in popularity of "NoSQL"-style databases. NoSQL databases don't define rigid data structures in the way relational databases do. NoSQL can be a better choice for data storage scenarios that involve rapidly changing or inconsistent types of data.
- Software-defined storage. Increasingly, storage is abstracted from the underlying disks by software-defined storage systems. Software-defined storage is more flexible and scalable, but it can make it more complicated to determine where on your cluster of servers a given piece of data actually lives.
- Cloud-based storage. The costs of storing data in the cloud have gone down -- AWS lowered the cost of its popular S3 storage service last fall, for example -- and there's a good chance that trend will continue as the competition between cloud vendors becomes even fiercer. This means more and more companies are moving their data to the public cloud. It also means more companies are facing the compliance and security challenges that come with cloud storage.
- In-memory and solid-state storage. In order to deliver better performance, some storage systems are now run using RAM or solid-state disks. These media offer higher input/output rates than traditional disks. But they can also create new challenges -- especially in the case of in-memory storage platforms, which run a higher risk of permanent data loss if they are not designed to handle crashes and hardware failures sufficiently well.
- Container storage systems. Docker containers present special challenges when it comes to data storage. As a result, they are breeding a new generation of storage systems designed specifically for containers, like Torus.
What do these changes mean for MSPs?
First, they mean that you need to know much more about storage than you did in the past in order to choose the storage solutions that best meet the needs of your business and your customers.
At the same time, however, the increasingly complex world of data storage also presents an opportunity.
By becoming experts in the storage options outlined above, MSPs can provide good service by helping clients to understand data storage questions that many consumers can no longer answer for themselves.
There's opportunity in complexity, especially for MSPs.