After the NoSQL movement, the data storage has now given the many applications for the data storage which can be equally be used by the programmers and architects for better data storage solutions.
|Scooped by Custom Tel|
ARE YOU LOOKING FOR THE RIGHT DATA SOLUTION FOR YOUR APPLICATION?
A big credit should be given to NoSQL movement, But still the data storage solution problem is not solved yet. Many of the agencies are now working hard to build an innovative and new data storage solution, and many of them are willing to use them. On the other hand, either you are a programmer or a solution architect who needs a data solution for application, there will be a daunting task which you need to handle while weighing and understanding the tradeoffs required for the applications and after that you need to make a decision. This article will help you in exploring the data needs of the end user and the other tradeoffs. This article will help you in providing the guideline for selecting criteria of data storage choices which will allow the developer and the architects to make an informed decision.
A few years back, when the most systems were small and relational databases could have handled those systems without any trouble. The storage choices for the developers and the architects were used to be simple. But now the scale of these systems has grown significantly over the last few years. On the other hand, the high tech companies like Google and Amazon have faced the scale challenge before others. They have observed that the relational databases are unable to handle those use cases in a wide scale.
This phenomenon can easily be explained by the CAP theorem given by the Eric Brewer which stated as “ a distributed system can only have two of the three properties which are Consistency, Availability and Partition Tolerance. After that CAP theorem , both Google and Amazon came up with the two data solution,which are named as Big Table and Dynamo respectively. They have supported the NoSQL movement which in turn lasted in many data solutions.
If the pre-NoSQL is an era which has considered to be boring , then the NoSQL or you can say NewSQL era is totally opposite. The architects have now used a wide variety of many storage choices which are Local memory, distributed cache, relational files, name value pairs, document storage, queue, graph DBs and server registries.
But on the programmer’s side, they are now facing the dilemma when it comes to choosing from the different choices, Different applications have their own specific needs from a data store. For an example, a transaction will be needed for an online retail application whereas e search application like Google will be needing the high scalability and consistent data as well.
Many of the hybrid solutions are unable to support transactions across the many multiple data stores but the transactions can also be made by using an external transaction manager.