App Data Management over the Cloud – DOs and DON’Ts

Date:
2013-07-19 17:41:23
   Author:
10Gtek
  
Tag:

 App data management is not a simple task if your App is having a high volume of users’ activity. There are different ways of managing App data based on the types of data your App is generating in the cloud. Data format is the key point in making the strategy for App data management over the cloud. Here are different data formats that your App can be

generated:                           

1. File data                     
2. RDBMS data
3. NOSQL  data
4. In Memory Data

File based Data

 

When to use it

1. You need to store the individual user data in each separate file like audio, text, images etc.
2. User can directly hit the file and access it from his device/ machine.
3. Searching of the file content is not important from the App prospective.


When not to use it

1. Real time search functionality based on file content
2. Need BI and Analytic on the file data

 

Implementation Options Available

1. App42 File Upload, Amazon S3, CDN, EBS, SAN etc


RDBMS data


When to use it

1. Need a structural data storage based on entities and their relationship in your App defined as schema.
2. ACID property is an important aspect for App Data.

 

When not to use it

1. RDBMS can fit into most of your requirements, however if your App data is not dependent on schema you can opt for other options.
2. Since horizontal sharding is not directly supported in most of RDBMS, data scaling can become tedious for very large volume databases.

 

Implementation Options Available

1. Amazon RDS or other hosted RDBMS provider like Rackspace etc.

 

NoSQL Data

NoSQL can be used in place of RDBMS databases. If your App data is not schema dependent and not fitting in RDBMS requirement, you can go for it. Also, scaling with high volume of App data is much easier than RDBMS.

 

Implementation Options Available

1. App42 Storage Service, Amazon DynamoDB etc

In Memory Data

 

When to use it

1. Your user App data is short-lived.
2. Data durability and recovery is not important aspect.
3. Frequent insertion and deletion of data

 

When not to use it

1. Data backup and recovery is needed

 

Implementation Options Available

1. Amazon Elastic Cache