Subscribe For Free Updates!

We'll not spam mate! We promise.

Tuesday, 10 February 2015

How to scale-out DB?



If you want to add more computers to your DB network, the best software you can find is Windows Azure. Windows azure is a software or a platform mainly used in cloud computing and was released by Microsoft in February , 2010 for creating and managing services relating to cloud computing over the world wide web through a network of data centers that are managed by Microsoft. It provides Infrastructure as a service (commonly known as IAAS) and Platform as a service (commonly known as PAAS) so as to support different programming languages, platforms and tools that are managed by Microsoft or any third party.
On windows Azure the power of managing huge work load is similar in meaning to the scale-out of the database. This is because there is a sharing of workload among the various data servers that are connected to the data centers of the network. Frankly speaking scale-out DB is a technique to solve problems that are related to performance and capacity of a network. There is always a need of scale-out when talking about large number of data and databases, which are independent of the fact that one user is using the system of 1000 users using.

 How to scale-out DB on Windows Azure?

 On windows azure scale out DB can be achieved in many ways but the most frequently used is the SQL database federation. This technique is the main concept is breaking the table. This means that there is a horizontal shrading of the database and thus the row splits one or more tables. This leads to partitioning of the federation table databases. But it is not that all problem can be solved by this method. There are simpler cases in which such complex techniques may not work or even are not required. So clearly this is not the solution to every scalability problem. According to characteristics and requirements of the users different approaches may be considered.
Here we provide you some major available solutions according to their complexity.

Below is a list of the available solutions according to their complexity:

1.   Increasing the database size

Evidently each and every data base have its size fixed. For web edition we can find 5 GB databases while for business edition we can find 150 GB database. Clearly if we increase the size of the database we can sort out most of the problem. Thus we can change the of database base as well increase the size of the database to solve this kind of problem.

2.   Using multiple locate users and databases

In some cases which may not be found generally you have the option to create multiple databases and allocate different user and login to each of the database. This was very common in the era before federations came in to existence which changed the scenario. Which were used to share the workload. The best thing about this technique is you can use it for the databases you are using for shorter period of time. When the work is completed you can easily merge all of them into the primary database which was used earlier .Thus it is a very good approach for small databases.

3.   Using federations:

Use of SQL database federation helps in gaining very high performance and sharing of resources in the network.  One or even more databases are broken into large number of smaller databases by the row and then each of them are partitioned to different databases where each of the database is called to be a federation member. This process is commonly known as sharding and can be considered a huge invention in management of large application databases. But this federation has a limitation that in case of scale-out DB, that it can be used only for the business edition databases.

Use other storage forms:

Windows Azure also supports use of BLOB storage which is used to store and retrieve binary large objects or BLO which gives the name BLOB. This is very powerful tool in case of large data. This can also be used to create databases for .NET websites. For more information on how to use BLOBs you may click here. You may also use table storage.

SOCIALIZE IT →
FOLLOW US →
SHARE IT →

0 comments:

Post a Comment