<img height="1" width="1" style="display:none;" alt="" src="https://dc.ads.linkedin.com/collect/?pid=24166&amp;fmt=gif">

AtScale Blog

Above the Cloud: Big Data and BI

Posted by on
Find me on:

MIC-ChartData_04-01

It’s becoming more and more crucial for businesses to find ways to gather, use, and analyze big data. After all, big data is by far the most effective tool for keeping accurate and thorough track of customers and employees, as well as the most efficient way to predict future behaviors. However, big data is notoriously complex, unstructured, and massive, making it extremely difficult to manage without the proper tools.

This is where Hadoop comes in. Hadoop simplifies the entire process, making it easy for businesses to obtain and collect big data, convert it into forms that are compatible with those companies’ software programs, and then use the data in whatever way they like.

Relational Database Management Systems

It is very common for a company to use a traditional relational database management systems, or a RDBMS, but there are many benefits to utilizing Hadoop and MapReduce instead. For starters, when it comes to updates to the system, an RDBMS must be read and written many times. With Hadoop and MapReduce, however, you will be able to read many times after writing only once. Additionally, an RDBMS has a static, inflexible schema, while Hadoop employs a dynamic, intuitive schema that can adapt to any given data structure, making it very simple to work with any type of data.

RDBMS are also not the best option for handling big data because as a company continues to process larger and larger amounts and types of data, that company will still want and need a highly interactive response. If that company uses an RDBMS, they will require more expensive hardware that can handle such high quantities of data. Hadoop and MapReduce allow for very large, robust data processing without requiring an extreme hardware upgrade or expansion; instead. more nodes are simply added. The more nodes that get added, the greater the amount of data that can be calculated--and at a higher speed.

Company Examples 

One example of a company that has significantly benefited from Hadoop is Facebook. Facebook uses Hadoop in many innovative ways: for example, they recently needed to come up with a way to transport a 30 petabyte cluster of data. One petabyte is the equivalent of one million gigabytes, so this was a very daunting task. As GIGAOM reports, “The move was necessary because Facebook had run out of both power and space to expand the cluster--very likely the largest in the world--and had to find it a new home.” Facebook decided replication was the best solution to their problem, so their data team began the attempt to copy over all of the data. GIGAOM explains that “the challenges were in developing a system that could handle the size of the warehouse...the rate of object creation meant that [their] previous system couldn’t keep up.” Fortunately, Hadoop’s abilities to handle extremely large data clusters allowed Facebook to efficiently replicate the data and store it all.

Hadoop is a perfect solution for a company, such as Facebook, that needs to accommodate such huge clusters of data. BI with Hadoop also provides effective disaster recovery, giving companies the knowledge and reassurance that even if the worst occurs, the data of your company and your clients will be recoverable and safe.

New Call-to-action

Source:

https://gigaom.com/2011/07/27/facebook-hadoop-cluster/

 

Topics: Hadoop

Learn about BI & Hadoop

The AtScale Blog is the one-stop shop for cutting edge news and insights about BI on Hadoop and all things AtScale.

Subscribe to Email Updates

Recent Posts