In the past, there was no concept of file, DBMS, RDBMS and SQL. The content was present in the magnetic tapes, with random access. In random access, seek and transfer activities are done. The transfer rate has been improved to 100 MB than the improvement of the ‘seek’.

So, seek time has been a limitationin the performance. On top of it, the data growth has lead storing and processing data to be complex. Here, storing data is accomplished through DFS, by means of HDFS and processing data by HadoopMapReduce.

Strengths of Hadoop became the huge data storage in peta byte scale and processing huge files, however has a limitation withbatch processing. However the requirement demanded is to get the results in the form of RDBMS. It stemmed the new software like Cadandra, Mongo DB and HBase and similar ones to be developed.

So, according to this HBase is the NoSQL that gives the output in the form of RDBMS.

NoSQL System

HBase is open source software and so there would be a large scope of distribution and development. It is horizontally scalable and its sorted map data are generally builtupon the top of Apache Hadoop. The data model used by it is quite similar to Google’s Big Table. The goal and fruit of this HBase is to provide very fast random read access or write access to the huge amount of structured data. It is found to be layered over HDFS. The best thing that makes many users reliable on HBase is that it is tolerant of failures of machines.HBase follows strong consistency model.

HBase is built on top of the HDFS and the best thing about it is that it provides very quick record lookups as well as updates for the larger tables too. The data is internally stored in the ‘StoreFiles, which are indexed and existed on HDFS for better and higher speed lookups. HBase is an incremental database that enables incremental data.

HBase Architecture:

HBase is on top of the HDFS, so, HBase uses HDFS as the storage layer, which is reliable.

When it comes to the organization of the data in terms of tables, the data is not stored like in columns and rows just like in the spreadsheet model. Instead HBase uses tag metaphor. The availability of the information is now according to the specific tag. 

To known more about Hbase Architecture and Hbase tutorials browse Intellipaat.in.

Use cases of HBase

-          The importance of HBase with respect to the internet is that it is used to store the web pages as it is crawling over the web.

-          The row key it considers in this case is the reversed URL for the specific web page. Like for example, www.hbase.org is taken as org.hbase.www.

-          The column family does include the storage of the actual HTML code, the contents family, the outgoing link and inbound storage in the form of anchor and to store the metadata similar to the languages.

-          The oldest copies of HTML can be stored with the help of multiple versions related to the content family. It helps the analysis after or during the page changes.  

 To known more about how to use Hbase and Hbase tutorials browse Intellipaat.uk.