Downloads: 398 | Views: 567 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1

Research Paper | Computer Science & Engineering | India | Volume 7 Issue 11, November 2018

Hadoop Performance Improvement using Metadata and Securing with Oauth Token

Swapnali A. Salunkhe ^[2] | Amol B. Rajmane ^[3]

Abstract: Initially Hadoop was designed without performance and security aspects. It was just used to process big data in parallel fashion. But now days user needs big data with high speed and with security features. But Hadoop has some limitations while executing the job. These limitations are reduces the efficiency of hadoop and increases the job execution time. It is mostly because of the job processing method of current hadoop system, scheduling and resource allocation. The proposed system replaces the current job processing method by using different one. Proposed system matches a new job with previously executed jobs. If a match found then same results are displayed to user and if not then it will execute new job. If a matching rate is high then execution time will automatically decreases. The proposed system also focuses on security constraints of current hadoop system. Current hadoop system does secure data while uploading and downloading it from system. Data in hadoop system is secured with OAuth authorization token token with AES algorithm for encryption and decryption. Proposed system authorizes hadoop user and encrypts data while uploading and downloading data from hadoop system. The encryption time of data is also less so it does not affect the performance of hadoop. So the proposed system decreases the execution time by metadata matching technique and secures with real time encryption algorithm

Keywords: Hadoop, Bigdata, Map Reduce, OAuth token, Real Time Encryption Algorithm

Edition: Volume 7 Issue 11, November 2018,

Pages: 1951 - 1955