Method for security reinforcement based on Hadhoop module

Abstract

The invention relates to a method for the security reinforcement based on a Hadhoop module. The method comprises the steps: taking a Kerberos server as a user external memory pool, and taking Kerberosas a verification protocol for the password verification of a user; permitting a user who successfully passes through the Kerberos verification to have a bill, and grading the user, using the servicein Hadhoop when the bill is verified to be effective, and dividing the uploaded data in the Kerberos into a plurality of data blocks according to the sensitivity of the data; executing the partitioned access authority for the users at different levels at step 2, performing the data encryption and hiding of the data blocks so as to guarantee the safety access of the authorized user to the sensitive data, and performing the data incoming and outgoing control for an unauthorized network; and isolating a big data system from other enterprise information systems through the network topology. According to the invention, the Hadhoop introduces the support for the Kerberos, and the method improves the safety of user access, guarantees the restriction on the access to various types of data throughthe encrypted dividing of the levels of data and users, and improves the safety performances.

Description

technical field [0001] The invention belongs to the field of big data, and in particular relates to a security reinforcement method based on hadoop components. Background technique [0002] Hadoop consists of multiple components, so the security of Hadoop cluster means ensuring the security of each component, which makes Hadoop cluster security a complex task. [0003] Hadoop's initial design is to run in a trusted environment, which assumes that all cluster users are trusted, they can correctly identify themselves and will not try to obtain more permissions. This enables a simple security model, which is the default authentication system in Hadoop. In simple security mode, Hadoop trusts the user identity provided by the operating system. Unlike most relational databases, Hadoop does not have any centralized user and privilege storage mechanism. There is no concept of authenticating a user via username and password in Hadoop. Hadoop accepts and trusts the user name provi...

AI Technical Summary

Problems solved by technology

For most companies, this simple user authentication method does not meet their requirements