Login | Register   
LinkedIn
Google+
Twitter
RSS Feed
Download our iPhone app
TODAY'S HEADLINES  |   ARTICLE ARCHIVE  |   FORUMS  |   TIP BANK
Browse DevX
Sign up for e-mail newsletters from DevX


advertisement
 

Best Practices to Protect Big Data

Learn more about the different steps and tools employed to protect Big Data applications.


advertisement

The security aspects of Big Data applications are often ignored or treated as a secondary requirement. But data security has a tremendous impact on the application as it is handling data. Learn more about the different steps and tools used to protect Big Data applications.

As Big Data is spreading across different domains, the security aspect is getting more and more attention. Formerly, we had end-point centric security systems, but that is not sufficient to protect your application from intrusions. Big Data brings with it a different set of security concerns that are very different from normal applications.

In today's world, security is very difficult to explore and navigate. It is also very expensive to implement a proper end-to-end security system throughout the software system. And there is always a possibility of a breach to the security no matter what policy or system you follow. Organizations taking part Big Data initiatives should plan accordingly, based on their budget and policies, to adopt modern and up-to-date security practices.

Security Risks in Big Data Environment

In this age of Big Data there has been significant growth in data volume, data velocity and data variety, as well as growth in cloud models, mobile apps and other interconnected applications. The data flows from one point to another through different systems, applications and environments. This data explosion offers meaningful insight to the business, but it also exposes the business data to various systems, processes, and people. As this huge volume of data is stored, processed, analyzed and shared in different collaborating systems, there is always a chance of a security breach.



Big Data is collected from different sources and various types of business intelligence tools are used to analyze it and get meaningful insight. This information is accessed and used by the decision makers. Sometime the data is also used for collaboration. The tools used for collaboration and processing are also having security limitations. So, there is always a probability of exposing sensitive data/content. Once the value elements of Big Data are identified, it can be accessed, updated or even changed by the users. This can cause serious security issues and threats to the organization.

Advanced security measures can ensure information security in a collaborative environment. Big Data organizations need to be more precise on controlling and balancing business requirements and data protection. The following are some recommended steps to protect your data:

  • Break Big Data into small data: In this way the system will be better able to handle the volume, velocity and variety of the data. As a result, organizations will also be able to make faster and more accurate business decisions.
  • Identify the context of the information: Organizations need to identify the employees, partners, vendors, or any other third party, involved in this collaboration and also the communication channel. This gives a detailed idea about the collaboration environment and its stake holders.
  • Deploy data controls: Data controls are very important to deploy at strategic locations. This will secure data protection and collaboration.
  • Deploy control for cloud and mobile environment: Cloud and mobile collaboration is an essential part of any application and its deployment and constitutes one of the highest risk areas. Organizations need to understand and identify how the data is shared in cloud and mobile environments.

Big Data Security Tools

In previous years, most organizations had a single software vendor and a single database (SAP, Oracle, PeopleSoft, etc.) for the entire organization. As a result, security issues were more visible and easily manageable. But in the current scenario, where we have Big Data, cloud, mobile devices, and more, the number of security holes in the system is an unknown and the possibility of a security breach is much higher.

In recent developments in information security, there are a number of software packages and vendors available to properly enforce security practices. Perimeter security strategy for Big Data is same as other systems, so in this section we will only discuss 'inside-the-network' tools.

  • Monitoring and logging: To monitor and log everything is the best strategy to detect unauthorized activities. Some logging systems like syslog (Linux), event log (Windows) can be effectively used. SNMP is also very useful to log network events. There are also different software packages available to aggregate logs and store them in a central location for analysis. These are known as Security Information and Event Management software (SIEM) packages.
  • Analysis and auditing: The main functionality of SIEM packages is to automatically detect unauthorized activities and generate warnings. But all SIEM software requires configuration to work properly. It is recommended to use pre-configured SIEM packages that are updated frequently and capable of identifying a major number of security breaches through log analysis. The latest SIEM packages are LogRhythm, Q1 Labs (IBM), McAfee, Splunk, etc.
  • Managing Identity: Identity and Access Management (IAM) is very important to protect Big Data. Because the data is accessed by employees/contractors using different channels such as mobile devices, SAAS model, and other services. It is very important to consider 'ídentity' as the new perimeter to identify who is accessing the sensitive data instead of concentrating on the physical location of the data. It is absolutely necessary to consider a collection of tools that will help us to deal with perimeter failure.
  • Masking the data: Data masking is another way of protecting data security. The data can be masked using encryption or tokenization. Some vendors also demand that their data masking tools do not follow encryption or tokenization but perform the entire masking dynamically.
  • Application security: The final step is to ensure security within the Big Data applications that are accessing sensitive information. This is very critical as most of the popular tools are not built keeping security factors in mind. Recently, most of the Big Data tools have improved significantly on the security side. The two most important factors are 'permissions at granular level' and 'data encryption'. The latest version of Hadoop is expected to support new security features and probably address some of these emerging issues.

Summary

In today's world, Big Data security is a huge concern. As we know Big Data systems are not like normal single vendor systems, so the security issues are much more complicated to handle. There is no single solution/tool/vendor that can protect your data, but you may need to use different security tools that are effective depending on what area is being secured. So the ultimate solution is to keep on using multiple effective tools over time. Eventually, you should have a good and comprehensive security system in place.

 

 

About the Author

Kaushik Pal is a technical architect with 15 years of experience in enterprise application and product development. He has expertise in web technologies, architecture/design, java/j2ee, Open source and big data technologies. You can find more of his work at www.techalpine.com and you can email him here.

 



   
Comment and Contribute

 

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

 

Sitemap
Thanks for your registration, follow us on our social networks to keep up-to-date