5 Big Data Security threats

Introduction

The cumulative growth of information has produced a gigantic pile of data in Terabytes(TB) or Petabyte(PB) that cannot be processed through traditional databases. The massive data is managed and processed through special big data processing technologies like Hadoop, NoSQL, etc.

Theses frequently used to discover the hidden pattern of data to understand the business trends for accurate decision making. The productive use of Big Data has brought a big revolution in many fields like business, healthcare, defense, research, etc. but in the meanwhile it is also facing some security threats.

These threats may create a vicious impact on processing results.

data security threats

Security threats include theft of stored information, DDoS attack, ransomware attack, or crash of servers. The security issue can be even worse when organizations store confidential or sensitive information. The information may include credit card numbers, login credentials, personal contacts, customer’s/clients information, this may include all such sensitive information.

In case of information loss, the company may face serious consequences like financial loss, litigation cost, penalties, sanctions etc.(Sisense 2020).

Overall, threats categorize into two categories;

  1. first is an intentional threat, in which data is infected by a human deliberately;
  2. second is an unintentional threat in which data is affected due to some natural reason like, hardware failure, software failure, natural disaster, etc.

It also includes sharing of data or leakage of data unintentionally, misconfiguration, loss of devices, etc. are the most happening threats that Big data normally face.

The following are major 5 security threats to Big data.

1.     Information leakage/sharing due to human error

Un-intentional or accidental threats are not deliberately posed by company workers. Normally they occur due to skill-based slips, misconfiguration or clerical errors e.g. pressing a wrong key, misapplication of valid rules, poor patch management, use of default usernames and passwords, or easy-to-guess passwords, etc.

Sometime these small mistakes may cause a big problem or we can say it is the vulnerability of the system that leads to wrong analytics based on wrong data(ENISA 2AD).

2.     Leaks of data via Web applications (unsecured APIs)

Big-data-api-leaks

Various experts claim that Big data is often built without proper security since Big data applications are developed according to the web services models. These models make use of APIs (application interfaces) that might be vulnerable to common attacks like Open Web Application Security Project (OWASP)(ENISA 2AD).

It is observed that normally data breaches occur due to insecure APIs in many companies particularly in social networking sites, photos, and video sharing services like Yahoo, Facebook, Snapchat, etc.

3.     Inadequate design and planning or incorrect adaptation

Big data processing is a very critical process in which a long-range of factors are considered by experts. In case of any mistake while deciding the variables and weightage of values may cause a serious issue.

Management of data redundancy is seen as a threat mitigation technique in which data is dispersed on multiple locations that may open to attacks, disasters, and outages issues of big data. HDFS plays an important role while designing a large scale storage system to manage data in Peta Bytes.

4.     Interception of information

Another critical threat in which an offender may intercept communication between nodes is by interfering with the communication link. According to some expert’s inter-node communication using new data, tools are often unsafe that can be hijacked by someone.

Such vulnerabilities are considered serious threats and flaws of the system.

5.     Failure of project

A specially trained team is required to implement the solution to the problem. They must know the statistical, mathematical, correlations, regressions, etc. in their mind to achieve the objectives. Failure of the project occurs due to insufficient organizational alignment and understanding lack of middle management adoption etc(Max Henrion 2019).

The failure of big data processing project may create a vicious impact on business or area for which big data was arranged. It normally happened due to a lack of experts with special big data processing skills.

The project leader must know the whole infrastructure and business domain to implement to the solution.  Sometime, the whole system might be collapsed due to the sudden failure of processing components.

 

References

ENISA. 2AD. Big Data Threat Landscape and Good Practice Guide. Vol. 104.

Max Henrion. 2019. “Why Most Big Data Analytics Projects Fail | ORMS Today.” Retrieved May 25, 2020 ().

Sisense. 2020. “What Is Big Data Security | Sisense.” Retrieved May 23, 2020