Attack Detection in Cloud Networks Based on Artificial Intelligence Approaches

Attack Detection in Cloud Networks Based on Artificial Intelligence Approaches

Zuleyha Yiner (Siirt University, Turkey), Nurefsan Sertbas (Istanbul University – Cerrahpaşa, Turkey), Safak Durukan-Odabasi (Istanbul University – Cerrahpaşa, Turkey) and Derya Yiltas-Kaplan (Istanbul University – Cerrahpaşa, Turkey)
DOI: 10.4018/978-1-5225-7353-1.ch003

Abstract

Cloud computing that aims to provide convenient, on-demand, network access to shared software and hardware resources has security as the greatest challenge. Data security is the main security concern followed by intrusion detection and prevention in cloud infrastructure. In this chapter, general information about cloud computing and its security issues are discussed. In order to prevent or avoid many attacks, a number of machine learning algorithms approaches are proposed. However, these approaches do not provide efficient results for identifying unknown types of attacks. Deep learning enables to learning features that are more complex, and thanks to the collection of big data as a training data, deep learning achieves more successful results. Many deep learning algorithms are proposed for attack detection. Deep networks architecture is divided into two categories, and descriptions for each architecture and its related attack detection studies are discussed in the following section of chapter.
Chapter Preview
Top

Cloud Architecture And Cloud Security Issues

There are three different layers in a cloud structure. These are Deployment Models, Service Models, and Essential Characteristics respectively from the bottom to the top. The classes in the Deployment Models are public, private, hybrid, and community. For any class of the deployment models, there are delivery models called Service Models, which involve Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). These models are the core of the cloud network and get several characteristics from the top level of the structure. The characteristics can be exemplified as measured service, on-demand self-service, and rapid elasticity (Said et al., 2016).

Figure 1.

Delivery models of cloud

978-1-5225-7353-1.ch003.f01

The large amounts of data and customers in a cloud environment cause the performance degradation and inaccessibility to the network. To solve any problem and also any security issue, the cloud computing requires specific methods. Because the system and its properties like sharing of the resources are different from the other networking types.

A brief description for security challenges in cloud and candidate solutions for each challenge are given in Table 1.

Key Terms in this Chapter

Trust Level: Access degree of end users to the cloud or any other computer infrastructure.

Data Center: Large areas or buildings that accommodate computer systems, data warehouse, and servers of the enterprises.

Complete Chapter List

Search this Book:
Reset