Deep Learning Approaches to Cloud Security. Группа авторов

privacy also increases. We use Deep Learning concepts to increase privacy levels. In this chapter, we understand the cloud computing concept within a Multi-Tenant Framework (MWF). In Multi-Tenant Frameworks, requirements for privacy and security concepts are developed using Deep Learning. The goal is to find privacy requirements across many factors in a Multi-Tenancy based systems using Deep Learning concepts. The services of Multi-Tenant based systems are aggregated due to the dynamic environment of Cloud Computing. Three consistencies will be maintained by privacy policies using Deep Learning. In Multi- Tenancy, a large number of users (tenants) use the same services required for privacy and security to maintain the durability and consistency of service.

      Keywords: Multi-tenant, privacy, framework, privacy policy, cloud computing, single tenant, public, private

2.1 Introduction

It is very important to understand the need for Multi-Tenancy in the Cloud Computing environment because today all work is done with the help of an internet transaction mode. In this environment, the database and schema used in the database change or modify frequently. Data is stored in different areas in different databases [1]. The cloud computing environment of every organisation is different. The real need of cloud environment-like private or public modes depends on the concept of data and tenant needs, organisation needs, or the configuration of a database depending on the structure of the database and the model used by the organisation for privacy policies. The security level of the Multi-Tenancy structure should maintain the durability of an application to maintain consistency of the system and avoid interruption of the regular based work done by the tenant. The concept of Multi-Tenancy in multiple databases requires more privacy between tenants both logically and physically to accommodate a greater the need for privacy and security in each work area. Using Deep Learning concepts can improve the complexity in accessing and managing a database. Using Deep Learning concepts should reduce access and increase efficiency, preventing the leak of data, hacking, or other risks because they can be found out easily and improve the quality of Multi-Tenant system efficiency. So, in Deep Learning, the privacy concept is improved and the transparency between the multiple tenants increases, maintaining privacy. Deep Learning concepts are very powerful and scalable for implementing. In Deep Learning, databases are managed efficiently. The Deep Learning algorithm makes this affordable. The Deep Learning concept has the ability to improve data driven predictions. It finds the patterns of privacy and security in databases in Multi-Tenancy. This can provide good or better values for organisation. In Deep Learning, the work is done for binary, category, and value predictions in Cloud Computing. In this chapter, we discuss the basic concept of Multi-Tenancy, privacy requirements, and the Cloud Computing concept with Deep Learning.

2.2 Basic Structure

It is very important to understand Cloud Computing, as it provides service accordingly to its user end. Whether it is a private or public cloud depends the requirement of user, like Multi-Tenancy or Single Tenancy. It is also an on demand service that depends on tenant requirement, resource availability, storage requirement, activity management, and which topology is required for the distributed system. The tenant requires either a centralised or decentralised framework. Security in Cloud Computing also depends on the concept of the database and the need for bigness and organisation. As the need arises, the organisation selects the structure of Cloud Computing. Then, they create the schema and select a data model and select a Single or Multi-Tenancy concept for work. Complexity and cost are also dependent on the requirements of the organisation’s needs. There are structures of Cloud Computing and the Multi-Tenancy concept available illustrating the impact of Multi-Tenancy in Cloud Computing. The cloud manages the concept of Multi-Tenancy, shares resources, and manages services with many tenants. Management and the utility services are provided by the structure according to need. This section discusses the basic structure of Cloud Computing and how the work is done in combination with Multi-Tenancy [2].

      2.2.1 Basic Structure of Cloud Computing

      Cloud Computing services are used by every service area, whether the business is small or large.

      The cloud facility is available according to the requirements of the business. The cloud enables the facility to store data, whether it be large or small, and provides access to data from any location and any hardware or a virtual environments for accessing or storing data.

      Cloud Computing provides the services of both service oriented and event driven architecture.

The given diagram represents the various services of the cloud. All services are inter-related to each other. The three main parties involved are the cloud service consumer, cloud service provider, and cloud broker shown in Figure 2.1.

Figure 2.1 Cloud computing services [3].

In cloud service, all consumer services are provided if the consumer belongs internally or, if the consumer belongs externally, it depends on the consumer requirement. At the time of consumer requirement, the service provides, as the diagram represents, the work of a cloud auditor.

      The cloud service provider provides all cloud services. Whether the type of service provided by the service provider is commercial or corporate is decided by the service provider [3].

      The cloud service provider can provide all services with physical or virtual resources that provide the cloud service and create applications according the requirement of services. All groups work together as a service orchestration.

      2.2.2 Concept of Multi-Tenancy

      The basic concept of Multi-Tenancy is required when many tenants work together with the use of internet and data can be transferred using digital services if the tenants are in different places; the concept is to work from one place to another place. This service totally depends on the needs of the organisation. The requirement of tenants and which type or place the tenant requires is up to the organisation’s needs [4].

Basically, there are two types of tenancy in an SaaS environment: Single and Multi-Tenancy. Figure 2.2 briefly discusses Multi-Tenancy’s use in real time application. In Single Tenancy, a single environment or platform is used, but in Multi-Tenancy, a common platform is used in different places or shares the platform of work and creates a known work place for a single organisation.

Figure 2.2 Types of tenant [4].

A Single Tenant service uses a single system (software) for a single tenant for one service. A single system service provides for a single building, but if its requirement is in a different building, the tenant needs to purchase again for the new building for the same services. In Multi-Tenancy, the client is not required to purchase a single service multiple times, which means single services can be used in different offices of a single organisation.

      In a Multi-Tenancy system, it is easy to modify and make changes as required by the tenant. Multi-Tenant systems combine the work place between the client and the tenant, but the concept or logic should be separate to each other. The organisation shares the data and configurations, as well as services of management with the user and tenant [5].

      There are basically three major concepts in Multi-Tenancy, as show in the Figure 2.2.

      1 1. A shared database which is used to manage the Multi-Tenancy system