In which cloud infrastructure and computing resources are made available to the general public over a public network?

Introduction

NIST defines four cloud deployment models: public clouds, private clouds, community clouds, and hybrid clouds. A cloud deployment model is defined according to where the infrastructure for the deployment resides and who has control over that infrastructure. Deciding which deployment model you will go with is one of the most important cloud deployment decisions you will make.

Each cloud deployment model satisfies different organizational needs, so it’s important that you choose a model that will satisfy the needs of your organization. Perhaps even more important is the fact that each cloud deployment model has a different value proposition and different costs associated with it. Therefore, in many cases, your choice of a cloud deployment model may simply come down to money. In any case, to be able to make an informed decision, you need to be aware of the characteristics of each environment.

Cloud Computing Deployment Models

Cloud deployment models indicate how the cloud services are made available to users. The four deployment models associated with cloud computing are as follows:

Public cloud As the name suggests, this type of cloud deployment model supports all users who want to make use of a computing resource, such as hardware (OS, CPU, memory, storage) or software (application server, database) on a subscription basis. Most common uses of public clouds are for application development and testing, non-mission-critical tasks such as file-sharing, and e-mail service.

Private cloud True to its name, a private cloud is typically infrastructure used by a single organization. Such infrastructure may be managed by the organization itself to support various user groups, or it could be managed by a service provider that takes care of it either on-site or off-site. Private clouds are more expensive than public clouds due to the capital expenditure involved in acquiring and maintaining them. However, private clouds are better able to address the security and privacy concerns of organizations today.

Hybrid cloud In a hybrid cloud, an organization makes use of interconnected private and public cloud infrastructure. Many organizations make use of this model when they need to scale up their IT infrastructure rapidly, such as when leveraging public clouds to supplement the capacity available within a private cloud. For example, if an online retailer needs more computing resources to run its Web applications during the holiday season it may attain those resources via public clouds.

Community cloud This deployment model supports multiple organizations sharing computing resources that are part of a community; examples include universities cooperating in certain areas of research, or police departments within a county or state sharing computing resources. Access to a community cloud environment is typically restricted to the members of the community.

With public clouds, the cost is typically low for the end user and there is no capital expenditure involved. Use of private clouds involves capital expenditure, but the expenditure is still lower than the cost of owning and operating the infrastructure due to private clouds' greater level of consolidation and resource pooling. Private clouds also offer more security and compliance support than public clouds. As such, some organizations may choose to use private clouds for their more mission-critical, secure applications and public clouds for basic tasks such as application development and testing environments, and e-mail services.

TIP

Using hypervisor-based virtualization software to provide isolation between different customer environments can lead to increased utilization of system resources such as CPU and memory. Using native virtualization technologies offered by hardware vendors, such as Solaris Zones when using the Oracle Solaris operating system, can be much more effective and efficient depending on the customer environment. Native virtualization technologies offered by hardware vendors are more restrictive in terms of what is supported than hypervisor-based virtualization software.

Figure 1.1 summarizes the computing architecture evolution.

As Figure 1.1 shows, cloud computing primarily impacts how IT infrastructure and platforms are set up, deployed, and provisioned from an end-user perspective. The applications running in a cloud environment should be able to seamlessly interact with the cloud ecosystem, including other applications within or outside the cloud environment.

Table 1.1 highlights the pros and cons of different computing architectures.

Table 1.1. Pros and Cons of Different Computing Architectures

To take advantage of cloud computing, legacy applications such as those developed using mainframe client/server technologies need to be adapted or migrated to modern languages and APIs so that they can interact with other applications regardless of where they are deployed. Cloud-enabling an application requires that the application be able to interact with databases, middleware, and other applications using standards-based mechanisms such as Web services. Most legacy and client/server applications today do not have this capability natively. Typically, these legacy applications require adapters and wrapper software to make them accessible via Web services.

Cloud Deployment Models

Mell and Grance next define the four Cloud Deployment models8:

Private cloud. The cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on premise or off premise.

Community cloud. The cloud infrastructure is shared by several organizations and supports a specific community that has shared concerns (e.g., mission, security requirements, policy, and compliance considerations). It may be managed by the organizations or a third party and may exist on premise or off premise.

Public cloud. The cloud infrastructure is made available to the general public or a large industry group and is owned by an organization selling cloud services.

Hybrid cloud. The cloud infrastructure is a composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load-balancing between clouds).

These four deployment models can see significant variation depending on other factors that we will discuss in the next section, but they serve to address the broad questions as to how one can deploy pooled cloud resources. Before we move on, it is important to make two points about the NIST Cloud Model:

A customer or tenant can have greater security control over more resources as one moves from SaaS to PaaS and again from PaaS to the IaaS service model.

A customer or tenant can achieve greater security control over more resources when moving from a Public cloud to a community cloud and again from a community cloud to a Private cloud.

Figure 2.3 is an adaption of the NIST Cloud Computing Model, which has been annotated to reflect the discussion in this section on customer and tenant control. We will examine the issue of control in greater detail in the next section.

Change Control

Changes to an operational environment are inevitable as a system undergoes routine maintenance. However, some changes may cause significant impacts to the security posture of the cloud service.63 Therefore, the CSP is required to report “changes in the CSP’s point of contact with FedRAMP, changes in the CSP’s risk posture, changes to any applications residing on the cloud system, and/or changes to the cloud system infrastructure” [6], and submit any residual artifacts associated with significant changes such as the SSP, security impacts analysis, and a re-assessment by a 3PAO to the FedRAMP PMO.

Summary

We covered a lot of material in this chapter. The main points you need to take away are:

Cloud service delivery models: The three basic types are SaaS, IaaS, and PaaS

Cloud deployment models: The four basic types are Public, Private, Community, and Hybrid

Cloud logging: It’s a young field which is growing every day. Be sure you understand what it is you and your organization require before you embark on logging all your log data to the cloud. Be sure to ask lots of questions of whatever perspective provider you are interested in using.

SIEM in the cloud: MSSPs have supported this model for years. Traditional Shrink-wrapped SIEM vendors are starting to realize the value in this model and are scrambling to build cloud capabilities and offerings into their products.

Security concerns: Be sure you understand how the provider approaches IT security. Also, make sure you understand your responsibility when it comes to legislative and regulatory compliance when operating in a cloud environment.

8 Migration phases

For the migration to a CSP under consideration that a remigration to another IT-Service provisioning must be possible, a Five-Phase-Model of secure cloud migration is introduced.

The migration process to a CSP is divided into five phases that correspond to each other. It is important that an organization, intending to move its IT-Services, and IT-Processes to a CSP is aware of all phases and develops and tests a strategy to get in each step of the phases during the whole process of migration and operation. It is also essential for an organization to think about cloud migration to develop a scenario to leave the CSP without any losses of IT-Services and data and with a predictable amount of investment and manpower. Applying the Five-Phase-Model of secure cloud migration will ensure this.

The five phases of cloud migration are:

Planning

Contracts

Migration

Operation

Termination

These five phases will not follow straightward each other. It depends on the success whether the next phase can be entered or not. Under every circumstance, it is necessary to develop the light gray (left in Figure 5) and gray strategy (right in Figure 5) to leave the CSP. The way back to an internal IT-Service is the safest backup, because the customer can define all SLAs and remigration conditions. But all assumed advantages of using a CSP would be lost. The gray strategy offers, on the one hand, the possibility to negotiate regularly new condition with the actual CSP. On the other hand, it is a clear strategy available to move to another CSP if the required service, the service quality, the maintenance, or the economical conditions are not sufficient anymore. In the following section, the single phases are described according to Figure 5. This subchapter gives an overview, which subtasks during each step of the Five-Phase-Model have to be fulfilled. A detailed description of all subtasks during the migration process is not scope of this chapter because the design of the subtasks depends strongly on the required IT-Services, the kind of data which shall be processed and which legal requirements have to be applied.

Summary

There are five key cloud characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service. A solution must exhibit these five characteristics to be considered a true cloud solution. There are four cloud deployment models: public, private, community, and hybrid. Each deployment model is defined according to where the infrastructure for the environment is located. There are three main cloud service models: Software as a Service, Platform as a Service, and Infrastructure as a Service. SaaS was the original cloud service model but the cloud has continued to grow and expand. Now a vast array of service models is available.

There are many factors pushing organizations toward the cloud, as well as many factors that are keeping organizations away. Each organization must evaluate cloud offerings for itself to see what best fits its needs.

Managing the Risks of Public Clouds

Though a public cloud deployment is suitable for most uses that are nonsensitive, migrating sensitive, mission critical, or proprietary data into any cloud environment that is not certified and designed for handling such data introduces high risk. A customer should first select a cloud deployment model and then make sure that sufficient security controls are in place. These actions should be followed by a reasonable risk assessment:

Data and encryption: If the data is stored unencrypted in the cloud, data privacy is at risk. There is the risk for unauthorized access either by a malicious employee on the cloud service provider side or an intruder gaining access to the infrastructure from the outside.

Data retention: When the data is migrated or removed by the cloud provider or customer, there may be data residues that might expose sensitive data to unauthorized parties.

Compliance requirements: Various countries have varying regulations for data privacy. Because some public cloud providers don't provide information about the location of the data, it is crucial to consider the legal and regulatory requirements about where data can be stored.

Multitenancy risks: The shared nature of public cloud environments increases security risks, such as unauthorized viewing of data by other customers using the same hardware platform. A shared environment also presents resource competition problems whenever one of the customers uses most of the resources either due to need or due to being exposed to targeted attacks, such as DDoS.

Control and visibility: Customers have restricted control and visibility over the cloud resources because the cloud provider is responsible for administering the infrastructure. This introduces additional security concerns that originate from the lack of transparency. Customers need to rethink the way they operate as they surrender the control of their IT infrastructure to an external party while utilizing public cloud services.

Security responsibility: In a cloud the vendor and the user share the responsibility of securing the environment. The amount of responsibility shouldered by each party can change depending on the cloud model adopted.

Managing the Risks of Public Clouds

Although a public cloud deployment is suitable for most uses that are nonsensitive, migrating sensitive, mission-critical, or proprietary data into any cloud environment that is not certified and designed for handling such data introduces high risk. A customer should first select a cloud deployment model and then make sure that sufficient security controls are in place. These actions should be followed by a reasonable risk assessment:

Data and encryption: If the data is stored unencrypted in the cloud, data privacy is at risk. There is the risk for unauthorized access either by a malicious employee on the cloud service provider side or an intruder gaining access to the infrastructure from the outside.

Data retention: When the data is migrated or removed by the cloud provider or customer, there may be data residues that might expose sensitive data to unauthorized parties.

Compliance requirements: Various countries have varying regulations for data privacy. Because some public cloud providers don’t offer information about the location of the data, it is crucial to consider the legal and regulatory requirements about where data can be stored.

Multi-tenancy risks: The shared nature of public cloud environments increases security risks, such as unauthorized viewing of data by other customers using the same hardware platform. A shared environment also presents resource competition problems whenever one of the customers uses most of the resources due either to need or to being exposed to targeted attacks, such as DDoS (distributed denial of service).

Control and visibility: Customers have restricted control and visibility over the cloud resources because the cloud provider is responsible for administering the infrastructure. This introduces additional security concerns that originate from the lack of transparency. Customers need to rethink the way they operate as they surrender the control of their IT infrastructure to an external party while utilizing public cloud services.

Security responsibility: In a cloud the vendor and the user share responsibility forsecuring the environment. The amount of responsibility shouldered by each party can change depending on the cloud model adopted.

Summary

This chapter has focused on many concepts that will be important in the rest of the book. First, the NIST definition of cloud computing and the three cloud computing models defined by NIST (Infrastructure as a Service or IaaS, Platform as a Service or PaaS, Software as a Service or SaaS) have been described. Next, the four major cloud deployment models – private cloud, public cloud, community cloud, and hybrid cloud, were surveyed and described. This was followed by an analysis of the economics of cloud computing and the business drivers. It was pointed out that in order to quantify the benefits of cloud computing, detailed financial analysis is needed. Finally, the chapter discussed the major technological challenges faced in cloud computing – scalability of both computing and storage, multi-tenancy, and availability. In the rest of the book, while discussing technology, the focus will be on how different cloud solutions address these challenges, thereby allowing readers to compare and contrast the different solutions on a technological level.

Go ahead – enjoy the technology chapters now and demystify the cloud!