Key Capabilities in Cloud Service Management

All the players in the cloud management ecosystem have a role to play in ensuring consistent delivery and operation of cloud services. The cloud provider has primary responsibility for the management functions of its cloud service offerings. The cloud consumer needs to monitor the operation of these services with a focus on end-user experience. The cloud broker can act as an intermediary between cloud providers and the cloud consumer to ensure that the cloud consumer has a consistent management view across its hybrid cloud environment.

To effectively manage a hybrid environment requires that all elements of the environment be managed in an automated and standardized way. Therefore, the six primary capabilities that any service provider must include are

check.png Management of customer accounts

check.png Management of provisioning and configuring resources

check.png Management of service catalogs

check.png Management of performance

check.png Management of security

check.png Management of interoperability

We describe the key capabilities in hybrid cloud management in this section.

Managing customer accounts

Cloud providers need to ensure that the basic business relationship with its customers runs smoothly. Due to the self-service nature of cloud service delivery, an essential aspect of cloud management deals with automating processes of customer account management. These processes include opening and closing accounts; establishing user accounts and authorizations; establishing, managing, and metering pricing; tracking usage; and managing service levels according to policies.

Provisioning and configuring services

Provisioning automates the process of accessing the right computing resources needed by the service provider or the consumer of services. Provisioning enables pools of resources ranging from compute, storage, networking, or application services, to be instantly available based on assigned business policy rules.

Without this level of automation, the provisioning process can take months since it often requires organizations to acquire, configure, and set up resources manually. By focusing on standardization and automation, cloud providers can make the provisioning process easier, faster, and more cost effective. In order to ensure that the right services are provisioned, it is critical to manage change. Therefore there must be configuration services in place to keep track of changes to everything from actual applications to operating systems, and the like.

Service catalogs

A service catalog plays an important role in ensuring that service providers adequately document the characteristics of services so that service consumers know where to look for the offerings they need. The catalog is essentially a list of internal and external services available to an organization. It is more cost efficient and faster for cloud consumers to select a standardized offering from a catalog. Ideally, an enterprise IT organization wants to provide a comprehensive catalog of services that is easily understood and used by its customers.

In some situations, an organization may need to provide different catalogs for line of business users and developers to make sure each group gets what they need from the catalog. Although, if a cloud consumer has multiple service catalogs to manage, this can make the process more cumbersome.

In a hybrid world where cloud consumers need to choose between services offered across the internal data center, private and public clouds, and hosted environments, a service catalog can provide the means to increase the overall agility of the organization. Using a service catalog can help to ensure that cloud consumers select the right set of offerings based on the rules and priorities of the business.

Some of the information available in a service catalog is as follows:

check.png Definition of the service and what it means to business users

check.png Various options for each service and any limitations

check.png The requirements for executing the service

check.png The cost of the service and whether costs vary by service level options

check.png Creator of the service

check.png Who can use the service

check.png How to request a service

check.png Whether there are bundles of services (service bundles can protect consumers from ordering groups of services that don’t work together, as in unsupported platforms)

check.png The associated service level and how the service performs against this service level

Management of performance

Cloud providers need to monitor cloud services to ensure they meet agreed-upon service levels. This means that the performance of servers, networks, and virtualized images in the cloud providers’ environment need to be measured and monitored — both individually and collectively — to ensure the environment is tuned to satisfy all business requirements.

A standardized and automated system needs to be in place to track, trace, and audit all aspects of performance. For example, bandwidth, connectivity, and scalability are performance characteristics that should be monitored. In addition, automated systems should be designed to quickly identify the root cause of the hardware or software failures so performance can be restored.

These systems should answer questions such as:

check.png Is the cloud infrastructure performing as expected?

check.png Are identified performance problems occurring randomly at regular intervals?

check.png Which performance problems are most severe and need to be given top priority to find the root cause and resolve the issue?

check.png How can performance be improved?

Cloud providers vary in how much performance level detail they make available to cloud users. Even when performance statistics are shared at a very granular level, the cloud user can’t control the provider’s environment. This lack of insight and control of cloud service performance can present significant challenges to the cloud user organization that is also a cloud provider to its customers. For example, a PaaS platform provider might not meet its service level requirements for its customers if its cloud infrastructure provider suffers a service disruption. Additionally, an enterprise IT organization acting as a cloud provider to users of its hybrid cloud needs to guarantee service to its customers without having control over the performance of public cloud services in its environment.

You can put management services in place to monitor performance of your data center and private cloud, but you still need a way to monitor statistics about service performance from your public cloud service providers. There are some vendors that provide monitoring and management products that will help to improve your ability to monitor performance of your public cloud resources and, hence, improve the overall service level of your hybrid cloud.

Security

Cloud providers are responsible for securing the physical and logical aspects of the infrastructure and operation system in the cloud environment. For example, cloud providers need to validate appropriate levels of network, operating system, and middleware security to prevent intrusion and denial-of-service attacks. Comprehensive governance and security strategies are a non-negotiable requirement for a cloud provider to maintain good customer relationships. A proactive approach is required to protect against security threats that change constantly. That being said, each cloud provider will take a different approach to security, and it is the responsibility of the organization consuming the cloud services to ensure that its security requirements are met. Security management must be viewed as a shared responsibility between the cloud provider and the cloud user. (See Chapter 15 for more detail on cloud security.)

Interoperability and portability

In hybrid cloud environments, interoperability and portability refer to how cloud users can move their tools, applications, virtual images, and so on, between the data center and private and public clouds. A high level of interoperability means that cloud users can easily move workloads from one environment to another with very few integration issues. Interoperability also means that an application will work the same if it is moved from one cloud environment to another. Portability is a related concept that is often used to describe what is involved when a cloud user wants to move a SaaS or PaaS based application from one vendor to another.

There are many reasons why cloud providers and cloud consumers need interoperability. Cloud providers need interoperability in order to quickly add additional resources from another cloud provider if they need to ramp up to meet customer demand. In hybrid environments, cloud consumers want the flexibility to move workloads between traditional and cloud resources based on which environment is best suited for the specifics of that workload. The optimal environment may change over time based on business priorities and the specifics of the resource environment. In addition, consumers want the flexibility to move from one cloud vendor to another if they are dissatisfied with the services of the vendor or their business requirements change. In addition, a cloud provider may want to use more than one cloud provider for failover protection. In addition, a cloud user may contract with one IaaS provider and wants to easily switch to another IaaS for additional capacity.

Achieving interoperability and portability in hybrid clouds can be very complex. What are cloud providers doing to make it easier to move applications between clouds? Cloud providers like IBM, Rackspace, and Amazon are building in templates to help with the integration process. These templates can provide some level of governance over where you should place a workload. However, there is still a lot of work to be done in this area.

There are many groups working to improve standards so that it is easier for organizations to share data and applications across multiple cloud environments. One organization making progress in developing standards for interoperability and portability is the Open Services for Lifecycle Collaboration (OSLC). Linked Data — using the web to connect related data that was not previously linked — is being used to allow for increasing levels of interoperability between applications and workloads in hybrid clouds. The OSLC is working on the specifications for linked data to be used to federate information and capabilities across cloud services and systems. This will become an evolutionary step in cloud management. It will help to eliminate the time-consuming and complex coding required moving applications and components across environments.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.138.117.75