A guide for manufacturers to build mobile cloud computing services based on behavior

The natural subtle chemical relationship between mobility and behavioral services has made a major profit-making opportunity, as the dynamic nature of mobile services and mobile usage patterns makes this relationship fully manifest in cloud computing.

But choosing the right technology and functionality to seize this rare opportunity is not easy. It requires strategic thinking and forward-looking planning for virtualization patterns, network architecture, management, and billing.

The challenge of optimizing the cloud architecture for behavioral mobile cloud computing services is that the necessary service content is still evolving. In the past, running the Operations Support System (OSS), Billing support System (BSS), http://www.aliyun.com/zixun/aggregation/14045.html ">" on a unified platform Network management systems and service delivery platform (SDP) applications do not feel much pressure. As a result, most carriers have a variety of hardware, operating system (OS), and middleware components that build basic modules for the early building of mobile services.

What is the correct virtualization model for mobile cloud computing services?

It is likely that a new service feature must be created specifically for the cloud computing environment, but that depends on what type of cloud computing is supported. Today, platform diversity seems to be a testament to the infrastructure of Service (IaaS) cloud computing based on virtualization and support for multiple customer operating systems, but future requirements for flexible application portfolios require a service-oriented architecture (SOA).

The challenge of optimizing the cloud architecture for behavioral mobile cloud computing services is that the necessary service content is still evolving.

In a sense, IaaS is considered to be managed virtualization with two-tier software-one for creating an execution environment for an application and the other for managing and scheduling applications in a pool of resources created by virtualization.

There are also two modes of virtualization: a stand-alone virtual machine (VM) that shares the hypervisor mode of the hardware platform and a managed operating system mode that the virtual machine runs as a task in the managed operating system. The difference between these two patterns is how they balance the need to host multiple application platform requirements and the need to create an optimized SOA framework for the future.

Hypervisor virtualization creates a highly flexible IaaS cloud computing, but it may also present a greater challenge in building a mobile cloud-based cloud service. The biggest advantage of IaaS mode is that it can cloud hosting at least some BSS components--that's where the virtualization platform is compatible--but each virtual machine can effectively become a stand-alone computer that can be managed. As a result, this model generates operating costs. In addition, interprocess communication links components of advanced Mobile cloud computing services may be more difficult to collaborate in hypervisor mode because the virtual machine addresses for application components may change as resources are redistributed.

While similar devops rules are designed to reduce these problems in cloud computing, if these new, mobile cloud-based services require a lot of new development work, the Cloud management program pattern will be phased out over time.

In virtualized managed operating system mode, a virtual machine is a "container" in a normal operating system and is essentially handled as a task. The operating system's multitasking capabilities then enable it to host applications specifically written for the behavior-based mobile cloud computing service. Applications and components that require virtualization can then share resources with applications and components that do not need to be virtualized.

As a result, managed operating system mode is more efficient than running on customer-specific cloud computing applications and components, as it does not replicate the operating system and middleware as it replicates the IaaS virtualization solution. However, compared to the IaaS mode, the virtualized managed operating system pattern implements a lower degree of isolation between virtual machines (or other applications running on the managed operating system) between VMs.

Choose the best network architecture for mobile cloud computing services

The need to resolve and isolate mobile application components, especially those provided by third parties, is an important factor in choosing the best network architecture to support the behavior-based mobile cloud computing services.

Functionally, this cloud computing is comparable to a content delivery network (CDN), which must contain an internal information path and a proxy for the user to gain access to the service. Like a CDN, the architecture that supports the behavior-based mobile cloud computing services must provide the operator with a comparable dimension in building those services ' private networks.

Some operators and vendors are considering Software Definition Network (SDN) technologies such as OpenFlow to improve quality of service (QoS) through improved traffic engineering, and improve security by explicitly managing traffic across all components. Other vendors may want to build their own strategy around MPLS or IPV6.

Regardless of architectural considerations, if the cloud Portal is used to provide mobile cloud-based services to users, the data, processing, and even content needs of those services can be connected to the portal using an optimized private network rather than an open Internet. Developers and partners can be authorized to be trusted insiders; Of course, an open Internet can also be used for these connections.

Are there too many unresolved technical challenges?

There are also many areas where operators may be required to customize their work or use external professional services.

As mentioned above, DevOps is an emerging subject designed to be able to implement cloud computing service components as application slices in the container. You can then manage these containers on the server and use automation policies to connect over the network. Open source cloud computing is the main way to do this, and it's easy to understand their current status by using DevOps as a keyword for online searches on these projects ' Web sites.

(Responsible editor: The good of the Legacy)