Three models and challenges of mobile cloud computing

In the past year, cloud computing technology has been seen as the "future of the Web," according to Sadie Consultants, 2010-2013, China's cloud computing services market size of the annual composite growth rate will reach 91.5%. The development of cloud computing is not limited to PCs, and with the booming mobile internet, cloud computing services such as mobile terminals based on mobile phones have emerged. Mobile cloud computing has become a part of our daily lives.

Mobile Internet is the main thrust of the development of mobile cloud computing, with the popularization of 3G, mobile cloud computing will attract rapid development. The Ministry of Industry and Information Technology released March 30 of the Communications industry Operation report shows that 1-April, the national 3G users a net increase of 30.547 million households, to 158.971 million households.

Mobile Cloud Computing has a wide range of applications. Common applications in the business sector include job dispatch, scheduling, internal mail, workflow and other mobile business management related services, as well as logistics, inventory control and other mobile commerce applications. The application of individual users is generally the mobile network access, E-commerce (shopping, mobile payment), entertainment and other personalized needs.

At the same time, the Government also in the public service, military and other areas of the application of mobile cloud computing, such as: e-government, electronic health services, tourism (electronic map navigation, tourism positioning services), intelligent transportation, environmental monitoring, wartime communications.

Three modes of mobile cloud computing

In my opinion, mobile cloud computing mainly has three kinds of modes: the mobile device is connected to cloud, Micro Cloud (Micro Cloud) and cloud to increase the efficiency of mobile device.

As we all know, the services provided by computers involve three parts: storage, calculation, operation/display. For example, playing a video, requires storage space to store the video, then need to compute resources to decode, and finally have a monitor and keyboard to achieve image output and user action. Cloud computing is about making multiple high-performance machines responsible for one or more of these parts. The author also defines different patterns based on the assignment of tasks.

Accessing the cloud via mobile devices

Access to the various SaaS (software as services) provided by the cloud via mobile devices. The common cases of access through mobile devices are: icloud, mobile search services, Google Mail, Google Maps, backup, and so on, for Apple users to provide data storage and synchronization services. From these applications, mobile cloud computing is often misunderstood as a mere SaaS, and in fact the various data processing associated with it is carried out in the cloud, which involves the service of IAAS (infrastructure, service).

How do you achieve intelligent access to the cloud? On the one hand, on-demand network connectivity technology, currently can be implemented by a wide range of wireless access technology support heterogeneous data access to achieve on-demand network connectivity, such as: GPRS,WCDMA/HSPA (high-speed packet access), LTE (3GPP long-term evolution Technology), microwave access to global interoperability, CDMA2000, wireless local area network. On the other hand, wireless access to the allocation of resources, through the terminal environment configuration, network nodes, or user-side sensors to reduce the waste of wireless resources and efficient management of wireless access to the allocation.

Micro Cloud (Micro Cloud)

The mobility of cloud computing through a single adaptive network based on mobile devices, such as the use of personal mobile device resources to provide a virtual mobile cloud-micro cloud. In this mode, mobile devices become part or all of the cloud services hardware, which eliminates the need for access to the Internet's cloud, does not require a central server, and enables data storage and processing services.

The Hyrax project at Carnegie Mellon University is now porting the Hapdoop operating framework to Android handsets. Multiple Android phones are distributed to store and process data through mutual collaboration. There are two main applications of this model:

First, the application of sensor data. The smartphone uses its sensors (such as GPS, accelerometer, optical sensor, temperature sensor, compass, etc.) to collect data, analyze and process the data to provide query services for the application. For example, use the location and mobile information collected by a user's mobile phone in different geographic locations to predict current traffic conditions.

Second, multimedia social applications. Multimedia data includes the use of smartphones, audio and video files recorded by the camera, and entertainment files such as music or videos that the user stores beforehand. For example, a user enters a picture or a section of audio or video, which is used to look for files that most closely match the user input from multiple user phones.

However, there is a lack of universality in the micro-cloud model, on the one hand, there is a lack of effective management of storage and computing resources. On the other hand, the process of data storage and access lacks optimization.

Cloud to increase the efficiency of mobile devices

This mode seamlessly and transparently migrates some or all of the data storage and computing tasks of mobile devices to the cloud in order to overcome the problem of mobile device resource constraints. At present, most of the data storage and processing needs of mobile phone applications are done on mobile devices, through the expansion of the cloud channel, data storage and processing can be implemented in the cloud outside the mobile phone, so that the mobile device itself to overcome the limitations. This pattern of mobile cloud computing is often viewed as IAAS (infrastructure as a service) or PAAs (platform as a service).

Because of its intensive performance, mobile computing has greatly improved the recognition ability of clients in speech, natural language processing, computer vision and drawing, machine learning, planning and decision making. However, mobile devices are not sufficient to cope with the need for large amounts of data storage and processing, and these features are more power-hungry and mobile devices will soon be off.

Real-time monitoring of the environment through mobile devices is based on the cloud-efficiency model, which began to be applied in the field of intelligent transportation. It is realized by a set of 3D transmission system, and the real-time monitoring of traffic condition is realized according to the two-dimensional image data collected by mobile devices in different locations and the corresponding sensing data. Data in the completion of the code upload to the cloud, in the cloud to complete storage, decoding, 3D graphics rendering, and finally presented to the user in 3D.

Cloudlet (Yun) is also a common cloud model for mobile device efficiency. The main idea of Yun is to migrate computing tasks on mobile devices to servers in a local area network. The infrastructure in this LAN that provides computing services to users is called the cloud. Compared with cloud services on the Internet, small cloud has the advantage of high data transmission rate, thus reducing user response time. In addition, Yun uses virtualization technology to achieve a brief customization of cloud infrastructure services. Users need to customize the cloud service resources before use, and then clean up the resource.

The main advantage of Cloudlet is that it solves the security problem partly, Yun the computing environment between users and users, and the user computing environment and the system service environment, so as to improve the application security. But Cloudlet is not perfect, it only allows a small number of users at the same time access, while its management by the merchant responsible for many times the lack of necessary professionalism.