Hypervisor Virtualization Technology And Data Centers: Essay On Characteristics, Scalability, And Weakness.

This assignment assesses the following Unit Learning Outcomes; students should

1. be able to demonstrate their achievements in following.
2. Apply socio-technical contexts in modern network applications and management;
3. Utilise common and emerging types of middleware to design and manage networked applications;

Write a report addressing following question with the help of figures and example scenarios.

1. What is a Data Centre and discuss different operation performed in data centre, how data centres divided in Different tiers based on what criteria’s? Give example of few famous Australia based Data centres.
2. What is Hypervisor-Based Server Virtualization Technology, its use and explain at least two examples of it. Also, compare both examples based on their features and limitations.
3. What is container-based virtualization and why we need it? Discuss at least two examples of it and also, compare both examples based on their features and limitations.

Demonstration: Students should prepare presentation slides and take screen shots of slide and paste at the end of assignment draft. The tutor will ask each student 2-3 questions from submitted work. The demonstration from each student should not be more than 8 minutes.

Task 1

Write a report covering and detailing above applications while emphasizing on the following aspects:

• Field and Purpose of the application
• Working model of the application
• Scalability issues
• Features and Weaknesses of the application
• Socio-Technical aspects

Task 2

Prepare a PowerPoint presentation discussing the main features of the given applications. The PowerPoint presentation should have at least 10 slides and students should present for 6-8 minutes during lab time in Week 8.

Hypervisor virtualization technology is described as computer applications that create a virtual machine and runs it. It is a technology that is commonly used and it has enabled more than one operating system to run on one physical platform. On the other hand, data centers can be described as a facility within an organization that is central, located to carry out IT activities within an organization. These activities may include storage, retrieval and processing of data. Data centers are important to an organization since they promote communication through enhancing data exchange. These technologies have increased communication and business operations in many enterprises. This report describes their characteristics, scalability and their weakness in general.

This is a technology that explains how computers create virtual machines and run them. Operating system designers face serious problems while isolating software and allowing sharing of applications between them [1]. This shows that each application has different layouts but share some objects. Hypervisor strives to provide full separation between virtual machines by providing support in sharing of resources in computers. The technology has made single machines to host multiple software that are unrelated which run independent of the organization. The applications do not need to actually share data, they run on the same kernel as a single operating system. Currently, one can have different types of operating system installed in his machine and use them without facing any problems.

There are many ideas that enhance the use of hypervisor virtualization technology. They include [2].

• The technology is used in securing work environment in laptops.
• It enhances efficient use of hardware components by reducing usage of hardware systems. This lowers maintenance costs and power usage in any organization.
• Detection of malicious attacks in the real time.
• It increases availability especially on the server side by providing features that are not found on the server.
• Used in disaster recovery like data back up in case there is any loss.
• Debugging failing systems.
• The technology is also used by programmers in development of software and testing them.
• IT centers use the technology in merging servers onto effective hardware.
• Virtual technology is used in selling private servers to companies.
• It creates space in server rooms or data centers.

If the server needs to execute a hypervisor, it loads the technology into a client operating system   of the VM. Then the technology provides correct CPU resource to memory bandwidth for the machine. A VM can create requests to the hypervisor using many ways including API calls.

There two kinds of this technology:

1. Embedded or hosted hypervisors - They run as software using operating systems such as windows
2. Bare metal or native hypervisors - this occurs when hypervisors run on hosts hardware components to manage the VM.

Comparison between embedded and native hypervisors

• Native run directly on the server while embedded run on operating systems

It was difficulty to replicate hypervisors before. Currently they are highly scalable and flexible. This means that they can be changed and moved if it is required. For this to occur, one needs to know about methods that are used in data replication. Volume in machines must be replicated to enable the technology to run more efficiently.

The technology has following features and weakness [3].

 Features

• It can be easily installed
• Easy to use

Weakness

• Extra costs are involved - installation of the technology requires additional hardware and software thus increasing the cost.
• Software licensing - the software that is used with this technology requires license which increases the cost of setting up and using the technology.
• Its installation and usage require skilled personnel on the technology. This creates extra labor.

Technical aspects of the network is that it provides high data security especially when it is used on the server side. It works with high scalability and flexibility.

Description

The growing need of organizations to store and retrieve data for future use has led to development and establishment of data centers. Today data centers contain thousands of computers with appropriate and significant bandwidth requirements. Different organizations will have different layouts of their data centers depending on their requirements and the type of data that is required to be stored for future use [4]. Well-designed data centers meet the following requirements:

• Scalable bandwidth - for faster communication between two hosts that are found within the network.
• Economics of scale - the cost of installation must be very low.
• Backward compatibility - the system should be backwardly compatible with the hosts and other resources in the network.

Data centers have the following functions in any network [5].

• Backup of data - it backups data automatically on the server.
• Compliance - it defines the normal roles of the organization.
• It determines the software that is running in the network.
• Configuration - it is used for changing management of devices that are being used in the network.
• Remediation - it identifies and fixes issues in the network.
• It provides consistency in the network
• It is used in patching where by it automatically downloads and installs any updates that are available in the system.
• It’s used in application management and printer management.
• It’s used as a service desk where by users can request for some details and wait for provisioning.

A data center is made up of power, storage and applications that are required to support an organization. The data center’s infrastructure is at the center of the IT from which all the content is passed through. A good model of data center is highly planned and scalable. It has high performance.

The figure below shows a structure of a working model of data center that is linked to the cloud for data backup. This design is called basic layer design.

Figure 1. Basic layer design of a data center as illustrated in [11, fig 1]

A data center should be highly flexible to deploy and support new features. This is because flexible data centers have ability to support new application features. We have different design models of data centers as explained below.

Multi-Tier Model

This is the common model of data center mainly used in Http–based applications. The model includes web, applications and database tiers of servers. It uses a software that runs as a different process on a machine using interposes communication when the process are build such that they run from different  machines.  It promotes security and resilience of the data centers. The security of the data centers is increased because an attacker can access a web server without breaking into application and database servers [6]. Loading of a balance network traffic promotes resilience between the tiers. Firewalls are placed between the tiers to provide security in the data center. The following diagram shows a multi-tier model of a data center.

Fig 2. Multi-tier model of a data center as illustrated in [11, fig 2]

Server Cluster Model

These data centers are used for many purposes because they are highly available and have increased power. This model enhances high performance of clusters which have many forms. Clusters have common goals of combining multiple CPUS together. This promotes high speed of the network. This kind of data models is mostly used in universities and in military research for unique applications. This kind of data center is also used in weather stations and in seismology for analysis.

Purpose of the Application

Data centers have some scalability issues as explained below [7].

Data center scalability is the ability of data center to keep working even if its volume and size is changed. Data centers increase the speed of data processing if they are increased into a large size. These companies maintain data centers in order to save money. When a company wants to improve their data centers, they determine the extent to which the data center can be improved to. If a data center is over built, then there will be wastage of resources. Although sometimes business organization overbuild data centers because it is not easy to estimate the required size of a data center.

More so, the scalability and the flexibility of a data center enable the organization to build quickly at scale and in the nick of time to meet their client's expectations. Likewise, the server farm versatility is additionally bringing the advantage of high-thickness, vitality productivity and also the capacity to lessen the expenses over the lifetime of data center speculation.

Moreover, one approach to make an adaptable server farm is by keeping up its area. A few organizations are generally searching for a territory to locate the server farm. Indeed, there is an open door for the future development that can be assume control without having more risky things for its extension, or with no issue of the expensive comforts.

The development of informal communication and thriving utilization of cell phones is the undeniable reason of the need of scalability. Data centers adaptability itself likewise expands the high security for records that the organizations keep. It influences the organizations to end up more dependable and trust-commendable. It is additionally about the requirement for speed. As the outcome to be focused, the organizations ought to have a focused on approach for server farm flexibility and versatility.

Features

• Has high availability
• Highly secured
• Requires dependable power
• Stores large amount of data.
• It has high compliance
• It requires room for expansion.

   Weakness

• It reduces one to one communications among customers
• Security related issues increase due to its reliability
• Experiences challenges due to natural disasters.
• Data center is located far way hence incur travel expenses.

The purpose for the Data Center and Server Room Standards is to describe the base prerequisites for outlining, introducing, anchoring, checking, keeping up, securing, and decommissioning a server farm

Data centers in Australia

• Global switch Sidney west
• Equinix
• Nextdc

Also called operating system virtualization is where the layer being virtualize is running as application within an operating system. The operating system runs on the kernel and the hardware node is isolated. The isolated hardware is called the container [8].

• It’s used as a backup for data.
• It provides environment for applications to run
• It provides security to its contents
• They compute a lot of data on the same server
• Can be used to locate resources easily
• They are used in decreasing the cost of operating system
• They are the best to be used in micro services

This technology works by sharing the host kernel with its contents e.g. Containers and hosts.

Fig 3. The Container technology architecture [8, fig 2]

Working Model

Containers provide necessary conditions for applications to run on the host operating system. Applications like servers which require direct access to hardware components will directly affect the functioning of the system. Container helps bypassing the emulation layers.

Scalability issues

Containers tend to have high scalability and adaptability but they are not perfect for every workload that they handle. Containers have received high virtualization technology that has great impact on its scalability [10]. It supports cloud computing and application development

Features

• It has increased scalability resources
• It has increased running periods
• Has significance over other technologies

Weakness

• It is not perfect for all task
• it grapes with dependence
• It has weak isolation
• It has limited tools

The reason for the container virtualization technology is to portray the base requirements for sketching out, presenting, tying down, checking, keeping up, anchoring and decommissioning the kernel.

Conclusion

Data centers should be highly protected since they store very important information of an organizations. Other technologies like hypervisor technology are of much significant to the society since they allow two different operating systems to run on the same machine. This has made life much simpler since individuals can now purchase only one PC and use different operating systems on the same laptop.

References

[1]Hwang, J.Y., Suh, S.B., Hoe, S.K., Park, C.J., Ryun, J.M., Park, S.Y. and Kim, C.R. Xin on ARM: System virtualization using Xin hypervisor for ARM-based secure mobile phones. In Consumer Communications and Networking Conference, 2008. CCNC 2008. 5th IEEE (pp. 257-261). IEEE, 2016

[2] Younge, A.J., Henschel, R., Brown, J.T., Von Laszewski, G., Qiu, J. and Fox, G.C. Analysis of virtualization technologies for high performance computing environments. In Cloud Computing (CLOUD), 2011 IEEE International Conference on (pp. 9-16). IEEE, 2010

[3] Abramson, D., Jackson, J., Muthrasanallur, S., Neiger, G., Regnier, G., Sankaran, R., Schoinas, I., Uhlig, R., Vembu, B. and Wiegert, J. Intel Virtualization Technology for Directed I/O. Intel technology journal, 10(3).,2016

[4]Al-Fares, M., Loukissas, A. and Vahdat, A, August. A scalable, commodity data center network architecture. In ACM SIGCOMM Computer Communication Review (Vol. 38, No. 4, pp. 63-74). ACM, 2008

[5] Koomey, J. Growth in data center electricity use 2005 to 2010. A report by Analytical Press, completed at the request of The New York Times, 9, 2013

[6] Al-Fares, M., Radhakrishnan, S., Raghavan, B., Huang, N. and Vahdat, A. Hedera: dynamic flow scheduling for data center networks. In Nsdi (Vol. 10, No. 8, pp. 89-92). 2009.

[7] Meng, X., Pappas, V. and Zhang, L., Improving the scalability of data center networks with traffic-aware virtual machine placement. In INFOCOM, 2010 Proceedings IEEE (pp. 1-9). IEEE, 2017.

[8] Xavier, M.G., Neves, M.V., Rossi, F.D., Ferreto, T.C., Lange, T. and De Rose, C.A, Performance evaluation of container-based virtualization for high performance computing environments. In Parallel, Distributed and Network-Based Processing (PDP), 2013 21st Euromicro International Conference on (pp. 233-240). IEEE, 2010.

[9] Adufu, T., Choi, J. and Kim, Y. Is container-based technology a winner for high performance scientific applications? In Network Operations and Management Symposium (APNOMS), 2015 17th Asia-Pacific (pp. 507-510). IEEE, 2015.

[10]. Gerlach, W., Tang, W., Keegan, K., Harrison, T., Wilke, A., Bishop, J., D'Souza, M., Devoid, S., Murphy-Olson, D., Desai, N. and Meyer, F. Skyport: container-based execution environment management for multi-cloud scientific workflows. In Proceedings of the 5th International Workshop on Data-Intensive Computing in the Clouds (pp. 25-32). IEEE Press, 2014.

[11] Barroso, L.A., Clidaras, J. and Hölzle, U. The datacenter as a computer: An introduction to the design of warehouse-scale machines. Synthesis lectures on computer architecture, 8(3), pp.1-154. 2013.