Understanding Data Centers: Purpose, Working Model, And Operational Aspects

Purpose of Data Center Networking

This paper intends to discuss about the data center and discuss its different operations. The data center is defined as the house of network, storage and computational resources which are interconnected with the help of a network connection. The resources of the data centers are interconnected with each other by the help of data center network (DCN) [1]. The growing demand of the cloud services has led to increase the need and efficiency of connecting a large number of servers. The data center is divided in to three tiers based on the tree network topology and it includes three network switches. The three network switches are aggregate, access and core layers. There are about 236 data centers in Australia which includes the data centers of Sydney, Brisbane, Melbourne, Adelaide and Perth. The examples of famous Australia based data center are Equinix of Sydney, Global Switch Sydney West, Global switch Sydney East, NEXTDC and VOCUS Communications [2]. This report would discuss about the purpose, working model, scalability issues, features and weakness and Socio-Technical aspects of the data centers. It can be concluded that this report would be effective in providing all the aspects of data centers.

Save Time On Research and Writing
Hire a Pro to Write You a 100% Plagiarism-Free Paper.
Get My Paper

The purpose of data center networking is to provide accommodation to multiple data center tenants with different types of jobs. The servers in such a network are the constituents that provide users with the service requested by them. The data center network are composed of the servers that manages the job and attends the request of the client. It contains switches for connecting the devices and the routers for the purpose of forwarding the packets. It also contains the controllers which are used for managing the workflow among the network devices and the gateways that serves as the junction between the data centers [12]. The field of data center is concerned with the architecture operation and the relevant services. The applications are composed of multiple number of hosts which runs a single component. The field of application of data center offsite backup. The organizations are concerned about subscribing to the backup service with the help of data center. Backups can be taken in combination with the location tapes and local servers on the tape. The basic application of the data centers are CRM and ERP system.

The data centers are centralized area where equipment for networking and computing are stored and they are used for collecting, processing, storing, distributing and providing access to the required data [14]. It contains a large number of servers for the better operation and this server are stacked in rows. Thus data centers are also referred to as server farm. The infrastructure allows the large numbers of servers being controlled by small number of people.

Basic Operations of Data Centers

The basic operation is dependent on the working between the clients and the servers in a network using the network resources. The resources on this network are likely to share a mapping system which is common for everyone and they are based on the standards of networking technologies. The map sharing is based on the Ethernet, Internet protocol and different networking topologies for the modern networking system. It uses the layer IP address or the IP routing which are designed for providing a transitional forwarding agent [4]. This agent required in the process is called routers.  It allows the moving of packets to data. In the next step the TCP and IP protocol is used by the routers for passing the data packet to each other. The common technology of data center that can be used in the operation is called Ethernet. It is utilized for connecting the devices with each other with the help of media access control or the MAC address. These basic networking technology have specific limitation which can be overcome by using additional protocols. Some of the example of this additional protocols are OpenFlow and VXLAN [3]. It can be said that these component are efficient in forming the infrastructure for the operation of the data centers. With the development of the technologies like virtualization the need of these components are likely to be fulfilled by the recently developed software [13].

The scalability is defined as the ability of the network to construct and expand with the help of the design which that can provide accommodation to the new device and the increase traffics without having any impact on the applications. The scalability should not have effect on the cost per port and workflow of the data center. The process of scaling is expected to be linear in the context of cost and the performance [8].

Save Time On Research and Writing
Hire a Pro to Write You a 100% Plagiarism-Free Paper.
Get My Paper

The scalability is the biggest challenge for the data center network connection. The increase in the use of cloud services have led to this issue. It requires the data center to scale up the number of the nodes which are thousands in the number. In order to achieve higher scalability, there is a need for the data center network to deliver a bandwidth of high cross-section [5]. The current data center architecture which are available is three tier architecture. This architecture provides high over-subscription ratio and poor cross section bandwidth. The use of flat three data center network can oversubscription in ratio of 1:1 and cross section bandwidth which is very high. It faces issue of low scalability which is limited to total number of cell. The recommendation for avoiding the issue of scalability is use of D cell as it can offer more scalability. The limitation of D cell is that it provides bad performance in the condition of network load. The issue is found to be high in the case of one to many network service.

Scalability Issues

Security

The security is an important feature of data center networking. The infrastructure of data center are built in the secure structure and in the safe zones. The security includes the multiple security layer, biometric access control for the employees and the contractors, video surveillance in data centers. The data centers ensures that the access is provided to only the authorized and appropriate people [6].  The details of access are retained and logged and the logging details are saved in the database of the data centers. The data centers of the Australia such as Equinix of Sydney, Global Switch Sydney West, Global switch Sydney East, NEXTDC and VOCUS Communications are highly secured. The geographical locations of these data centers are also in the main areas of Sydney, Brisbane, Melbourne, Adelaide and Perth.

Availability

The standard data centers are redundant to power supply and climate control. The availability of resources and favorable weather conditions ensures the proper working of the critical services provided by the data centers [15]. The availability of backup can help as a motivating factor in the performance of specific mission critical service and urgent data operations.

Compliance

It includes the internal and external audits of the client. This feature helps the data centers to demonstrate compliance to their standards and commitments. The external assessment used by the data centers are two per year. This compliance feature helps in accessing the facilities of the data centers [7].

Environmental Features

The data centers have started using the cold-aisle control system considering the environmental aspects in the mind. It uses a mix of mechanical cooling and climate cooling to generate reasonable, consistent and effective conditions of environment. The data centers of Australia are implementing this techniques for being sustainable and best. The networking data center of Sydney and Brisbane uses this technologies for the efficient environmental conditions.

Power

The data centers are using multiple power feeds from number of different sources. The data centers are also availing back from the multiple numbers of diesel power generator and UPS in order to avoid the situation of power loss [16]. The issue of power loss can affect this huge data center and the service providers and clients to a large extent. Thus this feature are effective in avoiding the power loss condition in the mission critical operation.

Network Operation Centers

The data center operation requires the features of the Network operation Centers in order to ensure the better performance of the centers. It is used to ensure the standards of security and assistance to the client request [9]. The experienced team of network operator are assigned to each data centers. The data centers of Australia are likely to have in-house network operation team for instant support.

Security

The weakness of the data centers are that one to one communication with the server and the client is reduced. There are specific security related issues as the dependency on the data service provider are increasing. The other weakness are infrastructure creation and expenditure of cost. The challenges that are faced by the data centers are due to natural calamities, operation modes and the climate conditions [11]. Thus it is recommended to the set up the data centers at the favorable locations. The challenges are faced when the location of the data centers are at very distance from the network hub. It can cause trouble to the in-house executives from the networking team.

The social aspects of the data center networking are efficiency in energy consumption, monitoring of the data centers, capacity planning and performance management. The technical aspects of the data center networking are data security, real time data reporting, and resource management. The data centers should have independent infrastructure and facilities which are not outsourced. The other aspects are maintainability and the computerized maintenance process. The social aspects of the data center are that the staffs should be trained and should be kept near to data centers [10]. The networking team and expertise should be available in the data center. It can provide better performance and aided back up to the centers. The fulfillment of social and technical aspects of the data centers are subjected to successful operation and management and data center. It can enhance the delivery of the services to the various data provider and cloud service provider. The technical aspects are capable of providing the quality to the service and networking requirements of the clients.

Conclusion

It can be concluded that this paper is efficient in providing the report on the data center networking. It defines the data center as the house of network, storage and computational resources which are interconnected with the help of a network connection. The resources of the data centers are interconnected with each other by the help of data center network (DCN). The report provides the details about the purpose and field of data center and the working model of the data center. It provides the three tiers in which data centers networking infrastructure id divided. It is effective while explaining the scalability issue in data center and it defines scalability as the ability of the network to construct and expand with the help of the design which that can provide accommodation to the new device and the increase traffics without having any impact on the applications. It also provides a nutshell description of the features and the weakness of the data center in networking. The features of data center described in this paper are availability, power, environmental aspects, compliance and security. The paper ends with providing the socio-technical aspects of the data centers. Therefore it can be summarized that this paper appropriately illustrates the concepts of data center and its aspects.

Availability

References

Bari, Md Faizul, Raouf Boutaba, Rafael Esteves, Lisandro Zambenedetti Granville, Maxim Podlesny, Md Golam Rabbani, Qi Zhang, and Mohamed Faten Zhani, “Data center network virtualization: A survey,” IEEE Communications Surveys & Tutorials15, no. 2, 2013: 909-928.

Kliazovich, Dzmitry, Pascal Bouvry, and Samee Ullah Khan, “DENS: data center energy-efficient network-aware scheduling,” Cluster computing16, no. 1, 2013: 65-75.

Zheng, Kuangyu, Xiaodong Wang, Li Li, and Xiaorui Wang, “Joint power optimization of data center network and servers with correlation analysis,” In INFOCOM, 2014 Proceedings IEEE, pp. 2598-2606. IEEE, 2014.

Perelló, Jordi, Salvatore Spadaro, Sergio Ricciardi, Davide Careglio, Shuping Peng, Reza Nejabati, George Zervas et al, “All-optical packet/circuit switching-based data center network for enhanced scalability, latency, and throughput,” IEEE Network27, no. 6, 2013: 14-22.

Ports, Dan RK, Jialin Li, Vincent Liu, Naveen Kr Sharma, and Arvind Krishnamurthy, “Designing Distributed Systems Using Approximate Synchrony in Data Center Networks,” In NSDI, pp. 43-57, 2015.

Porter, George, Richard Strong, Nathan Farrington, Alex Forencich, Pang Chen-Sun, Tajana Rosing, Yeshaiahu Fainman, George Papen, and Amin Vahdat, Integrating microsecond circuit switching into the data center. Vol. 43, no. 4. ACM, 2013.

Liu, Zhongjin, Yong Li, Li Su, Depeng Jin, and Lieguang Zeng, “M2cloud: Software defined multi-site data center network control framework for multi-tenant,” In ACM SIGCOMM Computer Communication Review, vol. 43, no. 4, pp. 517-518. ACM, 2013.

Chen, Min, Hai Jin, Yonggang Wen, and Victor CM Leung, “Enabling technologies for future data center networking: a primer,” Ieee Network27, no. 4, 2013: 8-15.

Yao, Fan, Jingxin Wu, Guru Venkataramani, and Suresh Subramaniam, “A comparative analysis of data center network architectures,” In Communications (ICC), 2014 IEEE International Conference on, pp. 3106-3111. IEEE, 2014.

Bilal, Kashif, Marc Manzano, Samee U. Khan, Eusebi Calle, Keqin Li, and Albert Y. Zomaya, “On the characterization of the structural robustness of data center networks,” IEEE Transactions on Cloud Computing1, no. 1, 2013: 1-1.

Chen, Zhen, Wenyu Dong, Hang Li, Peng Zhang, Xinming Chen, and Junwei Cao, “Collaborative network security in multi-tenant data center for cloud computing,” Tsinghua Science and Technology19, no. 1, 2014: 82-94.

de Souza Couto, Rodrigo, Stefano Secci, Miguel Elias Mitre Campista, and Luís Henrique Maciel Kosmalski Costa,”Reliability and survivability analysis of data center network topologies,” Journal of Network and Systems Management24, no. 2, 2016: 346-392.

Yeganeh, Soheil Hassas, Amin Tootoonchian, and Yashar Ganjali, “On scalability of software-defined networking,” IEEE Communications Magazine51, no. 2, 2013: 136-141.

Barroso, Luiz André, Jimmy Clidaras, and Urs Hölzle, “The datacenter as a computer: An introduction to the design of warehouse-scale machines,” Synthesis lectures on computer architecture8, no. 3, 2013: 1-154.

Wang, Peng, and Hong Xu. “Expeditus: Distributed load balancing with global congestion information in data center networks.” In Proceedings of the 2014 CoNEXT on Student Workshop, pp. 1-3. ACM, 2014.

Lu, Yifei. “SED: An SDN-based explicit-deadline-aware TCP for cloud Data Center Networks,” Tsinghua Science and Technology21, no. 5, 2016: 491-499.

Understanding Data Centers: Purpose, Working Model, And Operational Aspects

Purpose of Data Center Networking

This paper intends to discuss about the data center and discuss its different operations. The data center is defined as the house of network, storage and computational resources which are interconnected with the help of a network connection. The resources of the data centers are interconnected with each other by the help of data center network (DCN) [1]. The growing demand of the cloud services has led to increase the need and efficiency of connecting a large number of servers. The data center is divided in to three tiers based on the tree network topology and it includes three network switches. The three network switches are aggregate, access and core layers. There are about 236 data centers in Australia which includes the data centers of Sydney, Brisbane, Melbourne, Adelaide and Perth. The examples of famous Australia based data center are Equinix of Sydney, Global Switch Sydney West, Global switch Sydney East, NEXTDC and VOCUS Communications [2]. This report would discuss about the purpose, working model, scalability issues, features and weakness and Socio-Technical aspects of the data centers. It can be concluded that this report would be effective in providing all the aspects of data centers.

Save Time On Research and Writing
Hire a Pro to Write You a 100% Plagiarism-Free Paper.
Get My Paper

The purpose of data center networking is to provide accommodation to multiple data center tenants with different types of jobs. The servers in such a network are the constituents that provide users with the service requested by them. The data center network are composed of the servers that manages the job and attends the request of the client. It contains switches for connecting the devices and the routers for the purpose of forwarding the packets. It also contains the controllers which are used for managing the workflow among the network devices and the gateways that serves as the junction between the data centers [12]. The field of data center is concerned with the architecture operation and the relevant services. The applications are composed of multiple number of hosts which runs a single component. The field of application of data center offsite backup. The organizations are concerned about subscribing to the backup service with the help of data center. Backups can be taken in combination with the location tapes and local servers on the tape. The basic application of the data centers are CRM and ERP system.

The data centers are centralized area where equipment for networking and computing are stored and they are used for collecting, processing, storing, distributing and providing access to the required data [14]. It contains a large number of servers for the better operation and this server are stacked in rows. Thus data centers are also referred to as server farm. The infrastructure allows the large numbers of servers being controlled by small number of people.

Basic Operations of Data Centers

The basic operation is dependent on the working between the clients and the servers in a network using the network resources. The resources on this network are likely to share a mapping system which is common for everyone and they are based on the standards of networking technologies. The map sharing is based on the Ethernet, Internet protocol and different networking topologies for the modern networking system. It uses the layer IP address or the IP routing which are designed for providing a transitional forwarding agent [4]. This agent required in the process is called routers.  It allows the moving of packets to data. In the next step the TCP and IP protocol is used by the routers for passing the data packet to each other. The common technology of data center that can be used in the operation is called Ethernet. It is utilized for connecting the devices with each other with the help of media access control or the MAC address. These basic networking technology have specific limitation which can be overcome by using additional protocols. Some of the example of this additional protocols are OpenFlow and VXLAN [3]. It can be said that these component are efficient in forming the infrastructure for the operation of the data centers. With the development of the technologies like virtualization the need of these components are likely to be fulfilled by the recently developed software [13].

The scalability is defined as the ability of the network to construct and expand with the help of the design which that can provide accommodation to the new device and the increase traffics without having any impact on the applications. The scalability should not have effect on the cost per port and workflow of the data center. The process of scaling is expected to be linear in the context of cost and the performance [8].

Save Time On Research and Writing
Hire a Pro to Write You a 100% Plagiarism-Free Paper.
Get My Paper

The scalability is the biggest challenge for the data center network connection. The increase in the use of cloud services have led to this issue. It requires the data center to scale up the number of the nodes which are thousands in the number. In order to achieve higher scalability, there is a need for the data center network to deliver a bandwidth of high cross-section [5]. The current data center architecture which are available is three tier architecture. This architecture provides high over-subscription ratio and poor cross section bandwidth. The use of flat three data center network can oversubscription in ratio of 1:1 and cross section bandwidth which is very high. It faces issue of low scalability which is limited to total number of cell. The recommendation for avoiding the issue of scalability is use of D cell as it can offer more scalability. The limitation of D cell is that it provides bad performance in the condition of network load. The issue is found to be high in the case of one to many network service.

Scalability Issues

Security

The security is an important feature of data center networking. The infrastructure of data center are built in the secure structure and in the safe zones. The security includes the multiple security layer, biometric access control for the employees and the contractors, video surveillance in data centers. The data centers ensures that the access is provided to only the authorized and appropriate people [6].  The details of access are retained and logged and the logging details are saved in the database of the data centers. The data centers of the Australia such as Equinix of Sydney, Global Switch Sydney West, Global switch Sydney East, NEXTDC and VOCUS Communications are highly secured. The geographical locations of these data centers are also in the main areas of Sydney, Brisbane, Melbourne, Adelaide and Perth.

Availability

The standard data centers are redundant to power supply and climate control. The availability of resources and favorable weather conditions ensures the proper working of the critical services provided by the data centers [15]. The availability of backup can help as a motivating factor in the performance of specific mission critical service and urgent data operations.

Compliance

It includes the internal and external audits of the client. This feature helps the data centers to demonstrate compliance to their standards and commitments. The external assessment used by the data centers are two per year. This compliance feature helps in accessing the facilities of the data centers [7].

Environmental Features

The data centers have started using the cold-aisle control system considering the environmental aspects in the mind. It uses a mix of mechanical cooling and climate cooling to generate reasonable, consistent and effective conditions of environment. The data centers of Australia are implementing this techniques for being sustainable and best. The networking data center of Sydney and Brisbane uses this technologies for the efficient environmental conditions.

Power

The data centers are using multiple power feeds from number of different sources. The data centers are also availing back from the multiple numbers of diesel power generator and UPS in order to avoid the situation of power loss [16]. The issue of power loss can affect this huge data center and the service providers and clients to a large extent. Thus this feature are effective in avoiding the power loss condition in the mission critical operation.

Network Operation Centers

The data center operation requires the features of the Network operation Centers in order to ensure the better performance of the centers. It is used to ensure the standards of security and assistance to the client request [9]. The experienced team of network operator are assigned to each data centers. The data centers of Australia are likely to have in-house network operation team for instant support.

Security

The weakness of the data centers are that one to one communication with the server and the client is reduced. There are specific security related issues as the dependency on the data service provider are increasing. The other weakness are infrastructure creation and expenditure of cost. The challenges that are faced by the data centers are due to natural calamities, operation modes and the climate conditions [11]. Thus it is recommended to the set up the data centers at the favorable locations. The challenges are faced when the location of the data centers are at very distance from the network hub. It can cause trouble to the in-house executives from the networking team.

The social aspects of the data center networking are efficiency in energy consumption, monitoring of the data centers, capacity planning and performance management. The technical aspects of the data center networking are data security, real time data reporting, and resource management. The data centers should have independent infrastructure and facilities which are not outsourced. The other aspects are maintainability and the computerized maintenance process. The social aspects of the data center are that the staffs should be trained and should be kept near to data centers [10]. The networking team and expertise should be available in the data center. It can provide better performance and aided back up to the centers. The fulfillment of social and technical aspects of the data centers are subjected to successful operation and management and data center. It can enhance the delivery of the services to the various data provider and cloud service provider. The technical aspects are capable of providing the quality to the service and networking requirements of the clients.

Conclusion

It can be concluded that this paper is efficient in providing the report on the data center networking. It defines the data center as the house of network, storage and computational resources which are interconnected with the help of a network connection. The resources of the data centers are interconnected with each other by the help of data center network (DCN). The report provides the details about the purpose and field of data center and the working model of the data center. It provides the three tiers in which data centers networking infrastructure id divided. It is effective while explaining the scalability issue in data center and it defines scalability as the ability of the network to construct and expand with the help of the design which that can provide accommodation to the new device and the increase traffics without having any impact on the applications. It also provides a nutshell description of the features and the weakness of the data center in networking. The features of data center described in this paper are availability, power, environmental aspects, compliance and security. The paper ends with providing the socio-technical aspects of the data centers. Therefore it can be summarized that this paper appropriately illustrates the concepts of data center and its aspects.

Availability

References

Bari, Md Faizul, Raouf Boutaba, Rafael Esteves, Lisandro Zambenedetti Granville, Maxim Podlesny, Md Golam Rabbani, Qi Zhang, and Mohamed Faten Zhani, “Data center network virtualization: A survey,” IEEE Communications Surveys & Tutorials15, no. 2, 2013: 909-928.

Kliazovich, Dzmitry, Pascal Bouvry, and Samee Ullah Khan, “DENS: data center energy-efficient network-aware scheduling,” Cluster computing16, no. 1, 2013: 65-75.

Zheng, Kuangyu, Xiaodong Wang, Li Li, and Xiaorui Wang, “Joint power optimization of data center network and servers with correlation analysis,” In INFOCOM, 2014 Proceedings IEEE, pp. 2598-2606. IEEE, 2014.

Perelló, Jordi, Salvatore Spadaro, Sergio Ricciardi, Davide Careglio, Shuping Peng, Reza Nejabati, George Zervas et al, “All-optical packet/circuit switching-based data center network for enhanced scalability, latency, and throughput,” IEEE Network27, no. 6, 2013: 14-22.

Ports, Dan RK, Jialin Li, Vincent Liu, Naveen Kr Sharma, and Arvind Krishnamurthy, “Designing Distributed Systems Using Approximate Synchrony in Data Center Networks,” In NSDI, pp. 43-57, 2015.

Porter, George, Richard Strong, Nathan Farrington, Alex Forencich, Pang Chen-Sun, Tajana Rosing, Yeshaiahu Fainman, George Papen, and Amin Vahdat, Integrating microsecond circuit switching into the data center. Vol. 43, no. 4. ACM, 2013.

Liu, Zhongjin, Yong Li, Li Su, Depeng Jin, and Lieguang Zeng, “M2cloud: Software defined multi-site data center network control framework for multi-tenant,” In ACM SIGCOMM Computer Communication Review, vol. 43, no. 4, pp. 517-518. ACM, 2013.

Chen, Min, Hai Jin, Yonggang Wen, and Victor CM Leung, “Enabling technologies for future data center networking: a primer,” Ieee Network27, no. 4, 2013: 8-15.

Yao, Fan, Jingxin Wu, Guru Venkataramani, and Suresh Subramaniam, “A comparative analysis of data center network architectures,” In Communications (ICC), 2014 IEEE International Conference on, pp. 3106-3111. IEEE, 2014.

Bilal, Kashif, Marc Manzano, Samee U. Khan, Eusebi Calle, Keqin Li, and Albert Y. Zomaya, “On the characterization of the structural robustness of data center networks,” IEEE Transactions on Cloud Computing1, no. 1, 2013: 1-1.

Chen, Zhen, Wenyu Dong, Hang Li, Peng Zhang, Xinming Chen, and Junwei Cao, “Collaborative network security in multi-tenant data center for cloud computing,” Tsinghua Science and Technology19, no. 1, 2014: 82-94.

de Souza Couto, Rodrigo, Stefano Secci, Miguel Elias Mitre Campista, and Luís Henrique Maciel Kosmalski Costa,”Reliability and survivability analysis of data center network topologies,” Journal of Network and Systems Management24, no. 2, 2016: 346-392.

Yeganeh, Soheil Hassas, Amin Tootoonchian, and Yashar Ganjali, “On scalability of software-defined networking,” IEEE Communications Magazine51, no. 2, 2013: 136-141.

Barroso, Luiz André, Jimmy Clidaras, and Urs Hölzle, “The datacenter as a computer: An introduction to the design of warehouse-scale machines,” Synthesis lectures on computer architecture8, no. 3, 2013: 1-154.

Wang, Peng, and Hong Xu. “Expeditus: Distributed load balancing with global congestion information in data center networks.” In Proceedings of the 2014 CoNEXT on Student Workshop, pp. 1-3. ACM, 2014.

Lu, Yifei. “SED: An SDN-based explicit-deadline-aware TCP for cloud Data Center Networks,” Tsinghua Science and Technology21, no. 5, 2016: 491-499.