BPO and Cloud Computing - Case Study Example

? BPO and Cloud Computing (case study) Data Centers Data centers are very essential for computing resources in a controlled environment and under centralized management which helps an organization to operate around the clock or according to the needs of the business. These computing resources include web and application servers, messaging servers, file and print servers, mainframes, application software and the operating systems that are used to run them, storage sub systems, network infrastructure and storage area network. These applications range from external e commerce and business to business application to internal financial and human resources. An enterprise can have one or more than one data centers. The benefits of the data centers includes supporting business operations round the clock, rapid deployment of applications and consolidation of computing resources and maintenance for sustaining business functions. Enterprises make special arrangement of both the facilities that house the equipment and personnel required for such an operation. These facilities support high concentration of network infrastructure and server resources. A data center is based on the availability of power capacity, cabling, cooling capacity, temperature and humidity control, physical security like restricted access and surveillance systems, fire and smoke systems, rack space and raised floor. Data centers support high speed communication between servers, storage and devices of storage (Arregoces & Portolani, 2004, P.21). It also increases the availability of mirroring, clustering and replication between database systems and storage devices. These facilities helps in storing the data in multiple locations thus lowering the chances of site failure that prevents chances of normal system operations. Site failures are recovered by the replica of data at different sites thereby creating the need for distributed data centers, distributed server firms and obvious transport technologies for enabling communication between them. Data centers can be local as well as one that is located overseas. A local data center does not rely on connectivity over the internet and therefore negates a possible failure point. Local data collection, correlation and processing are easier with local data centers. Corrective actions can be immediately taken to fix problems. The closer the management components are to the source of the problems, the3 more independent the firms can be for supporting their infrastructure. Corrective management activities can be taken even if the network to the management server is down. Continuous and autonomous monitoring is possible with the help of local data centers. A local database has access to all data sources which are not remotely accessible. In addition to remote interfaces like SNMP, scripting, WMI, the local data centers also have sensors for log files, discovery information, performance data sources- basically anything that is available locally. A local data centre can perform correlation which helps in reducing flood of data into meaningful set of events. Instead of sending thousand log file records over the network to the server for analysis, local data centers filters out extraneous information and allows important problems to be sent to a centralized event console. It helps in reducing false alarms and increases the productivity and efficiency as short term peeks are ignored and no time is spent in dealing with false alarms. It also helps in suppressing duplicate events that is based on time interval, count of similar events or both. Identical messages that occur within a very short period of time are suppressed and the first event is only forwarded to management server (Murugesan & Gangadharan, 2012, P. 21). Local actions can be executed without the intervention from the management server. The disadvantage of local data center is that keeping each client side database in sync is very difficult. Programming, administration and maintenance of data become nearly impossible. Data integrity, security, data and information theft, data corruption can easily be replicated if databases are used to update other client systems.  Outsourcing of Data Centers The outsourcing options available for data center services are numerous and have many benefits. The myriad service providers offer everything from simple data storage plans to a complete transition to cloud. The risk associated with outsourcing a data center is huge. Adequate building space for the data center, maintenance of adequate security, hiring a staff to manage it, keep pace with government regulations and many such factors are associated with it. Data center outsourcing is an option for many companies today and it can be done successfully if it is carefully planned. However sending all IT resources into the cloud, with no assessment of the potential hidden expenses and pitfalls can be disastrous. Costs can quickly increase thereby making traditional in-house data center approach look much more appealing. Potential Risks of Outsourcing Data centers. Security- It is a perennial concern for data center outsourcing. By allowing an outside party to provide IT services for an enterprise, other people can look into the loop of all data and computer activities of that enterprise. Although many service providers are very scrupulous about securing both physical and virtual facilities, the risks are still present and in some of the cases, it is compounded. Data center service providers provide all the usual security measures like malware protection, physical site security, authenticated access, data encryption and employee background checks. However other security risks may be peculiar to these providers. For example collocation providers may allow customers access to the data center which means that unless special precautions are taken, competitors or unknown people can see or access data that can compromise data security. The fact that more people have access to company specific data automatically increases the risk of a data leak or other breach. Security violations can lead to financial damage bringing unwanted attention from government regulators. Compliance- Data security measures should meet certain standards. When data centers are locally situated level of compliance with regulations is easy, but when it is outsourced, one has to trust another company to maintain that level. If the service provider serves a number of customers in different industries, it is unlikely to be aware of regulations specific to any particular business. If data of any specific organization is housed in a server or storage device along with that of someone else, the chances of violation some law or regulation will be very high. This can lead to confiscation of equipment shared by several customers i.e. alleged violator’s assets are seized along with those of companies who have nothing to do with the situation (Clark, 2011, P. 11). Accountability and Service- One virtually unavoidable and inherent problem with outsourcing is that the service provider serves more than one company. As a result, the provider has to split its attention among a number of companies and individuals. Thus the value received from outsourcing may not be what an organization expects. Gaining a certain level of accountability with a provider can require complicated legal documentation and disputes can be expensive to resolve. Disaster Recovery- Another area where necessary legal documentation and various allegiances leads to trouble while outsourcing is disaster recovery. If the service provider suffers some disaster then recovery may require attention to numerous customers. Priority of a particular organization may depend on a number of factors like how the provider operates, existing agreements etc. The main concern is how the service provider handles a disaster and its response to expectations. Availability- Service providers often have scheduled maintenance affects the ability to access certain data or services. Even unplanned outages can hamper the productivity. Even an in-house data center requires downtime for maintenance and recovery from unplanned event, but outsourcing means trusting someone else with service availability. Switching providers can be expensive and time-consuming, but poor service can also cost. Adding more links to the chain- Business relies on data center services to meet both internal needs and the needs of customers. When your data center is kept in house, the chain that connects the organization to required resources is generally shorter. When outside companies are involved to provide services, links are added to the chain. More links means a greater chance of a failure somewhere along the line (Nash, Darukhanawalla & Bellagamba, 2011, P. 17). Cloud Service Data Centers Cloud service data centers are termed as mega data centers, having on the order of thousands or more servers that draws tens of Mega-Watts of power at peak (Greenberg, Hamilton, Maltz, Patel, 2009, P. 14) Massive data analysis applications are a natural fit for a mega data center where problems require huge amounts of fast RAM, massive numbers of CPU cycles, and massive disk I/O bandwidth. Cloud service applications are often build on one another. Having large numbers of servers in the same location eases the systems design and lowers the cost of efficiently supporting applications with and associated communication needs and multiple dependencies. The advantages of using such data centers are- Large economies of scale- The size of cloud scale data centers, some of which are approaching 1,00,000 severs presents an opportunity to the organization to leverage economies of scale not present in the data centers of enterprises, though the upfront costs are high. Scale Out- Organizations often optimize physical space and number of devices, consolidating the workload into a small number of high-price servers and hardware devices. Cloud service data centers helps in distributing workload over large numbers of low cost hardware and servers. The innovations in cloud service data centers can benefit the enterprises through outsourcing of computing and storage to cloud service providers like scaling down and adapting technologies and business models from cloud service providers. As Bob, I would adopt cloud service data centers because as estimated, the growth of the enterprise will also increase its need of storing large amount of data. Both approaches to data centers local and outsourced have their benefits and drawbacks. Budget is also a critical concern. Although cloud computing can save lots of money, it depends on particular business and IT needs. Cloud or any other outsourcing approach can end up to be costing more than the in-house approaches. Unexpected costs owing to lack of availability, security breaches, failed audits and inadequate disaster recovery can easily blow away any savings from relying on service providers. References Arregoces, M &Portolani, M. 2004. Data Center Fundamentals. USA: Cisco Press Murugesan, S & Gangadharan G. R. 2012. Harnessing Green IT: Principles and Practices. USA: John Wiley & Sons. Greenberg, A, Hamilton, J, Maltz D. A, Patel, P. 2009. The Cost of a Cloud: Research Problems in Data Center Networks. Nash, D, Darukhanawalla, N & Bellagamba, P. 2011. Interconnecting Data Centers Using VPLS. New Delhi: Pearson Education India Clark, J. 2012. Risks of Outsourcing Your Data Center [Available on] [Assessed on 23 March 2013] Read More