Planning an Application Server Migration - Case Study Example

PLANNING AN APPLICATION SERVER MIGRATION Planning an application server migration is an inevitable but complex process that must be well documented for organizational success. The organization must therefore understand the risks, savings and the impacts of the migration process. This paper explains the steps needed for the organization to achieve the goals of its intended migration process. Application Server Migration Use Terminal Services to help meet criteria for providing applications and support for the remote offices. The company has to implement a network application server through which employees will connect via remote desktop services. The preferred application server pool is windows server 2003 as it is able to provide load balancing, fail over and clustering capability that will give us the opportunity to put more emphasis on business logic. This server will store all of the organizations files, data, and information. Applications will not run on the users’ workstation but on the server. The user’s workstation will send keystrokes and mouse movements to the remote system while the server will reply by sending screen images to the user’s workstation. This remote session will enable the users run applications and desktop environments that they would not have run because of the high quality hardware or bandwidth required. The server will execute applications as either pooled sessions, or as virtual machine environment infrastructure also known as (VDI). This will allow the employees connect to the server via any device and collaborate (share their files) with any other employee on the server. In terminal services, programs are installed only once and at one location i.e. the server thus easing task handling process. Programs that require a lot of bandwidth and processing power will also function better with the terminal service. Additionally some programs tend to execute poorly when accessed from the internet but will execute just fine when hosted with terminal servers because they are processed from the server and the user only sees the screen. Consequently, all the users will have excellent performance regardless of their connection type. Dial up users on hosted terminals can use the internet as well as execute applications with speeds of high speed internet access (Menzel, 2006) Remote servers are controlled environments working continuously. They are highly secured by locked doors, cabinets, and video surveillance. Moisture content is also monitored. For this system to be successful, the company must implement the following technologies: i. Implementation of a full fidelity user experience to enhance user experience. This includes the implementation of a multi-monitor support, ability to redirect support for windows media player, uploading and downloading audio capability, in addition to higher bitmap acceleration for the #D applications and high quality media content like silver-light and flash. ii. RD connection broker that allows the implementation of a unified administrator experience for terminal services and emerging virtual machine based remote desktops iii. RemoteApp and desktop connections to ease the access to resources. It is simpler to configure and update programs from the satellite offices. Employees will be able to print from their remote desktops because the server will act as a cloud desktop thus giving the employees access to the network printer (Pivot, 2014). The terminal services will have two modes of operation i.e. remote administration and application server mode. Remote administration mode will allow concurrent connections while at the same time give the administrators remote access to manage the server via terminal services client. This minimizes the impact on the server’s performance. The application Server mode will provide access to a Windows 2000-based computer desktop and its programs for employees connecting via Terminal Services client so that programs response time is optimized. Determine how application pools will help to reduce costs The company will run IIS 6.0 in worker process isolation mode therefore it will be able to group web applications and websites into application pools. A worker process is shared among applications inside an application pool. Individual worker processes in turn execute as separate instances of the worker process executable thus the worker processes that service an application process are always separated from the other worker processes that service other application processes. Thus makes user that there exists process boundaries such that whenever applications are assigned to one application pool, the processes outside it do not affect it. This reduces cost of repairing the whole system whenever there are failures. IIS 6.0 was designed to offer high security, reliability and improved performance. Incidentally we will use IIS 7.0 because of its ability to convert the high quality base of IIS 6.0 web server to a modular and extensible platform with needed support for the vital modem deployment and management scenario. IIS 7.0 also offers a high degree of performance improvements in the socket layer and windows authentication. It also includes better functionality in terms of flexibility. It is implemented modularly hence gives the user the ability to add on or remove some of the features without affecting its performance and other features. The organization will hence be capable of adding and removing features to improve functionality and reducing threat risks (Technet Magazine, 2008) Ad and NTFS permissions Ad groups and NTFS file system use should be configured for security and appropriate access. Active directory offers a secure environment that is mainly used for communication with networked components in the organizational server. Active desktop have central components that are used for sharing data and availability. Improving security to the active desktop involves adding more attributes to the database so that the active desktop transmits different environments to the forest so that security is enhanced. All of the users must be given single but unique usernames and password as a NetID so that the system the system can give the foundation for the exchange and SharePoint services. NTFS consists of both standard and special permissions. NTFS is an operating system part that can be remotely used by the user to find needed files. Within windows server 2003 NTFS employs the ‘closed software code’ in dynamic file. The home and public folders are put on separate volumes from the OS prior to permission setting. This makes the backup [process simple as the organization will not be required to back up public folders and applications. It also provides the organizations recommended user access restriction. The organization then creates groups and allocates user accounts to each employee with regard to member priorities. They are then assigned permissions such as read and write. The owner of the group is assigned full control of the files. This access permission ensures that employees have access to files created by their counterparts remotely. Windows SharePoint Services to provide support for the additional remote users Microsoft SharePoint is the preferred web application framework. It incorporates intranet, content management and document management. It can offer collaboration, and file management functionality. It includes a fully functional web technology together with standard APIs. The application offers a central management as an application platform in addition security controls for the implementation of its requirements. One of its major advantages is that in integrates directly to the IIS 7.0 therefore easing the management and provision of servers as required by web site hosts. This organizational will use it to store and track electronic files and applications that have been used but its employees. It will then keep records of the different versions made by the remote users. The users have the ability to quickly see who made what changes to files that interest them. Metadata- data about data is another component of the file thus users can quickly set up views to look at how the data has changed periodically. It has the ability to centrally store information, a functionality that can be exploited to delete duplicate files in the organization. It also connects to extranets thus provide a functionality that can be used to offer encrypted web facing access to the remote users (Husman, 2007). The organization can then utilize this to offer a collaboration environment to its employees. The organization will also use it to provide an alternate access mapping that allows a site to be accessed at different urls by different employees so that it provides remote access to the employees while at the same time strengthening the security because the organization will authenticate each url separately to offer more security. This can also be used to change the site into an intranet to employees of a certain department and as an extranet to the rest. Providing high availability is a complex process that involves striking a balance between scale maximization and isolation. This calls for the implementation of the combination of the pool models. The organization implements multiple pool models. Individual application pool is given a unique purpose e.g. webservers consisting of static content being contained in one shared application pool. This may include web servers that host applications. This cannot affect the performance because static content cannot be affected by security and performance issues that are often linked to dynamic content. The remaining application pools are dedicated to the webservers that house dynamic content. This enables the organization to distribute enough resources to the dynamic servers. The organization will use shared hosting environment therefore the only configuration will be a hybrid. Examine how you can use high-availability technologies to manage costs High availability technology will allow the company perform maintenance with minimal downtime. The increasing density can affect the remote users. However, with high availability technology single firewalls, switch or single PDU failure cannot affect the company’s availability. This will guarantee continuous availability thus increased productivity and cost savings. The Oracle cluster ware is also another single high availability software solution that the organization will use thus do away with the cost of maintaining other software. Consequently, this will reduce the cost of maintenance. Replication is a high availability technology through which the company can maintain the state and configuration across its entire web front ends. Mass hosting, HA environments requires that the organization maintains the state and the configuration among the web front ends in the organization. Even though there exist some other configurations that must be implemented on the web fronts, the webservers configuration is the most vital. Application pools such as IIS 7.0 will minimize the cost by reducing the bandwidth costs. IIS 7.0 uses the HTTP compression thereby minimizing the size of the response time significantly. To add on, it is supported by all desktops and its decompression cost on the desktop hardware is low in comparison to the latency in savings from sending less data. Since compression relies on Content Encoding negotiation as defined in the HTTP 1.1 protocol, enabling it is secure for clients that don’t have compression capability. The clients will get the same version of the content that hasn’t been compressed. It will also serve to provide the two compression features. Static compression will compress the files before the files are send and will also save the files in the disk. This will enable future requests to send the stored and compressed files without the additional cost and time of compressing them. In dynamic compression they are compressed in real time and therefore compress for the response by the applications. High availability technologies are very expensive to acquire. In contrast, technologies like the oracle Grid infrastructure can reduce the costs while at the same time offering greater scalability and optimize the organizations use of its networked resources (Oracle® Database, 2011). This organization must cautiously analyze the capabilities of the infrastructure to make sure that it produces the optimal combination of high availability so that the business can run without problems. The IT infrastructure that supports the user’s welfare must have high availability and should also be disaster proof so that costs of maintenance are reduced as the risk of loss of data is also reduced. Failure is always inevitable. To counter this organization must always back up data at regular intervals at their local servers and their replicas send to the redundant remote high speed connected servers. There must also be redundancy in the server and the servers must be synchronized at regular intervals. If there are any chances of failures this data should be backed up at regular intervals at their local servers and their replicas send to the redundant remote high speed connected servers. There is redundancy provided in the remote servers and they are synchronized. Take advantage of the change in licensing considerations Licensing terms will be changed because all of the applications will be installed in the application server. Employees and other users will access services from the server. This means that licensing will only be done in the server thus the deployment to the remote desktops will not require licensing. Often, when applications are installed on organizational workstations they require individual licensing that is costly. To add on, some applications require licensing for each user thus sky rocketing the costs. The organization takes advantage of the change in licensing by installing application in the application server and giving users remote access. This significantly reduces the cost. Manage network load balancing and avoid slowdowns in performance Load balancing enhances network efficiency by offering redundancy, scalability and better performance (Austin, 2013). Load balancing guarantees that network load is shared among the servers to eliminate network bottlenecks. Installing multiple servers to offer redundancy i.e. serve as alternative routes during network failure. There are numerous Web front ends, emphasis must be put on choosing the best option across the web servers. The configuration relies on the choice of load-distribution model. There exist numerous models of load distribution. The chosen two are load distribution and aggregate load distribution. Application Load Distribution This is a mechanism of distributing the load among the web fronts with reference to the servers function. It leverages on the application layer routing functionalities with the load balancers. This functionality will be used to subdivide the web farm. Static content will then be assigned to servers that do not have dynamic content. Servers that serve dynamic content such as ASP.NET will be assigned dynamic content. Furthermore, dynamic content will be subdivided with regard to their functions. PHP servers will be allocated PHP content whereas AS.NET servers will be allocated ASP.NET content. Some of the dynamic servers will be designed by aggregate load distribution model so that they share the same configuration for easier management. Configuring the print server Before the print server is configured the following must be done i. Check to ensure that the operating system has the right configuration. In windows server 2003 the print server relies on the configuration of the operating system and the operating systems services. Windows server 2003 default configuration works fine. However, for upgrades the settings should be confirmed with the default settings for services (microsoft, 2005) ii. The workstation must be joined to the Active Directory domain as a member server. Printer access can be restricted so that some domains can print while others cannot, print server to publish shared printers to Active Directory such that the domain users can search for the printers. However, when all of these is not needed, then there is no need for connecting to the domain iii. Ensure that all disk volumes use NTFS file system because it is more secure that FAT 32 iv. The printer must be deployed by the Casper Admin in order to grab the mappings. Then install LPD settings. To add drivers and mappings on each of the users use Casper remote. References Technet Magazine. (2008, September). Retrieved from Top 10 Performance Improvements in IIS 7.0: http://technet.microsoft.com/en-us/magazine/2008.09.iis.aspx Austin. (2013, September 25). avende. Retrieved June 12, 2014, from The need for Network Traffic Management(Load Balancing): http://avende.com/the-need-for-network-traffic-management-load-balancing/ Husman, G. (. (2007). Beginning sharepoint administration: Windows sharepoint services and sharepoint portal server. New Jersey: John Wiley & Sons. Menzel, W. P. (2006). REMOTE SENSING APPLICATIONS WITH. Madison, WI: University of Wisconsin. microsoft. (2005, January 21). microsoft technet. Retrieved from Print server role: Configuring a print server: http://technet.microsoft.com/en-us/library/cc775791%28v=ws.10%29.aspx Oracle® Database. (2011). Oracle® Database High Availability Overview. Retrieved from Oracle® Database High Availability Overview: http://docs.oracle.com/cd/B28359_01/server.111/b28281/hadesign.htm#g1007388 Pivot. (2014, June 12). Pivot. Retrieved june 12, 2014, from Hosted Remote Desktop Application Server: http://pivotnetworks.com/hosted-cloud/hosted-remote-desktop-application-server/ Sutton, C. (2008). J2EE development. Oxford express. Read More