Computer Network Server Configuration - Assignment Example

Building a Server HU Name Institution Professor Course Date Building a Server HU The computer network server is configurable efficient and reliable per the needs of a given business. The business having a critical nature inclusive of consulates, banks, corporate companies or the police department, the reliability cost of the server increases respectively. There are the majority of objects involved in the computer network including routers, hubs, bridges, Gateway, ISDN adapters, transceivers, Modems and firewalls, which reliably depends on a given server sharing resources and server. There are subsystems involved to enable the reliability of the server. Following tips and recommendations involved in the construction of secure and higher performing server, the business requires the business to consider server components (McCabe, 2007). Hard disk subsystem reliability The main server component prone to a failure is a hard disk meant for storing the server data. The hard disk is a considerable mechanical component in the server. In the process of selecting a disk drive, there should be a consideration of Reporting and the Self-monitoring technology. The technology enables the proper monitoring of disk drive parameters and capable of predicting a failure in the disk drives. Majority of the server vendors avails the hot swap disk drives meaning that they are capable of replacement while the existing server is still under the working mode (McCabe, 2007). For the provision of fault tolerance, implementation RAID (inexpensive disk redundant array) is necessary. The RAID is pliable under implementation through the installation of RAID disk controller present in the server. The RAID 5 is capable of providing fault tolerance through the addition of parity information of the written data on the present disk drives. The available parity information is utilizable in the regeneration of the data in case the disk experiences failure. A minimum of three disk drives is necessary for the implementation of the RAID 5 (parity striped set) with the maximum disk number involved being 32 in the RAID (McCabe, 2007). Uninterruptible power supply functions The functioning of a computer always needs the presence of power supply hence there is a need to determine the type of power supply that the computer will need. However, many companies bring their computers into the market with multiple power supplies. The multiple powers can always supply more than two computer servers in addition to sharing electrical load of the computer server (Aycock, 2008). Most of the computer users always prefer the use of multiple power supplies since the servers operate independently causing the continuous supply of power. There is no disruption of power since the failure of one supply does not always disrupt the functioning of the other power supplies. Another advantage of the supply server is their activities, which present it to be hot swappable, just like other hot swappable hardware (Aycock, 2008). Computer server software The server software always serves as an important element in the functioning of the compound. The software put on the server should always be steadfast implying that it should have not virus. This calls for the need of antivirus software that should always be updated according to the manufacturer’s direction. There is also need for a good firewall, anti spyware, malware and adware should be available on the server (Aycock, 2008). The server will also need backup software to counteract any effect of emergency. These will ensure that there is adequate security for the server subsequently making the server to work continuously. There is need for removal of software that is of much suspicion as malicious and can pave way for the viruses. Viruses are a threat to server since if a virus gets into the server it will always be easy for it to spread all over the server. The accessory of the server requires the service of NIC adapter. NIC adapter is crucial in the execution of attachment of the server. NIC is vital to the development and functionality of the server to the extent that failure of the adapter would result into malfunction of the server. There is minimal connection between the server and other computers with network services. Numerous servers offer the needed support to the redundant NIC adapters. These servers can undergo implementation to minimize the failure of the connection with other computers with network connections. If one NIC adapter fails to initiate communication between the server and other computers with the networks, one NIC adapter is applied to enhance communication with the server while other NIC adapters held reserve until the other NIC fails and the other NIC adapter undergoes automatic activation process. The NIC adapters are usually activated with the aim of improving the reliability and performance of the communication servers. This indicates that the servers would enjoy maximum connection with computers with network settings. The PCI hot swap slot provides the capacity to replace the faulty NIC adapters with minimal effects such as shutting down the server. Clustering is crucial to the development and realization of redundancy. Memory Performance The memory technology rarely fails to work following the step up it has attained. It is necessary for system network administrators to purchase a memory for the server. The memory should be in a position to support Error Correcting Code (ECC). The technology of Error Correcting Code has the capacity to detect and fix the particular bit error while at the same time reporting the multiple bit memory errors (Rosenthal 2004). Apparently, a server requires a substantial amount of memory for smooth performance of its functions. A lower amount of memory would disrupt the smooth functioning of the memory. The ability of the server to perform its functions not only depends on the level of memory but also other specific to the server. A slower memory as indicative of certain servers may results into slow boot up. Similarly, there would be slow response time as well as possibility of the slower memory bringing system processes to a halt. Hence, memory is an essential element for every computer server (Rosenthal 2004). The functions of Uninterruptible Power Supply (UPS) Uninterruptible Power Supply (UPS) play very crucial role in the event that there is a power failure. UPS usually provides power back up for a period of between three and four hours. During this particular duration, the computer user easily setup system server to save the data. All users of the network may also access the store of imperative data on the server. Software always controls the server’s shut down process. The software is installed on the server. The Uninterruptible Power Supply is vital in areas with sufficient network coverage. Such environment could also be having considerable number of UPS in a single network. UPS can share the load and in the process provide constant power supply to the supply (Rosenthal 2004). Backup Generator Working In cases where the server is required to work throughout then machines like the mass server, web, email, the server that has the role of backing up power, in addition to the power backup, generator are essential. The power backup systems are machines that run on diesel fuel to produce power to be in use by other servers. A great number of manufacturing companies are participating in the construction of these emergency power generators (Bender, 2010). One advantage of these emergency power generator systems is that they have the function allowing them to auto start the moment the main power source is cut shot. Most of them take approximately nine seconds to start running. If a company has installations of UPS backup then the generator starts after the UPS consumption is complete. Cooling Fans Role Having plenty of hardware installed in a computer chassis, there is a need of having a cooling system. Most machines construction is with a fan to play the role of a cooling system in a machine. Other developers are using water-cooling systems for their gaming systems. When comparing these two cooling systems, one notices that the water-cooling system is apparently more expensive than the fans (Bender, 2010). With this reason, most manufacturing companies opt to settle for the fans as a mode of a central processing unit cooling system. If the cooling system fails, then the machine will automatically shut down, this results from excessive heat buildup in the server. Data server Location When setting up a server unit, it is advisable that the main server be stored in a well-ventilated room that eases the cooling system thus; the room should be dust free. The room where the server is placed requires the best security systems in that only authorized personnel’s have access to it. This helps avoid any damages or tempers with the server that may cause its failure. Server Design Analysis The server hardware components show several trends in baseline power utilization, and cooling in the server configuration. The costs in power consumption and cooling are comparable to hardware costs. The CPU hardware, power and cooling is the only components with a total cost of 20 percent to 22 percent. In addressing the achievement a compelling performance in the server, design solution holistically, address multiple components (Ekman and Stenstrom, 2005) This holistic solution in the design of the server focuses on four factors: reducing overall CPU hardware and power cost by using high-volume lower-cost lower-power non-server processor, reduce power burden cost through packaging solutions, reducing cost of memory sharing across a cluster, and reducing overall disk component cost by utilizing lower power disks with emerging nonvolatile memory (Ekman and Stenstrom, 2005). Te memory design and architecture minimizes consumption of power in two ways: reducing the total amount of DRAM, and correspondingly the power, enabling an extended utilization of lower-power DRAM components. The memory blades access is less frequent, and at page granularity. Transfer time across the PCI link dominates the latency. One can leave the memory-blade DRAM in active power-down mode, and reduce power to about 90% in the DDR2. Architecturally, the server memory sharing design further improves by having a DMA I/O going directly across the second level memory (Ranganathan et al, 2006). The server cost model data sources are from a variety of contacts, mostly the public. Some of the values are from confidential interview communication and industrial estimates. References Aycock, J. (2008). Spyware and Adware. Berlin: Springer US. McCabe, J. D. (2007). Network analysis, architecture, and design. San Francisco Calif.: Morgan Kaufmann. Rainardi, V. (2008). Building a data warehouse with examples in SQL Server. Berkeley, CA: Apress. Caputo, T. C. (2003). Build your own server. New York: McGraw-Hill/Osborne. Rosenthal, J., Irwin, K., & Ebrary, Inc. (2004). PC repair and maintenance: A practical guide. Hingham, Mass: Charles River Media. Bender, M. (2010). MCTS guide to Microsoft Windows Server 2008 network infrastructure configuration. Boston, MA: Course Technology, Cengage Learning. Ekman, M. and Stenstrom., P. (2005). A Cost-Effective Main Memory Organization for Future Servers. Ranganathan, P., Leech, P., Irwin, D., and J. Chase. (2006). Ensemble Level Power Management for Dense Blade Servers. Read More