System Availability Management for SMEs - Essay Example

System Availability management for small to medium sized marine shipping companies in Hong Kong. Problem ment System availability means availability of a complete application when some one wants to make use of it (Michael Hawking 2002). Availability is measured end to end for any of the system application. All the components that make up the system or the application should be available to mark it as system available. Defining metrics and measuring the right availability of the system under such conditions is a major problem for small and medium industries. This is so because they do not have the needed skills to study the peculiarities of their business and the IT availability. This study aims at identifying the system availability and measuring it; enabling the company to improve upon their own performance. Major Issues System availability is normally considered as the uptime of the hardware. But for the users, it is not just the PC hardware that matters; all the components that make the application including software and other networking / printing components also ensure that the system availability is dependent on them. It becomes imperative that the system availability is evaluated under such conditions. According to Piedad and Hawkins (2001), availability of the systems is how it is perceived by the users. Hence users need to be checked out even before drawing the SLAs to ensure that the requirements of the users are driven rather than the thoughts of the CIO. This will ensure that the system is made available to the users in the right form and style. This lack of understanding and not knowing what is expected out of the IT systems is not known to a number of designers. This leads to a problem. The second issue arises when critical issues are not addressed and in lieu of the right fixes for the critical issues, lean solutions are used to fix them. This increases the probability of a failure at the earliest point of time. The chain is as strong as its weakest link! The four key elements that make up the complete chain needs to be tackled according to Piedad and Hawkins (2001), Processes, Data, Tools and Organization. For instance, a failure in the process could also lead to a downtime. It is not only the hardware that has to be up and running but also the entire system should be up. The software, availability of the data at the right instance, tools to run and use the applications needed and the entire organization of the system should be in line with the availability. A failure to maintain the entire chain up and running would lead to a down time (Robin Sahner 1996). The third issue comes when appropriate standby facilities are not maintained. This happens when there is no backup of data or the repairing features for the software under fault conditions are not maintained. All these issues need to be sorted out if the uptime or the system availability is to be maintained. In addition to this, the cost of ownership needs to be kept within the budgetary limits (Michael Hawkins 25 May 2001). Solutions to the Problem The solutions to the issues listed above: 1. In order to fully understand the requirements of the users and also to appreciate the user standpoint, it is important that they are involved at all levels of system design and implementation. This can be achieved by ensuring that the requirements are individually listed and analyzed after identification. Requirements identification and capturing has taken different perceptions over a period of time (Vinay Shet 2006). Methods like Scenario based requirements capture has been successful particularly in identifying hard and soft requirements of the clients (Carl Olefson 2005). 2. The weakest link is the one that would break first. In the entire system design, the availability has to be taken care of on the whole and not individually. It is therefore, important for the administrator to ensure that the entire system is taken into consideration and the issues concerning all of them are noticed and sent across sorted out (LSI 2006). Only then the issues would not arise. In the complete system, the failures can happen on the hardware or on the software or a combination of both in terms of system software. Only then the failure rate in this case could be avoided and the uptime maintained for the system. Clustering of hardware is another method of avoiding failures in hardware (Lixin Shen et al Apr 2005). 3. The second issue also takes the form of the third issue when the failure happens without appropriate standby facilities (Schiesser 2002). Many of the corporate entities create a hot standby for the server and the database for the system to be available with the continuous uptime and system availability. However, an appropriate data backup plan should be in place to ensure that data is never lost. This can be taken care of by providing for an online backup facility that would keep the system online at the same time taking the backup for safety of the data. The data could also get stored in a number of locations (more than one normally) to ensure that the data does not get corrupted (Cisco 2007) should something go wrong. All these issues can be sorted out with the methodology indicated below. Research Objectives This study will have the following research objectives: 1. The causes for the failure rates in Hong Kong for the small and medium sized shipping companies. This will be listed using the standard failure rates as identified by the observer during the survey. 2. The failure can also be caused by a lack of standard procedure for failure handling. A fault tolerant system should be designed and put in place if high dependability of the systems is noticed. This is important specifically in case of systems that are critical and are running on real time basis. The availability of the system, the data is also checked out and an evaluation process is worked out to identify the cause of the failure in all the surveyed and monitored locations. The system availability is monitored during the research. And the cause of the failures is also identified in addition to the systems that were in existence in the systems. 3. Cause of failures is taken in to consideration including human failure rates and the operational errors and other systemic errors that might have led to the system unavailability are also taken in to consideration during the research. This will help in identifying the various methods that are taken up for upkeep and the success rate of these methods. 4. Cost of all these activities and the methods to reduce total cost of ownership. Conclusion The research objectives will be taken up for the analysis and research that has to be done. Reasons behind drop in system availability if any and the cause for the failure of systems is taken up. In the course of the research, it is also taken into consideration, by conducting appropriate survey to collect data from the small and medium industries in Hong Kong specifically in the marine shipping businesses. In order to arrive at the right conclusion in the research, an unbiased study is done to fully understand the source of the problem. And also the solutions for the problems thus identified. References 1. Carl Olefson (2005) Meeting the Needs of the Small and Medium Sized Businesses. Oracle Corporation. Available at: http://www.oracle.com/database/idc_oraclesmb.pdf 2. Cisco (2007) Top Five Security Issues for small and medium sized businesses. Available at: http://www.cisco.com/en/US/netsol/ns644/networking_solutions_white_paper0900aecd804606fc.shtml 3. Floyd Piedad & Michael Hawkins (2001) High Availability: Design, Techniques and Processes. Prentice Hall, ISBN: 979-0-13-096288-1 4. Lixin Shen et al (2005) Achieving high availability and Performance computing with an HA-OSCAR cluster. Future Generation Computer Systems, Vol 21, Issue 4, pp 597 - 606. 5. LSI (2006) Continuous Application Availability for Small and Medium Businesses. Available at: http://www.lsi.com/documentation/storage/systems/1333_server_clustering_1106.pdf 6. Michael Hawkins (2002) Availability is a User Metric. Solutions for IT Professionals, Harris Kern's Enterprise Computing Institute. Available at: http://harriskern.com/index.phpm=p&pid=377&aid=76 7. Michael Hawkins (2001) The Total Cost of Ownership: The Driver for IT Infrastructure management. Available at: http://harriskern.com/index.phpm=p&pid=377&authorid=25&aid=7 8. Robin A. Sahner , Kishor S. Trivedi , Antonio Puliafito. (1996) Performance and reliability analysis of computer systems: an example-based approach using the SHARPE software package, Kluwer Academic Publishers, Norwell, MA. 9. Scheisser, Rich (2002) IT Systems Management: Designing, Implementing and Managing World Class Infrastructure. Prentice Hall. ISBN: 0-13-087678-X. 10. Vinay Shet (2006) Service Availability. Available at: http://www.cs.umd.edu/vinay/Presentations/818.ppt#1 Read More