Jan
10
2012

Database Performance

With the way information are created, transferred and consumed nowadays, it is not unimaginable the pains a company puts into their IT infrastructure. Every day you hear people investing in applications, software, physical servers and other IT-related stuff. And if you’re into HR or finance, you’ll know how much money goes out of the company in terms of consultancy services and payroll. Such steps, however, don’t make database performance management revolutionary because attempts to improve starts not with an initiative but with a distress call from an irate user. While all departments of the enterprise should carry the task of enhancing database performance, many companies left much of it in the hands of the ‘IT experts’ who are often too busy maintaining the daily tasks with little or no time left for proactive database performance measures. Companies and its heads should realize that, while performance of database is, on one part, dependent on the physical infrastructure, the physical aspect is only a part of the story.

What Factors to Look for

Every company should seriously look for improving the performance of the database in synch with the development of the company. The goals to improve performance should be to maximize throughout by reducing query time, network traffic, CPU time, etc. through appreciation of application resource requirements, logical and physical data structures, contention tradeoffs and other considerations. The company should seriously look on several factors, which includes the demand, capability, resources, optimization, contention and satisfaction.

Database Performance database performance

Demand

Demand is the workload that the company requires. Every demand differs from company to company, depending on the company and its activities. Demand can be measured based on the activities (e.g., online transactions, web traffic, etc.) or with a certain time frame (e.g., there are less people using the system during after-office hours).

Capability

Overall capability refers to the amount of data a system can process. The capability of a system is heavily dependent on the combination of several factors. These factors include processing speed, the machine’s parallel capabilities, I/O speed, and the overall efficiency of the hardware-software tandem.

Resources

Overall capability is also partly dependent on the database system’s resources. Resources are generally defined as the software and hardware components that are available for the database system’s performance. The system resources include database kernel, cache controllers, microcode, etc.

Optimization

Optimization of database performance is another factor that defines overall performance of database systems. While all systems can be optimized, every database system can be configured differently and can be configured up to various degrees.

Contention

The fifth factor for determining performance is contention. Contention is an event that occurs when two or more workload components are using a particular database resource at odds with each other. If these contentions occur, the overall capability is compromised.

Satisfaction

The last factor to consider would be satisfaction. Basically, there would be no limit by which performance optimization can be undertaken. If the company’s resources are used at its peak and the company is satisfied with the way they are used, then the database system need no further upgrade.



Comments are closed.

Advertisements