Next step in cloud Application Performance Management (APM)

Michael Kopp is a technical analyst at compu-Ware's APM center excellence. He has worked as a designer and developer in the Java/Jee field for more than 10 years. In addition, Kopp also specializes in the structure and performance of virtual and cloud large-scale production deployment.

?

The new APM emphasizes the cost of each transaction and helps testers increase their strategic role in maximizing investment

When an enterprise enters the cloud, it will produce many benefits, including more flexible business and significant cost savings. Of course, it can also increase profits. Continuous applicationsPerformance management(APM) supports the entire application lifecycle. In many aspects, it can be the key to achieving these benefits. Recently, the importance of APM in the cloud has attracted much attention, especially when it involves ensuring fast and reliable end-user application experience. The cloud is not transparent, that is,Cloud applicationsEnterprises often have no in-depth insight into their internal work and the capacity management decisions of the cloud service providers they choose. There is only one way for enterprises to ensure that end users are using cloud-based applications for a reliable experience. Testers need to estimate performance from the perspective of real end users on the other side of the cloud. Because they are on the "Front Line", The tester is the first to make a decision when performance problems occur in cloud-based applications. In-depth diagnostics can help determine the cause of performance issues, whether it's cloud, your data center, or another element in the application delivery chain. If the cause is cloud, enterprises can use this useful information to warn cloud service providers and maintain SLAs ).

Many companies and their testers have made significant improvements in using APM to ensure high levels of Member application performance. Their applications may be trustworthy enough, but there is no time to stop doing APM in the cloud to better understand the internal work of cloud-based applications, which can be of great benefit. The next step of APM in the cloud is to optimize the cost structure of cloud-based applications, just as application performance has a direct impact on cloud ROI. Once again, the tester has a strategic opportunity to help maximize cloud investment and raise the bottom line.

In other words, cloud is concerned that APM not only makes applications faster, but also makes applications, especially the most productive applications, more cost-effective. Every step of the application development process must be taken into account, because even the so-called best application is definitely not perfect. Continuous improvement, correction, and optimization can greatly affect the performance and cost effectiveness of a cloud-based application. Therefore, regression testing and the entire life cycle of shadow require both considerations.

In addition to the quick search function, the tester also needs to optimize it so that it can provide better results without being executed five or more times by each tester. This can be regarded as function optimization or execution cost reduction, because in the cloud, each transaction is worth one dollar. Each transaction reduces database access calls and does not accelerate the search at all, saving costs. This is because most cloud providers charge for the number of executions, such as Sol. So optimizing the number of sol may be more cost-effective than saving the CPU! In this way, the company completely bypasses the problem of resource optimization and directly hits the cost optimization problem. The tester can really control the price. In fact, enterprises do not need to care about the use of public cloud resources at all. They need to care about real end-user application SLA and cost effectiveness-more and more testers are contractors of this knowledge.

Another example is the purchase function. Does an end user spend a lot of time on the cloud and consume many resources? For example, product introduction or production image? By understanding how end users use their time, the tester has a deep understanding of what can increase prices in the cloud. In addition, by understanding the cost structure of a transaction and how much it will benefit, the tester can better set the priority so that the functions of the previous transaction can be changed, and the cost and performance can be optimized.

For many enterprises, the main obvious advantage of migrating to the cloud is variable, which avoids the old method of capacity planning and large upfront payment: otherwise, enterprises can, as their load increases, they can expand their environment scale. However, variability also has its drawbacks. It is easy to cause excessive consumption of the planned capacity, because there is no hard limit, this will exceed the cost estimation.

Therefore, it is important for testers to directly understand how end users interact with applications and how applications handle loads. In this way, testers can provide valuable information that can help guide more informed cloud capacity decisions. Without this information, execution is blind.

In short, the end user experience management enables testers to understand how the behavior and performance of end users affect the conversion rate and business. But on a public cloud, this is only part of APM. Only when enterprises can continue to reduce their costs and meet end users' expectations for fast and trusted applications can they succeed in the cloud and use the cloud to improve their business performance. By focusing on the cost of each transaction, the tester can make a greater strategic contribution to the company's profitability.

Copyright: This article from spasvo zezhong Software Testing Network: http://www.spasvo.com/news/html/2014922144040.html

For original works, you must use hyperlinks to indicate the original source, author information, and statement of this Article. Otherwise, you will be held legally liable.

Next step in cloud Application Performance Management (APM)