8-steps metrics program - an introduction to metrics

转载 2007年09月26日 16:12:00
8-steps metrics program - an introduction to metrics

What Are Software Metrics?
Software metrics are measures that are used to quantify software, software development resources, and/or the software development process. This includes items which are directly measurable, such as lines of code, as well as items which are calculated from measurements, such as earned value.

Everyone who develops software uses some kind of software metrics. However, when asked what software metrics are, the tendency is to restrict the response to software size measurements, such as lines of code or function points. In reality, software metrics include much more than primitive measures of program size.

Software metrics include calculations based on measurements of any or all components of software development. For example, consider the system integrator who wishes to determine the status of a project’s test phase. He or she will undoubtedly ask for information on the proportion of tests that have been executed, the proportion that were executed successfully, and the number of defects identified. These measures are all examples of primitive - yet useful - software metrics.

Consider the engineer who is responsible for improving the performance of a software product. He or she will consider items such as memory utilization, I/O rates, and the relative complexity of software components. These are also examples of software metrics.

There is nothing overly complicated about software metrics. In fact, the biggest challenge in establishing an effective metrics program has nothing to do with the formulas, statistics, and complex analysis that are often associated with metrics. Rather, the difficulty lies in determining which metrics are valuable to the company, and which procedures are most efficient for collecting and using these metrics.

The SPC conducted research with a number of software developers and managers, many from well-established companies -- few of whom had much experience with software metrics. A common assumption emerged: software metrics are used simply to measure programmer productivity against an industry standard.

Apart from providing little value, such comparisons are apt to foster resentment among the programming staff, and earn little support for a metrics program.

A healthy metrics program focuses on much more than the measurement of programmer productivity. Consider these areas of software development which can benefit from a well-planned metrics program:

  • project management
  • product quality
  • product performance
  • development process
  • cost and schedule estimation

The key to the effective use of software metrics within an organization is to prepare a plan describing how metrics will be used to meet strategic management goals.

This manual is a guide to devising and implementing such a plan. 

Next: A Metrics Program in 8 Easy Steps

举报

相关文章推荐

An Introduction to Interactive Programming in Python (Part 2) -- week 8

Sets — Groups of sprites Animation — Sprite animation

11. When to change dev/test sets and metrics 何时更改开发/测试集和评估指标(《MACHINE LEARNING YEARNING》翻译)

何时更改开发/测试集和评估指标当开始一个新项目时,我一般会试图快速选择一个 开发/测试集 ,因为这可以给团队制定一个明确的目标。我通常会要求我的团队在不到一周之内想出一个初始的开发/测试集和评估指标。...

我是如何成为一名python大咖的?

人生苦短,都说必须python,那么我分享下我是如何从小白成为Python资深开发者的吧。2014年我大学刚毕业..

K8s 多自定义计量(custom metrics)

这里介绍下如何在K8s设置多个自定义计量。K8s 默认的可用于PHA(Pod Horizon Autoscaling)只有CPU,自定义的计量feature一直处于alpha 测试阶段,而且不稳定,在...
返回顶部
收藏助手
不良信息举报
您举报文章:深度学习:神经网络中的前向传播和反向传播算法推导
举报原因:
原因补充:

(最多只允许输入30个字)