性能数据收集和可视化工具
我有一个性能测试套件,它在运行期间收集响应时间信息(以 CSV 格式)。此外,我还有一组监视脚本,用于收集服务器上正在运行的应用程序指标(也是 CSV 格式)。我想自动化这些数据集(基本上是时间序列)的可视化,例如一些网络应用程序,我可以在其中上传所有数据集并将它们显示在一个漂亮的图表中,可以按时间段过滤,在不同的切片中显示数据并与其他基于时间的数据集。你们知道类似的事情吗?
I have a performance test suite that during the run collects response time information (in a CSV format). Also I have a set of monitoring scripts that collect running application metrics (also in CSV format) on the server. I would like to automate a visualisation of those datasets (basically, time series) e.g. some web application where I can upload all datasets and get them displayed in a nice chart with posbility of filtering by time period, displaying data in different slices and correlating with other datasets based on time. Do you guys know of something like that?
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
用于此目的的一个特别流行的工具是 rrdtool。
如果您对最复杂的时间序列分析工具感兴趣,我建议您使用 R,并且对于 Web 界面,RApache 或 RStudio处于服务器模式。如果您主要对仪表板式工具感兴趣,我怀疑这有点过分了。
A particularly popular tool for just this purpose is rrdtool.
If you are interested in the most sophisticated analytical tools available for time series, I'd recommend R, and for a web interface, either RApache or RStudio in server mode. I suspect that that would be overkill, if you're primarily interested in a dashboard-style tool.
另一种选择是名为 KNIME 的工具,它是一个开源数据集成、处理、分析和探索平台。它有一堆时间序列节点用于处理和分析时间序列中的数据,以及用于读取和写入不同数据存储的节点。
可能对您的问题来说太过分了,但对于需要分析时间序列数据的其他人来说可能很有用。
编辑:我忘了提及,KNIME 还集成了 Java、Python 和 R 脚本节点,因此,如果您找不到使用其本机节点之一执行某些操作的方法,您可以滚动自己的脚本来完成工作。
Another option would be a tool called KNIME, an open-source data integration, processing, analysis, and exploration platform. It has a bunch of time series nodes for processing and analyzing data in a time series, as well as nodes for reading and writing to different data stores.
Probably overkill for your problem, but may be useful for someone else who needs to analyze data in a time series.
EDIT: I forgot to mention, that KNIME also has integrated Java, Python and R scripting nodes, so if you can't find a way to do something with one of its native nodes, you can roll your own script to get the job done.