Presenting a Picture with Analytics

Statistics are boring. Unless you're a baseball fan or银色银, statistics are the kind of mind numbing thing that make you want to lay down and take a long nap. Statistical data isn't going to win any converts by itself. The key is making statistical data tell a story.

Analytics As Words

When we look at the raw data that we can get from a network, whether it be wired or wireless, it is far too easy to dismiss what we're seeing as a jumble of numbers being thrown our way. That's because the raw data we collect from our devices is just a set of numbers. It's a picture of a point in time that has some labels attached to it.

例如，考虑Wi-Fi热图。那些漂亮的颜色没有随机绘制地图。每个点对应于通过噪声和环境中的其他因素修改的特定信号强度。接入点无法告诉您墙在环境中的位置。所有它可以报告的是，来自该天线的信号强度在此方向下降。

The key to making sense of that data is a platform designed to collect and report on that data. Much like a poet or an author will take words and use them to paint pictures and craft narratives, the raw analytics data from the network can be used by the platform to figure out important information and present it in such a way as to help the operators and administrators make important decisions. You can build a visual idea of data over time on a heat map or you can see patterns emerging in data that would have otherwise been invisible.

画一个数据图

A great example of this came at Aruba Atmosphere 2017. Aruba's Rasa platform is a great way to show off the power of analytics and what you can do with them.

In this video, David Morton of the University of Washington shows how Rasa can power active heat maps of users on campus. The challenge for the UW team was to figure out which clients were actually in the building versus which ones were just passing by the front on their way to class. Thanks to the analytic capabilities in Rasa they were able to figure it out. They were also able to install outdoor APs that helped users associate to the network as they were walking up to the building before class.

它不仅仅是在像这样的地图上看到图片。这也是在网络中看到更大的画面，为您的用户提供最佳决策。如视频中概述，Aruba RASA团队在系统中打开了机器学习功能，了解所需的更改，以帮助优化系统。正如你可以从大卫的回答中告诉初步的想法，“不，我们不会那样做。”信任软件算法的输入可能很难通过自己的本能和培训。然而，最终，UW团队能够看到RASA建议不仅正确，而且他们还长期帮助了网络。

长期来看是truly where analytics will shine. People get promoted, move out of the organization, and generally forget why they made changes. Analytics platforms are constantly on the job. They evaluate everything and ensure that the proper changes are made to increase stability or performance. And they can be documented at the time of the change to ensure that the administrator that takes the job long after you are gone understands why a thing was done the way it was.

Analytics and statistics don't have to be boring. Instead of focusing on the minutia of the data, you need to step back and let a platform like Rasa show you the big picture. That kind of visibility helps you better understand how you can use the data to make you network better in the future.