"Hello World"
As some of you might be aware, CA Performance Management is unveiling a new API strategy focused on simplicity and ease of use. In this Blog series I am going to spend some time talking about that strategy, how we are building our API to enable content creation, and how you can take advantage of a multitude of technologies freely available to divine insights into key performance characteristics of your infrastructure.
Photons and Network Performance Data
But first things first, why is this titled "First Light"? Great question! In astronomical circles "First Light" refers to that moment when a telescope gathers it's initial photons of light to present the viewer with the first astronomical image. This is usually a significant moment in the life of a telescope and I feel a connection to this moment in our forthcoming OpenAPI release. My favorite First Light image is of IC1396 included here for
your enjoyment along with a few other gems.
With our CAPM OpenAPI users will be able to easily query and export configuration and performance data for external consumption. While many tools allow this in some capacity our approach differs in that we are focusing on empowering our users to create content without requiring deep insights into our data models or database schema. We are abstracting our data models to allow for simple and intuitive queries, or as I like to say Questions.
What do I mean by questions?
Fundamentally data is collected, analyzed, stored, viewed, reported on, and alarmed on to serve a purpose, to answer a question related to the infrastructure. These questions can be as simple as "What is my throughput for interface x?" or more complex such as "which Gigabit (Gbps) interfaces in my Boston site which are experiencing less than 200 Mbps throughput (max) over the past 7 days?". Today a set of data is presented to the user to allow that question to be answered. In some areas this works well while in others it may be a bit more difficult than it should be. Users often have to run multiple reports, examine multiple dashboard views, perform external analysis in tools such as Excel or perform back of the napkin calculations. We feel this process can be easier, much easier.
Our goal is to allow those questions to be asked as easily as possible. We want to provide the answers in a meaningful way that will allow users to explore data sets to discern insights both to answer the specific question being asked but to also understand deeper relationships and patterns that ultimately will allow smarter decisions to be made to optimize infrastructure.
It's all about simplicity
You might now be thinking "sounds good but how is CAPM OpenAPI able to do this?". The answer is a bit more simple than you might imagine. We are presenting our API and the items defined within a logical schema rather than as a disparate set of query-able items. If you have leveraged APIs in the past I'm sure you are aware of the iterative nature of the request/response process. Basically you cast a wide net and then slowly refine the results set by interrogating each data set returned to build the next query to ask a question which is a bit more focused on the answer. Rinse and repeat ...
Personally I think this leaves a heck of a lot to be desired both from a simplicity perspective but also from a scale and performance perspective. Why require the user to run a gauntlet of various parsing and matching algorithms to create a simple, focused data set? Typically the answer is that it's easier to develop the API using this approach. Easier for the API creators but certainly not easier for the most important group of individuals, the consumers of the data. We want to expand the typical user profile beyond the most experienced users to those that simply have an idea on how they would like to examine or visualize a certain data set.
Or more plainly, get answers to a question.
Stay Tuned ...
Next up in the First Light series will be a deeper look into the logical schema and how complex queries can easily be created using natural relationships directly available to the user.
Oh, and before I forget, this is all possible with a single OpenAPI "query" or question.
Feel free to stop by and say hello if you are attending CA World. I will make myself available for demos and other sneak peeks into our upcoming OpenAPI functionality.
Until next time,
Jason