Stackify - application monitoring for DevOps

Stackify's CEO and VP of Marketing discuss what monitoring/management tools developers/operators need. In short, simplicity and the ability to dive deep into the software stack when needed.
Written by Dan Kusnetzky, Contributor

Stackify's CEO, Matt Watson, and VP of Marketing, Irit Gillatch, stopped by to introduce their company and demonstrate a tool they think would be the perfect aide for developers/operators. 

What are the needs of DevOps?

Stackify believes that developers and operators need related, but different things. Developers need to be able to quickly identify the root cause of performance issues and, when needed, go deeply into the stack of software supporting the application. Operators also need to be able to quickly identify performance issues, but may not need to go spelunking in the software caves underneath the application.

The key, they believe, is knowing what data is likely to be needed, gathering that data, organizing that data into useful, but cross-silo information, and making it easy for staff to view. Stackify believes that DevOps staff face significant challenges using today's tools including the following:

  • Organizations have acquired separate tools to monitor systems, networking, storage, database engines, application frameworks.
  • Each tool is designed to focus on one area and offer insight into what is happening there. Getting an overview can be a challenge.
  • DevOps need a tool to rapidly get the overview and yet drill down to get details when needed. Being forced to wade through the details to get an overview is counterproductive.

Stackify's tool

Stackify is offering a tool that combines functions for monitoring servers, applications, security and access. This tool offers a quick view into application metrics and performance, error tracking and even log file management.

The tool Stackify offers isn't trying to do everything for everyone, everywhere, always. It is focused on the needs of DevOps and makes it possible to quickly see that problems exist and make it possible notify the proper person or dive down into operational data for elements of the software stack supporting the application.

It offers application and server performance information including, monitoring performance of applications, servers, databases, webpages, and can even track custom metrics. Customers can manage log and error files, aggregate them and allow quite a number of useful functions including search through the logs, correlating log information from several  logs, drill in, see what logs are effected when an error occurs, see all the process and everything that happened when error occurred etc). What was really interesting to me was how simple the incident escalation and notification processes are. It appears that it would be quite straightforward for DevOps to manage issues.

Other components of the overall solution are likely to be monitored in future releases of the tool.


I'm reminded of the wonderful story of the blind men and the elephant. The men can't see the whole elephant. Each of them, however, thinks that they know what the elephant looks like by touch. One thinks the elephant is like a hose after touching its trunk. Another thinks it is like a snake, after touching its tail. Another thinks the elephant is like a tree, after touching one of the elephant's legs. While not technically wrong, none of them really understand the elephant.

During the past year, I've spoken with at least 20 different suppliers of monitoring tools for systems, networks, databases, applications, application frameworks, hypervisors and cloud frameworks. Each of them strongly believes that the answer to an organization's problems can be found in the area their product examines. Often they denigrate other products because they don't delve as deeply into their chosen area. Seldom do they really try to provide an overview making it possible for staff to find the root cause of a performance anomaly.

A few suppliers, such as Stackify and Nastel, are doing their best to offer a useful overview. Stackify's dashboard is designed to be easy to use, simple and give DevOps staff a way to look at the history of errors found in log files to quickly gain an understanding of what's happening and what's going wrong. Nastel feeds that sort of data into their complex event processing engine to get that understanding to another level.

I can recommend both of these tools for different reasons.

Editorial standards