As we read about in Part 1 of this blog entry, GalaxZ ’16 featured a panel of 3 long-time Zenoss customers: Huntington Bank (Kyle Kopp), Rackspace (David Mills), and NWN (Doug Syer).
Here is Part 2 covering the takeaways and advice from the panel participants.
You should reduce the number of tools you use – but a constant challenge for managers is fending off other tools
One of the first things that all of the managers on this panel did was to create a single pane of glass, using Zenoss Sevice Dynamics. The benefits? Increased organizational transparency, correlation between IT domains, and best of all, less finger-pointing.
Rackspace brought a staggering 13,000 processes down to a single process, while Huntington Bank decommissioned approximately 34 vendor tools, including all four Big 4 frameworks.
Regardless of the internal success and adoption of their monitoring methodology, they constantly have to defend their choice of monitoring solution and convince other groups to stick to the desired methodology.
Their tips for how to stay in control of the tool creep?
- Give your organization access to your main tool. Then, set a good example by using it and showing others how they can be more efficient with it too. They may be looking at fancy alternatives, but they may not understand that there is both a cost and a level of expertise associated with each time they use another tool.
- Show your execs that they can achieve a much better customer experience by maintaining a single monitoring platform and a single pane of glass on the backend.
- Understand the needs of your internal and external clients, and help them to see what they want to see using your tool. (But don’t help them too much and stretch your resources too thin, otherwise it will be assumed that your team has the bandwidth to take these tasks on as day-to-day responsibilities)
- Reinforce and reiterate the “right methodology” that will make them most efficient. It’s all about education.
- Explain how sucking in as much data into the main toolset as possible will provide the best insight, allowing you to most competently identify issues as well as their root cause.
Top advice for new monitoring projects
In closing, Brian Wilson asked each of the participants: “Pretend that you are once again starting from a clean slate with your monitoring strategy - what is the advice that you would give to others doing the same?” Here was their advice:
- Huntington Bank:
- Use the business case of eliminating tools to fund the switch to a new approach
- Start slow. There is no way to boil the ocean in a day or in a week. Don’t get caught up in how much is possible.
- Leverage your partners and vendors, especially to get jumpstarted so that you can see the value more quickly (leadership will appreciate that)
- Spend more time on the whiteboard focusing on the endgame rather than “the geek side” of it
- Bring in the partners that can help you to succeed at scale
- Make sure the folks focusing on the monitoring strategy are one step removed from day-to-day operations. The measurements of success for proactive and reactive job roles are very different and often at odds.
- Start with CMDB integration before taking on incident management
- Ensure that you can effectively translate engineering effort needed to support new delivery models and technologies into staffing requirements
- Convey complexity in simple terms, in a way that leadership can understand
Any thoughts on the expert advice? Write your comments below!
Please provide your email address below to subscribe and enjoy reading updates right from your inbox!