A system administrator’s roles and responsibilities span various dimensions of an IT organization. As a result, keeping tabs on what’s going on in the world of technology, including vendors and their products, latest product releases, end-user experiences, and troubleshooting performance issues are just some of the areas of focus. Over time, system administrators turn into thought leaders due to the technology, industry, and domain experience they gain and use. They pass on their knowledge to colleagues and technology aficionados. Even organizations turn to such experts to hear what they have to say about where IT is headed.
On that note, we at SolarWinds® are glad to have brought together IT gurus, geeks, and fellow system administrators to share their thoughts on system and application performance management. This event took place recently in the form of a #syschat on twitter. For those who didn’t get a chance to tune in, here are some highlights:
Application monitoring: Generally, there is a consensus that application downtime affects business performance. Given that businesses are paranoid about this, why hasn’t the adoption of application monitoring in some organizations taken off like it should? Experts like @standaloneSA and @LawrenceGarvin feel that, “Some of it has to do with need.” Or, as @patcable points out, “Admins don’t know what to monitor, and apps don’t provide the right data.” This is true for various reasons. Often, IT pros are given a mandate by business groups saying that all apps are critical. Therefore, they have to watch apps closely for performance issues. Before answering the “what to monitor” question, IT pros need to ask, “Why should I monitor these apps, are they really that critical?” Knowing the answer to this question eliminates the additional noise, and you can focus only on what to do with the really critical apps and ensure that you’re monitoring the right metrics.
Apps in the cloud: Monitoring the performance of apps in the cloud is, again, not a direct solution to solving a performance problem that can arise from your apps running in the cloud. As more applications are being deployed in the cloud, the level of difficulty in monitoring those apps gets higher. IT pros have to really get down to understanding the “how,” which takes time. For example, @vitroth said, “Ops finds it hard to monitor what engineering doesn't instrument. Give me counters, categories and severities!” When IT pros have difficulties managing apps running on a physical server, the cloud layer is certainly going to be an unfamiliar place, and new complications will arise.
Skills sets for SysAdmins: A lot of buzz is going around about whether SysAdmins will need to have coding skills one day. It may not be mandatory for IT pros to have programing skills, but they might want to develop these skills so they can create and automate tools. While this was only one opinion, others like @patcable suggested that “sysadmins are going to have to become more comfortable writing stuff in some language other than shell.” Learning and understanding your IT infrastructure and environment are essential. IT pros should be willing to learn and learn quickly because ‘things aren’t slowing down.’ Where gaining technical knowledge and skills is concerned, it always helps to “learn a programming language, version control w/git, config management, and keep an eye on Hadoop,” as recommended by @standaloneSA.
What are your thoughts on these topics? Where are you doing with application monitoring in your organization? What difficulties do you see with monitoring apps in the cloud? Do you see DevOps improving the adoption of application monitoring? We’re happy to hear your views and opinions. Follow us on @SWI_Systems to learn more.