Disclaimer: I recently attended Tech Field Day 19. My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event. Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.
Operations, And No Operators
VMware has a pretty comprehensive suite of management tools that you can use to manage your VMware cloud products, including:
- vRealize Automation
- vRealize Operations and Network Insight
- CloudHealth, Operations SaaS, Wavefront, Log Intelligence, Network Insight
One of the keys to building successful management and monitoring tools is delivering the ability to perform activities in an autonomous fashion. To wit, there are parts of your infrastructure that you want to be “self-driving”. Taruna Gandhi talked about the “4 Tenets of self-driving operations”. These are:
- Continuous Performance Optimisation – Assure application performance with atomic workload placement and balancing workloads based on business and operational intent
- Efficient Capacity Management – Run infrastructure like a public cloud – optimal densification, proactive planning, and procurement
- Intelligent Remediation – Predict, prevent, and troubleshoot across SDDC and multiple clouds, from apps to infrastructure
- Integrated Compliance – Reduce risk and enforce IT and regulatory standards with integrated compliance and automated remediation
The idea behind tools like vRealize Operations is that you can run your VMware-based infrastructure in an autonomous fashion.
It’s A Small Thing, But It’s Really Quite Useful
One small thing that VMware bought up was the ability to use tags for licensing enforcement and VM placement using DRS. You can read about how to do that here. I think the capability was first introduced in vROps 6.7. Why would you need to move workloads around for licensing enforcement? Just five years ago I was working with enterprise environments that had to have limited amounts of CPU sockets exposed to various operating systems (when virtualised) or line of business applications. The way to combat the requirement was to deploy dedicated clusters of compute for particular software packages. Which is pretty stupid when it comes to getting value from virtualisation. Nowadays the cluster is no longer the barrier to VM mobility, so you can move workloads around in an easier fashion. The general feeling on the Internet might be that the likes of Microsoft and Oracle have made these kinds of workarounds harder to do (and stay compliant), but there are still plenty of smaller software vendors that have odd requirements when it comes to the number of sockets consumed in virtual environments. Being able to leverage tags shounds like just the sort of thing that we’ve talked about for years in terms of operational overheads that shouldn’t be overheads. It strikes me as something that many enterprise customers could be interested in. As VMware pointed out though, some of the enterprises needing this capability ironically may not have upgraded yet to the required version yet.
Thoughts and Further Reading
I’m the first to admit that I haven’t spent nearly enough time keeping up to date on what VMware’s been delivering with the vRealize Operations product. I used it early on and then moved into roles where it became someone else’s problem. So it was nice to get up to speed on some of the capabilities they’ve added to the product in the past few years. It’s my opinion that if you don’t have to do certain operations in your environment, that’s a good thing. Infrastructure operations is a hectic business at the best of times, and the requirement to intervene in a manual way is not just potentially a burden on your workforce (particularly when something goes awry at 3 in the morning), it’s also an opportunity for other things to go wrong. The good thing about automating the management of infrastructure is that things get done in a consistent fashion. And there are, generally speaking, fewer opportunities for human error to creep in. This does require a certain amount of intelligence to be built into the platform, but VMware seem to have a pretty good grasp of what’s happening in the average vSphere environment, and they’ve coupled this with many years of field experience to build a platform that can get you out of a spot before you get in one.
vRealize Operations is more than just a glorified dashboard application with some cool traffic lights that keep management happy. If you’re running any type of reasonably sized virtual infrastructure, and you’re not leveraging vROps, I think you’re making things unnecessarily difficult for your operational staff. Obviously, vROps isn’t some silver bullet when it comes to IT operations, but it has a lot of power under the hood, and I think there’s some great potential that can be leveraged in the platform. You still need people to do stuff, but with tools like this you won’t need them to do quite as much of that tedious stuff. I’d also recommend you check out the other parts of VMware’s presentation at Tech Field Day 19, because they covered a lot of really cool stuff in terms of their vision for cloud management tools.