Technical support is very good.
The solution is very good as an infrastructure as a service.
The product offers very good visibility.
The solution is good for cloud cost management.
Technical support is very good.
The solution is very good as an infrastructure as a service.
The product offers very good visibility.
The solution is good for cloud cost management.
CloudHealth needs to start building out Turbonomic-types of features that help the customers who are using CloudHealth really understand everything down to the server level, the virtual machine level. They need to be able to illustrate the performance of the machine. As a customer, I want to be able to log into one tool and be able to look all the way down to a virtual machine that's running in Azure or AWS and see, is it running at 80% utilization or 20%? If it's only running 20% utilization, well, maybe I should move that workload off of that machine and combine it with another machine and remove that machine, therefore lowering my overall Azure spend as I've just eliminated a server. However, I need that insight in order to make that call.
Oftentimes, when you look at multiple application layers, if you can identify six or eight or ten different servers that are just underutilized and can eliminate them from your fleet, well, you just dropped your Azure bill by a thousand dollars a week. The savings they could offer clients could be quite substantial. However, CloudHealth doesn't have that capability. They can't reach down to that level right now. That's why you've got to bring in Turbonomic. They can reach in at that virtual machine level to tell me if the machine is being truly utilized. I'd be able to understand if I should go ahead and eliminate it from my fleet.
If I took the remediation recommendations from Turbonomic, and I implemented that into my Azure environment, and then came back and analyzed it from CloudHealth, well, now I could see that that overall Azure spend has dropped by 10%. However, ideally, CloudHealth should have those tools built-in just to help me optimize my fleet on the fly. And then I should see that cost justification.
CloudHealth could expand into is the SaaS side of it. Right now, CloudHealth is primarily focused on the infrastructure as a service players and is helping to analyze, how the infrastructure as a service virtual environment maps to the customers, to the needs. The opportunity that CloudHealth has is if you're talking to somebody about their public cloud spend, oftentimes the customer thinks of that as not just the infrastructure as a service, but they also look at that from a software as a service. One of the biggest challenges that we also see in the enterprise accounts that we work with is the SaaS sprawl is almost out of control. It's as bad as the infrastructure as a service sprawl, where people are given the credentials to their Azure tenant, and now they're spitting out virtual machines for all kinds of stuff.
Unless you really have better visibility and control in that, you just don't know what's going on until you get the bill. The same problem is happening in the SaaS side where, if you have 500 employees, traditionally that company goes out and buys Office 365 times 500, and they just give everybody the same thing. At the end of the day, maybe everybody doesn't need an Office 365 E5 plan. Maybe a third of the company needs E5, a third could get by with E3 and a third could get by with E1. If you optimize that Office 365 footprint, along with all of the other SaaS applications that you may be attaching to that Office 365, like Trend Micro for security, or Metallic for backup, there is more money to be saved.
Right now, CloudHealth doesn't give me any visibility into that SaaS side of the model. They give me a view into the infrastructure as a service side and the cost analytics and reporting. If I could use that one toolset that not only helps me optimize my infrastructure as a service and also have visibility across my SaaS environment, to me that would be the ideal situation. That way, I've got one platform, one toolset that helps me manage my cloud spend across SaaS and IaaS. Right now, for the tools that are out there, it's either IaaS or SaaS. And there's not a lot of players in the SaaS management space at this point, there are only four or five companies that I'm aware of.
We became a reseller about a year or so ago. We've been dealing with it for a while.
We've actually had to engage with technical support a few different times. Whether it's just configuring customer accounts correctly or troubleshooting, it's always been very good. There was a little bit of a hiccup during the transition when VMware was acquiring CloudHealth where I think there were just some of the procedural and operational processes where we had some account managers change.
When we reached out to the technical account manager that we were assigned at the time, he had been shuffled around a bit. We were routed from one person to another and had to keep getting people up to speed. However, that got ironed out.
The ones that I have been personally using are CloudHealth and CloudCheckr, however, those are more around the cloud cost analytics, the cloud cost management. Those companies are starting to, change a bit. You've got Turbonomic and companies like that, that are coming from the on-prem asset level up, where you would use Turbonomics to crawl internal VMware infrastructure types of environments and tag specific applications. Let's say that I'm trying to think of how I'm going to migrate my marketing application from my on-prem environment to the cloud. That marketing application is built at three tiers. You've got a web tier, an application tier, and a database tier. That's multiple servers, multiple layers of applications. How do you analyze what that footprint looks like so I can make a fair approach to what I'm going to need in an Azure or an AWS cloud environment to move that application?
Turbonomic has approached it from that perspective where they come in and they ask questions such as how many IOPS is that application consuming? What's the performance of the servers at a CPU and storage and a memory level, so that they can actually map that into equivalent Amazon EC2 servers or Azure servers of sizes of X, Y, or Z.
That's where Turbonomic has really approached it from that kind of foundational infrastructure up whereas a CloudHealth or a CloudCheckr have approached it more from an application already being in the cloud and drilling down to costs.
Now we're trying to figure out how to optimize that footprint. What I'm seeing is both those types of companies are starting to meet in the middle where Turbonomic is now starting to add features around cloud cost management and CloudHealth is starting to get down into the Turbonomic side of assessment of what you're trying to move to the cloud.
I can start to see the melding of that space over time, where there are more and more of these on-prem application performance management APS-type of applications. They are starting to bleed up into the cloud and the cloud and cloud applications are starting to bleed down that on-prem world.
We're a reseller. There are two different models. We're a reseller of CloudHealth. We're a reseller of Dell. We're a reseller of VMware. We also use those products internally.
We use it daily. It's one of our core tools within our Zones Cloud Platforms tool bag, when we're looking at how we help customers optimize their cloud spend across their Azure and AWS environments today.
It's a cloud-native application.
In terms of CloudHealth, I have not dug underneath the covers as to where do they actually host it. CloudHealth just got bought by VMware about six or eight months ago, or maybe almost a year ago now. I'm pretty sure that they built on a public cloud. I'm just not sure which public cloud they actually built it on.
I would advise new users to get a demo, take a look at it, dig into some of the analytics and the reporting. It's really one of those tools that you'd need to have hands-on time with. You can go through the presentations, you can look at the high level, but you really need to have a public account, an AWS account, a Google account, an Amazon account, and then tie that to your CloudHealth environment during the demo, or during your proof of concept. When you test, you can make changes in your environment and see how it replicates and how CloudHealth presents that data back to you. That real-time reviewing and seeing the cause and the effect is what really helps you understand what the platform can do. And it helps you have a better understanding of its features and capabilities and how in-depth the reporting and the analytics are compared to a CloudCheckr or other third-party applications that are out there.
The last piece I'd also throw out there is that one of the biggest challenges I think that people are having right now, when it comes to a CloudHealth or a cloud cost analytics product, is that most of the independents have been acquired. There was a product called Cloudyn. That got bought by Microsoft. Now Cloudyn's baked into Azure in what they call Azure Cost Management. Google's acquired a couple of companies over the years, and they have their Google Cloud Cost Management and multi-cloud orchestration and all this fun stuff. CloudHealth got bought by VMware. CloudCheckr is one of the only ones that's still independent in that cloud cost management space. Outside of that, there's some smaller niche players, however, it's getting hard to find an independent. You pretty much have to go and buy one of these products from VMware.
I'd rate the solution at an eight out of ten.
If they can start to improve how they work across other SaaS applications - not just infrastructure as a service - they would get higher marks. If I rated them just from an infrastructure as a service play, I'd probably put them at a nine or a 9.5 simply due to the fact that they do a really, really good job in that space. However, it's a bit of a niche. It's an important niche and it really is helpful in that specific spot, however, it leaves me having to want to find another tool to do the other half of the job and where ideally if it was one tool, I'd take that eight up to a nine and a half. As it is, right now, I rate it as an eight as it doesn't give me all of that features. I have to use multiple tools to get the end result I'm looking for.