What is our primary use case?
We're a cloud service provider, so we wanted a solution that would proactively be able to notify us of potential issues. We have four core cloud platforms that we provide. We wanted something that could look at both the network connectivity as well as the infrastructure and storage layers.
How has it helped my organization?
As a cloud provider, it's of paramount importance that we're connected to the internet and the cloud in general. If the data collectors ever go offline, there's an alert that's actually sent out to us. Because of the alert tiering, we can have it so that that's treated as an emergency alert and it goes to a different set of critical recipients. We've had it where it's assisted us when we were having issues with one of our IP transit providers, and we were able to use the logs that it provided to demonstrate that we had a definitive issue with the provider and their connectivity. That actually enabled us to push back on the IP transit provider and get quite a substantial claim approved because we were able to demonstrate how unstable the link was.
The other element that it's helped us with is in predicting the future. And another thing that it allows us to do very easily is to track our bandwidth usage that's going in and out of each of the data centers. We've been able to use that information to trend and predict when we need to get upgrades in place. Funny enough, we have an order now where we're increasing our connectivity at one of our data centers tenfold and that's being driven because Auvik's enabled us to understand that we're rapidly approaching our threshold.
The mapping tool does make it easy and convenient to access and get console-level access quickly and easily because of the way it works within Auvik, it embeds the credentials. It's a couple of clicks of the mouse button and you're on a console session. You don't have to go through that rigmarole of what's the IP address, how do I connect, what do I use PuTTy, do I use Telnet? What are the credentials? With Auvik it's very streamlined, click, point, click and you're on.
We've saved on intangible costs. The overhead of managing three different open-source platforms has now completely gone. We just have a SaaS platform, we pay our fee, and it does exactly what it says it will do.
It carries a high value ratio on time-to-value. The interesting thing with the price model is that that value ratio could change. It depends really on if you have a hundred switches, it's going to be a lot more expensive than a client that's only got two large switches. But for us, we find it's very high value for money and good value for money.
What is most valuable?
The auto-discovery and the mapping are quite nice. We can see how our data centers are connected. That was one of the immediate appeals.
The change control that's built into it for picking up network device changes and recording is something else that we found to be extremely useful.
It's extremely easy to use, although sometimes some features can be a little bit hidden. You have to know where to look, but generally nine times out of ten, it's very straightforward and quite intuitive.
Network discovery is very good. Like anything that does auto-discovery, it can at times get confused, but it's very easy to select to do an override. If it mis-detects a firewall as a generic network device, it's very easy to correct that on a manual basis. But that happens quite infrequently.
It automatically updates our network topology. We're quite lucky we don't have too many issues. It has given the guys on the desk confidence that they can see very quickly and access any system that we've got monitored. In the early days, we had a hesitancy to know if we could rely on Auvik, but over the last couple of years, it's proven itself time and time again. If it tells us there's an issue, we trust that.
In certain circumstances, it has decreased our mean time to resolution. The bulk of our issues unfortunately tend to be more of an application layer, which Auvik doesn't have visibility into.
Auvik enabled us to consolidate or replace other tools. In the early days, we used to use a SaaS platform called LogicMonitor, which we then reverted from and pulled to an in-house solution. That ultimately became three open-source in-house solutions. It was at the point that we wanted to look at something that could consolidate and give us more intelligence and that's where Auvik came into play.
What needs improvement?
We use network mapping slightly differently from a lot of MSPs who are more focused on using Auvik to maintain end-user environments. We're looking at it maintaining quite a complex data center environment. The mapping is good, but that can mean that it can get a little bit unwieldily. So having the ability to be able to have more manual control on how the map is organized, would be really useful for us.
It uses SNMP in its discovery process and how it pulls in data. But today it doesn't have an SNMP trap facility so you can't have your infrastructure devices push alerts into Auvik. And that for us would be a big feature that we would like to see.
The single sign-on piece that they have is really good. That works really well for us. Everything else we're really happy with. They have the chain of control stuff and configuration management piece, which was really nice to discover. We never knew about that. That was one of those things that we fell across and then make use of that quite extensively.
For how long have I used the solution?
We've been an Auvik customer for approximately two to two and a half years.
What do I think about the stability of the solution?
We haven't had any issues with availability. They do regular maintenance, but we always get proactively notified of it and it's never caused us an issue.
What do I think about the scalability of the solution?
We use it to monitor two data center sites that have somewhere in the region of about 300 to 500 infrastructure devices in each and we never had a problem with it. My understanding is that if we went out to 20 data centers, it would scale without any issues.
It requires zero maintenance. We would have to do regular patch management with our on-prem solution. It wouldn't take up a huge amount of time, but it was something that had to be scheduled on a monthly or bi-monthly basis.
How are customer service and technical support?
Technical support is extremely good. Any ticket that we've raised, whether it's a query or we feel that we've hit a bug, has been responded to promptly. They have an extensive knowledge base set of articles, which are invaluable for pointing you in the right direction.
Which solution did I use previously and why did I switch?
We saw Auvik through one of our partners. We sell to IT resellers, and it was seeing the ease that they could actually access some of the information for a shared client that put us onto doing the evaluation. The one-week evaluation turned into a purchase.
How was the initial setup?
The initial setup was very easy. We downloaded an OVF file, deployed, and connected. We had someone from their tech teams, once we had done the deployments, work with our service desk team and work through doing the initial config.
We had the collectors deployed in under a couple of hours and the configuration for each data center to set things up took us a week for each data center. That was a process that was hindered by us because we had to tweak and tune things to meet our requirements.
Compared to LogicMonitor, my experience was pretty much on par. The SaaS providers tend to have quite a streamlined model. You deploy a data collector, which they have as a single download, and then it starts to consume data into the SaaS platform. For SaaS to SaaS, it was pretty much the same. When you're doing your own on-prem deployment it's vastly more efficient.
What's my experience with pricing, setup cost, and licensing?
We think the pricing is actually really cool. Only certain network devices make the pricing really cost-effective for us. We can monitor 50 servers and essentially one server or 50 servers has no impact on costs. The one thing I think that's crucial is just to make sure that you understand how many billable network devices you have in your estate before you move forward.
Typically, in our environment, VM hosts, storage arrays, virtual machines, or physical like Windows or Linux machines, all have no impact on cost. The only things that really impact costs are our network switches and our firewalls.
What other advice do I have?
My advice would be to verify that it can cover all the devices that you want it to monitor. For us, it does virtually everything that we need and the odd exception hasn't caused us any major problems. We're still able to do basic monitoring. We just can't sometimes get the level of detail that we want. Go back to the environment and make sure that you understand your network and your network devices so that you can make sure that it's going to give you the value that you want.
The biggest lesson we've learned from Auvik is that we had an assumption that because it talks to the devices and discovers them using SNMP as one of its main mechanisms, we assumed that it would do this SNMP trap feature. We were surprised that it didn't. It hasn't caused us any major issues, but we do welcome the day that that's actually added as a feature.
I would rate it an eight out of ten. Not a ten because of the lack of the SNMP trapping and the fact that it's got a lot of flexibility on the devices it monitors, but there are a couple of holes. It's not a big issue for us.
Which deployment model are you using for this solution?