One of our clients asks for the “entry fee” for getting into the IT management platforms I’ve been talking about in the blog over the past year. The list is by no means exhaustive, just off the top of my head. In fact, if you’d like to see other folks listed, I’d love to update the table: add a comment below with the info or just who it is and I’ll look into them and update the table.
That disclaimer and invitation to help update the list out of the way: below is the price breakout for platforms that have public pricing. Instead of listing it per month, I extended it out to a year. There are several other notes and clarifications, esp. around what each vendor or group considers a “unit”; I’ll add notes for that below the table.
Pricing
| Product | Min. Price/year | Price/unit | Min. # of devices |
|---|---|---|---|
| GroundWork | $8,000.00 | $160.00 | 50 |
| Hyperic | $120.00 | $120.00 | 1 (socket) |
| Klir | $1,296.00 | $18.51 | 70 |
| OpenNMS Group | $4,995.00 | ||
| Versiera | $119.88 | $9.99 | 1 (host/server) |
| Zenoss | $5,000.00 | $100.00 | 50 |
Notes
- The above is the minimum price to start a year long subscription to for support and/or use (in the case of hosted platforms and/or closed source components) for one of the platforms above. The open source platforms can, obviously, be used at no charge if you don’t want support. Each vendor (apologies to the OpenNMS Group as they’re not a “vendor”) has many other service levels and services you can buy, like help with the initial setup. Also, of course, as you buy support and/or use for more “units,” you’ll may get price-breaks. See the linked to pages for more detail.
- Typically, the support/use you buy at the above prices is for business hours. 24/7 support requires one of the higher level plans for most vendors. See the linked to pages for more detail.
- Most of the vendors are providing support for open source software. What you’re buying is support per server, device, interface, or “unit” monitored and managed.
- How each vendor classifies the “unit” being monitored and managed varies. Some charge per server (IP addressable device) like Versiera, while others charge per socket (Hyperic), and others charge per interface (like OpenNMS). Check he linked to pages for more details on each.
- Most of the vendors require a minimum number of “units.” This is either phrased as, for example, a “minimum of 50” (Zenoss) or “up to 2,000” (OpenNMS Group). Either way, aside from Hyperic and Versiera, you can’t but just one.
- FiveRuns and Qlusters/openQRM aren’t included as their prices aren’t public. SpiceWorks isn’t listed as they’re free.
- I didn’t include log management vendors like LogLogic and Splunk as I see them as more additive than the above. What do you think?
Again, if I’ve excluded someone or made an error in the table above, please tell me and I’ll update the above table.
Disclaimer: GroundWork, Zenoss, SpiceWorks, Splunk, and LogLogic are clients.
Technorati Tags: pricing, redmonkclients, systemsmanagement2.0
On Klir, a couple of clarifications:
1) They're a SaaS solution vs inhouse install.
2) they have a free version that supports, I believe, 25 devices. If that's all you need, it's free forever (not a trial in other words).
Disclosure: I have consulted in the past for Klir and consider a couple of the execs friends.
Excellent point, Rick, and thanks for adding it. Versiera has a similar free option.
Interesting post, Michael. Hey — what do you make of all of the other projects out there (Cacti, NeDi, RRDTool, MRTG, Ganglia, etc.) for monitoring / mapping network performance? Seems like there are some vendors that use MRTG / RRDTool (even big 4 vendors) in their solution … then there are users out there building their own home grown monitoring systems with these, then writing their own front end for graphing, et al. Do you see these projects gathering any momentum out there? It would be cool if you guys did some sort of report in the future about all the other cool open source network monitoring tools out there and pointed to some of the ones with momentum. Most folks know about Nagios by now, but seems like a lot of those other ones are relatively obscure (despite the fact that some of them have good adoption).
Hey Michael,
For the GroundWork pricing, the $8,000 annual number is for up to 50 devices monitored. It's $16,000 annual for moniotring an unlimited number of devices (limited only by the performance capabilities of the monitoring server).
Tony
Travis: yes, people use that stuff. Quite a few of projects use RRDTool. A little under a year ago I visited a local data-center that duct-taped together their own setup just like you're describing. I was most impressed by the Cacti integration/customization they'd done. And, of course, GroundWork itself is a well polished version of combining and customizing available OSS projects and platforms. And I agree it'd be nice to start tracking these types of projects more formally, writing up notes as appropriate.
Tony: thanks for the note 😉 I combined the concepts of "minimum" and "up to" to help normalize the table a little bit.
Stumbled upon this site pretty much by accident. Thanks for including us in the list.
When we started providing support for OpenNMS, we didn't want to use the tired "per node" or "per license" method of pricing. "Per node" doesn't scale for anyone but the vendor, and since all of our code is available under the GPL, the license model didn't work either.
So we price our support per support contact. In the basic support subscription (the one you list for $4995) you get two contacts: a primary and a backup. They can manage as many OpenNMS instances as they want, and we encourage our clients to run a production as well as a development version.
Thus it makes it kind of hard to price per node. We have some clients with 100 devices, but we have one with nearly 50,000. Also, since we are client-less, the "work unit" is more along the lines of the number of interfaces and services being managed versus "nodes". The 50K device client has about 66K interfaces whereas another client with only 12K devices has nearly the same amount of interfaces since their devices are mainly routers.
As a pure services company we try to price according to the number of people accessing our services versus how "much" OpenNMS a client uses.
Thanks again for including us and I hope this clarifies things a bit.
Tarus: thanks for taking the time to clarify and add explanation. I've corrected the original post to note the unlimited quantity.
I was told that groundwork cost per year for entprise monitoring is $23000 and unlimited devices. I see here that $8000 for 50 devices. What do you mean by "limited only by the performance capabilities of the monitoring server"? So even in $23000 enterprise contract the customer has to pay additional charges?
Hyperic charges based on cpu's and not based on sockets.
Gokulnath: this pricing is from 2007, so it could be out of date both in how Hyperic slices up what to price and the price itself. So, thanks for the update!