ZVNcloud

Benefits:

It's relatively easy to use. I say "relatively" in that the UX is not particularly good but our main criterion for using it is that we can get senior developers to use it when deploying application updates across clusters rather than putting the burden on systems staff.

Drawbacks:

It's overpriced, buggy and not particularly well engineered. The APIs are poorly constructed and do not scale well when monitored.

In sum:

We had previously used a number of entirely systems-based load balancing systems and it was proving difficult to manage and coordinate application updates for clients. We needed something that would have some sort of management interface to reduce the possibility of errors and to make it more accessible to other team members. We'd tried the community version and while it was far from perfect did at least arrange things in a reasonably logical manner. The cost of the application was excessive but as the developers are clearly a small company this would be more favourable than a larger company for whom we'd be way down their list of priorities. To our horror, when we were approached after purchasing several licenses for the product if we had any questions, we asked about a migration path from the community version. They immediately refused and said they would only be able to help if we bought a separate support contract. They have made the business practises very clear. We won't be giving them any more money and at some point when a better product materialises, we'll go for that. For now we are using the product and manually fixing issues within their code (which is mostly Perl-based). This is still better than managing it all by hand, but far from ideal given how much we've paid for what is supposed to be an enterprise solution. Our advice would be to think extremely carefully before committing any budget to this.