Scalarium Update: User Accounts, Perfomance Improvements

2010/04/11

We just rolled out a new feature that have been a long time coming and sure were requested a lot by our customers. You're now able to define multiple users to access the same account, so you're a) not required anymore to hand over your credentials to someone else and b) can track who did what. Someone started an instance? You'll see who it was. Deployment kicked off? You'll find out who's to blame, no problem. But bear with us for the glory details, let's start at the beginning.

The new user management hides behind the settings menu item which you'll find at the top of every Scalarium page.

Adding a user is easy as pie, just specify name, email and password, and he's good to go, he'll even get an email letting him know that he's now a Scalarium user.

There's two types of users, administrators and normal users. By default, new users are created as normal users without administrator privileges.

You have to explicitly edit a user to set the administrator flag, which ensures you don't enable it by accident and only for users you really intent to have administrator privileges.

What do these privileges entail? For now, only administrators can see and manage assets directly related to EC2, e.g. SSH keys, AWS credentials, EBS volumes and Elastic IPs. When you log in as a user with no administrator privileges, this is what you see.

Neat, huh? Now, when Vlad decides to stay true to hist last name and deploy or start something, everyone will know what he's been up to.

Now, we know what you're going to ask for next, you'll want to have users with specific capabilities, only being able to manage clouds, instances or only allow them to deploy and nothing else. We want the same thing, and it's on our list. First and foremost we wanted to get the basics of user accounts out the door, and work from there. Hope you find it as useful as we do!

Another thing you'll hopefully notice is that the dashboard and cloud pages are much snappier. We made a lot of improvements to speed up the database communication and to reduce the amount of data we're querying (detailed blog post to follow), and with that drastically reduced the response time of most pages, but most notably the ones you're coming back to the most, clouds and the dashboard. We weren't very happy with how long some of the pages took to load, and we felt your pain too. It's only a start, but a pretty good one if you ask us.

Apart from that we're working heavily on improving our own infrastructure. As the number of instances managed by Scalarium grows we're seeing the hotspots and problem areas of our own backend, identifying the causes and working hard to fix the problems so that we can guarantee a great quality of service throughout.