As one of the world’s largest software companies, we often field a number of questions from people thinking about making the switch to Accurev for software configuration management. I thought I might post some of the common Accurev questions and our answers to them proactively to save a little time
How many users do they have and how big is your repository?
We have 125 users, spanning 3 product offerings, and the repository is about 20Gb.
What kind of storage are you using on the backend?
Nothing too fancy. Windows box with a 100Gb RAID array – mostly done for data redundancy. I think we could have gotten away with less, but we erred on the side of caution.
How fast has your Accurev repository grown since day one?
We have averaged a growth rate of roughly 200-250Mb per year over the past 6 years.
What is the performance of Accurev like?
I can’t say we’ve had many complaints. We have a setup where the primary server is in colocated with the largest US-based team, and we have AccuReplica servers at our other US sites and another at our site in India . My experience is that it is much faster over VPN than Perforce or ClearCase were.
How long did it take for users to become familiar with the streams concept?
Not long at all – hours at most. Our developers caught on right away and loved it. We ran some education sessions to help get things jump started, but most developers understood immediately – “Hey, I don’t have to make the same fix 4 different places anymore!”, and Accurev merging is so good that merge points are no longer a pain for developers.
How many admins do you have?
Zero. We setup some simple policies for behavior – some best practices that we tell everyone, and then my release engineer is in charge of the ‘mainline’ streams that will turn into product releases. It takes him less than an hour a week to do the admin type tasks in the system (update triggers, lock streams on occasion, create user accounts), but by and large the things a ClearCase or CVS admin would do are done by the individual developers. Our developers love it because it empowers them to do what they need to do.
What kind of response speed do you get from Accurev support?
Accurev’s email-based support is excellent. I can’t recall ever waiting more than an hour or two for someone to answer, and usually it takes only 1-2 emails to resolve the problem – the support team knows that they are doing. In 6 years I can’t recall EVER having to pickup the phone and call the support line because the email is answered so quickly.
How painful are server and client upgrades?
They are a piece of cake. The regular installer is smart enough to perform the upgrade in a wizard fashion and there are no skill testing questions! When we went to v4.6 it took less than an hour to perform the upgrade, push new clients and verify that everything was working, including our build system. If you have provisioning software you could deploy the new clients in moments. We have upgraded our servers all the way from v3.0 and have never experienced any issues at all with the upgrades.
What do you like least about Accurev?
Directory scanning can be more time consuming than I’d like. It would be nice if it were done as a background task all the time when the UI is running instead of only when you make a request. Other than that, I’m hard pressed to think of anything I don’t like.
What is the best thing about Accurev?
Developer Empowerment – far and away this tool lets my developers do what they need to do exactly when they need to do it.
Parrallel development – never have I worked on projects that required so little codeline coordination as we’ve enjoyed since taking on Accurev. Merging code is so easy that merge points are done in a fraction of the time. And Accurev makes it possible to merge your code so often that merges are less painful when they happen.
Visualization of change – with the dynamic process version browser and the annotation tool visualizing how your code has changed over time is a snap.