Posts from 2011

Today I'd like to announce a few changes to our Assembly index page that some customers could already see over the past few days and that are now rolled out to everybody.

We are happy to announce the release of our new /file/filter Robot that allows you to filter incoming files based on all of the file's metadata, and more.

We are happy to announce the official release of our /audio/encode Robot! Having been battle-tested in production by quite a few customers for several months now it should greatly help with your audio needs. It is priced at 25% of the video encode robot and starts at $1 per GB. As always, the more gigabytes you use, the less you pay per gigabyte.

Some customers have already had the pleasure of using it behind the scenes, but now we are finally ready to officially released our SFTP Robot. It makes storing files on your own servers a breeze and costs just the same as the /s3/store Robot. For all the intricate details, please check out its documentation and feel free to leave your comments on it.

Things have been nothing less than crazy over the last two months. We have gone from 300.000 uploads to almost 1.000.000. During peaks, we are now fully utilizing 20 machines (160 cores), and we are still continuing to grow quickly.

Update: The maintenance was a success, we were down for only 4 minutes.

Today, we are happy to announce the 1.0 release of two new Gems for the Ruby community.

If you are not using a storage Robot, because you have a setup where you prefer to manually fetch our result files from our temporary URLs, here is some good news:

The last few days have been crazy. As of today, we have processed 300.000 uploads, up from 250.000 only 7 days ago (!).

As some of you have noticed already - we have just launched a new design, which will hopefully enrich the browsing experience on our website!

We have some good news for people using CDNs or bucket names that are not well-formed. As of today, we are supporting a new parameter called url_prefix for our /s3/store Robot.

We have a new parameter for the /image/resize Robot. It is called frame and allows you to select a specific frame to use for the operation when you are dealing with animated GIF files.

Amazon S3 is great. Except when it isn't. When you are using it a lot - we have stored around one million customer objects so far - you begin to see some odd things that look more like "eventual madness" than "eventual consistency".

While most of our uploads successfully complete their journey across the world wide web, sometimes one can get lost along the way.

Today we are making the /image/resize Robot's zoom parameter public. Up until now it was an undocumented parameter called "dont_zoom" with a default value of "false" that only few knew about. We decided it makes more sense to make it public now so everyone can profit. It is now called "zoom" and has a default value of "true" - so everybody who used it so far, please change your Templates accordingly.

Today, we are officially announcing support for Google's WebM video format.

Please check out the new rotation parameter for the /image/resize Robot. It allows you to specify rotation degrees for your images and by default fixes any miss-rotated images. You can see it in action on the rotation demo page.

As you probably know, Transloadit has been running its production infra on Amazon's cloud since day 1 - our block stores, encoding servers, load balancers, the full monty. Except for 1 often forgotten, yet essential part of infra: our name servers.

After a few recent complaints from our customers, it became evident that our x264 encoding configuration was way too slow. Not only had we missed some critical optimizations when configuring the library, but we were not using optimal parameters for the encoding itself either.

Last night, we switched our default machine type from Amazon's c1.medium (2 core) instances, to x1.large.

Today the Transloadit team is happy to announce the arrival of a new feature: image cropping based on coordinates.

We are excited to announce that we have launched a major overhaul of our documentation. It is much more structured now, has a full example of a minimal integration and is all in all easier to grasp. Please check it out and give us your feedback.

For those of you curious about what it's like for us to run on top of Node.js, I gave a talk on the subject last night: