• Docker pitfalls

    Containers have been really trendy recently and I can certainly agree that they are changing the world of software delivery. Although the overall idea of putting applications inside immutable containers is not new, Docker is the first widely adopted implementation which makes it easy as a snap. Containerization has multiple benefits but the most important one (in my opinion) - it makes software easily deployable and therefore scalable.

  • Deploying Ruby on Rails applications with ansible

    capistrano was the best Ruby on Rails deployment tool for a years, but if you want to apply DevOps practices in your project then you might want to try out ansible. It allows you to significantly simplify your operations by creating easy YAML-based playbooks. It’s good for configuration automation, deployments and orchestration. And most important - ansible is very easy to learn.

  • Visualize Delicious Tags With Gephi

    I’ve recently was playing around with a nice graph visualization utility called gephi. It allows to easily visualize large networks and graphs using a set of pre-defined layouts. It’s possible to import some existing graph into gephi using one of the formats it understands. I’ve decided to try to visualize tags of my Delicious bookmarks with it.

  • Video Recording And Processing In Android

    Our Android app has an ability to record videos and also process frames as they are coming from camera. Processing includes traffic sign discovery and hard subtitles addition (GPS coordinates and timestamp). So, our final goal is to be able to access raw frames (byte[]) and to record video at the same time. It turned out that it’s quite hard to perform such kind of tasks on Android simultaneously, but we came up with our own way of doing it. Let’s start with what did not work.

  • Downloading Multiple Files In Ruby Simultaneously

    We’re saving photos of all recognized traffic signs in RoadAR on Amazon S3 bucket. There’s a regular process when some of those files should be downloaded, cropped and feeded to the machine learning algorithm which produces SVM output file used in our recognizer. The problem that there’s quite a few of such files to be downloaded in one batch (can be 300-400k at our current scale). I am going to cover 3 different ways of downloading multiple files in Ruby:

subscribe via RSS