Author Archives: Amund Tveit

Continuous Deployment for Python/Tornado with Jenkins, Selenium and PhantomJS

The purpose of Continuous Deployment is to increase Quality and Efficiency, see e.g. The Software Revolution behind Linkedin’t Gushing Profits or read on This posting presents an overview of Atbrox’ ongoing work on Automated Continuous Deployment. We develop in several … Continue reading

Posted in cloud computing, continuous deployment, infrastructure, tracer bullet development | Tagged , , | Leave a comment

Combining Hadoop/Elastic Mapreduce with AWS Redshift Data Warehouse

There are currently interesting developments of scalable (up to Petabytes), low-latency and affordable datawarehouse related solutions, e.g. AWS Redshift (cloud-based) [1] Cloudera’s Impala (open source) [2,3] Apache Thrill (open source) [4] This posting shows how one of them – AWS … Continue reading

Posted in analytics, cloud computing, Hadoop and Mapreduce | Tagged , , , , | 5 Comments

Mapreduce Algorithms – Presentation held at O’Reilly Strata Conference

My presentation held at O’Reilly Strata Conference in London, UK, October 1st 2012 Best regards, Amund Tveit

Posted in cloud computing | Leave a comment

Atbrox @ O’Reilly Strata Conference in London

Atbrox is participating and holding a Hadoop/Mapreduce algorithm related presentation at the O’Reilly Strata Conference in London October 1st and 2nd. If you are there and would like to meet Atbrox send an email to

Posted in big data, cloud computing | Leave a comment

A large-scale in-memory storage example – social network data

This posting is a follow-up to the large-scale low-latency (RAM-based) storage related price estimates in my previous posting Main takeaways from Accel’s Big Data Conference. Assume you were to store and index large amounts of social network updates in-memory, e.g. … Continue reading

Posted in in-memory, information retrieval, infrastructure, RAM | Leave a comment