Archiv für die Kategorie ‘Frank Celler’

NoSQL matters 2012

Thursday, March 29th, 2012

On May 29-30, NoSQL matters 2012, an international NoSQL-conference will take place in Cologne (Germany). The event is supposed to give pioneers in the field an opportunity to present recent innovations in the dynamic field of large, scalable, high-performance databases. The presentations cover the big NoSQL-topics, like Riak, MongoDB, Hadoop, Redis or CouchDB and shed …

Using RSpec for Testing a Rest-Application

Tuesday, March 20th, 2012

or Using RSpec without any Ruby Knowledge There are quite a lot of test frameworks out there. For C / C++ pieces I am using Boost. For testing the HTTP interface of one our software products I used phpunits. @moonbeamlabs pointed out to me that RSpec together with HTTParty provides a very nice setup for …

Benchmarking SSD with MongoDB and CouchDB, Part 4

Wednesday, March 7th, 2012

As noted in part 1, benchmarking is a black art. Most of the time it is unclear, what you want to measure and what you are measuring. Setting Up a Durable Document Store For today’s test, I want to investigate the following setup: I need a document store, which guarantees the client that the data …

Benchmarking SSD with MongoDB and CouchDB, Part 3

Sunday, October 23rd, 2011

I’m still waiting for the Firewire adapter to arrive. The first order got lost, the second order was canceled, and the third order has not yet arrived. Thanks to Amazon I got my money back without any problems. Therefore, I decided to ignore the Mac OS X for the moment. In addition to the OCZ …

Benchmarking SSD with MongoDB and CouchDB, Part 2

Wednesday, September 28th, 2011

In Part 1 I started to investigate the performance impact of SSD on MongoDB and CouchDB. However, I got sidetracked because of the strange behavior of Mac OS X. The Mac is lying about msync. You have to force a synchronization by explicitly calling fnctl with F_FULLSYNC for the underlying file descriptor. So, I decided …

Benchmarking SSD with MongoDB and CouchDB, Part 1

Saturday, September 24th, 2011

Doing benchmarks is always complicate because it is most of the time totally unclear what one wants to measure and what will be measured. There is article describing this dilemma, see Benchmarks-You-are-Doing-it-Wrong. The CouchDB – The Definite guide contains a longer version of this, see High Performance. I assume that this is also one of …