hg2012 - v. 1.6.2
haxogreen 2012
...not your regular mercury diet
| Speakers | |
|---|---|
|
|
Thibaut Britz |
| Schedule | |
|---|---|
| Day | Day 1 - 2012-07-26 |
| Room | Chalet |
| Start time | 19:00 |
| Duration | 00:45 |
| Info | |
| ID | 64 |
| Event type | Lecture |
| Track | Systems |
| Language used for presentation | English |
NOSQL in practice at medium scale
Operating a cluster of more than 250 NoSQL servers running Cassandra
I will explain how Trendiction operates a cluster of more than 250 Nosql servers running Cassandra. I'll detail how jobs are being executed on our cluster in order to crawl and analyse the collected data. The analysis includes automatic detection and normalization for content type, language and duplicates in order to finally being able to deliver content to our customers: market research institutes and media analysis companies.
I will also unveil an internal tool providing a global overview over the distributed job execution service, allowing to quickly determine the ongoing workload on the 250 nodes.