Creating a new Data organization, what is the best solution?

Hi,

I'm building a new structure for my data and i'm considering 2 solutions :

  • Using a DB to store my documents, and index them in ES
  • Using only ElasticSearch to store and search my documents (maybe
    stored in Hadoop)

Since ES can store securely my data, scale it and access it, what is the
point to have a DB on side of ES like CouchDB ?

--

Hi,

Some people simply trust more mature DBs or HDFS more than ES or feel more
comfortable with them. At the same time, others rely on just ES to hold
their data. If you trust the hardware you run ES on, if you have high
enough replication level, if you trust your team's ability to upgrade ES
smoothly, etc. then you could use just ES. One thing ES doesn't have
worked out well yet is backup, which may be important to you.

Otis

Search Analytics - Cloud Monitoring Tools & Services | Sematext
Scalable Performance Monitoring - Sematext Monitoring | Infrastructure Monitoring Service

On Tuesday, September 4, 2012 11:05:53 AM UTC-4, Looky wrote:

Hi,

I'm building a new structure for my data and i'm considering 2 solutions :

  • Using a DB to store my documents, and index them in ES
  • Using only Elasticsearch to store and search my documents (maybe
    stored in Hadoop)

Since ES can store securely my data, scale it and access it, what is the
point to have a DB on side of ES like CouchDB ?

--