Overall Satisfaction with Elasticsearch
In my organization, Elasticsearch is used as a fast and simple solution for providing search capability to text-based data and to easily perform analytics for our dashboard. Being a JSON-based response system, our APIs become simple and support multiple behaviors by translating to Elasticsearch queries. Not only does Elasticsearch act as our analytics platform, but also it serves as secondary database storage.
- Text-based searches on data
- Daily, weekly, monthly analytics on data
- Super easy scripting with painless scripting language
- Relational data query
- Sync data from SQL on table change (with hash maybe)
- Provide better tutorials for beginners
- Faster searches on our application have resulted in better usability and increased application use
- Analytics dashboard has given our managers a better understanding of day-to-day activities
- Being a backup data store, we need not touch SQL database while doing data dumps for local data science projects
Search and analytics capabilities of Elasticsearch are superior to its competitors. Being open source, it is a cheaper and faster solution than other competitors. Installation is straightforward and it can be potentially deployed anywhere and everywhere! There is no need for expensive subscriptions or pay per data.
Do you think Elasticsearch delivers good value for the price?
Are you happy with Elasticsearch's feature set?
Did Elasticsearch live up to sales and marketing promises?
Did implementation of Elasticsearch go as expected?
Would you buy Elasticsearch again?
Elasticsearch is best suited for search, analytics, aggregation, and consumption from single tabular structured data. It works best if you sync your data at regular intervals either with Logstash or any other custom sync process.
However, Elasticsearch still does not support relational queries out of the box. You could denormalize your data before every sync, but that has the potential for complicating the sync process very fast.