Elasticsearch

Kommentare: Excellent. The ELK stack is the foundation of our audit process, and it's held up really well. The installation can be automated, and archiving in the AWS environment works quite well. We have not yet had a problem in scaling resources to match our resource needs. The integrations with logstash, beats and kibana have been excellent, and used to great effect.

Vorteile: Scale: You can run this from a single server or even co-installed on a database or file server. I wouldn't recommend it, but let's just say it will run in a small box. On the top side, Elasticsearch will run in clusters managing multiple Terabytes of data.
Features: Excellent flexibility to absorb multiple types of data sources, and great integration with Logstash and Kibana.

Nachteile: The upgrades in terms of archived data. This platform changes significantly on an annual basis. If you do a ton of customization, automation, or work with a lot of data, this can be an issue as you will need to update ALL of your data for every upgrade.

Kommentare: I work with Elasticsearch as a developer on a daily bases for textual search. It is the greatest tool I've used in years. Really like it.

Vorteile: It is just the best back end for search engines in the market. A NoSQL database that is trustworthy. Also, it is open source. Incredibly easy to use.

Nachteile: Mostly for developers: other people would have a hard time with it.

Vorteile: I've been using Elasticsearch since early days, with very different things in mind. I started with simple text search - with some additional tweaks, stemming and other cool features it helped us drive enormous traffic to our website. I can't imaging pulling it off so easily with any other tool. Every day I use it for web server log analytics. Search and great visualizations make it an absolute essential in work my toolset. We also run a lot of algorithm analytics on top of our Elasticsearch cluster. If you're looking for managed options check AWS Elasticsearch Service, or the recently introduced Elastic Cloud.

Nachteile: My only concern with Elasticsearch is that it might get expensive to run pretty quickly. But with a certain amount of effort put into optimization it's gonna be worth it.

Kommentare: We use Elasticsearch to filter and sort search results in our marketplace. We've built out many complicated queries that allow us to do interesting things like geo-based queries, personalization, and time boxed deals.

Vorteile: Elasticsearch offers a very flexible system for adding search capability to your systems. It is also capable of much more. The REST API and great documentation makes getting started very simple. Elasticsearch was also designed with scaling in mind. Adding nodes and self balancing is quite easy. AWS offers hosted Elasticsearch that makes spinning up your first cluster as simple as a few clicks.

Nachteile: Writing complicated queries can be quite tedious at times. The JSON interface is not always easy to read when trying to match up parentheses. Upgrading from older versions is not a simple process.

Kommentare: We're ElasticSearch mainly to index large amount of logs from several servers. Its makes it very easy for us to index and search logs. Logstash sends the logs and with Kibana we access to logs and create nice dashboards. But, you have to manage your indexes. For our log an index is created every day which we reindex monthly to a new index, then we do a forcemerge and after that we delete the daily indexes. This keeps the number of shards low. If we don't do this we run into problems because if to many shards. But, you can schedule this via curl and with every new version of Kibana/ElasticSearch you can do more and more via the GUI. For us this is the number one tool to index and search gigabytes of logs on a daily base and we're able to keep months of logs and still be able to search through it.

Vorteile: Indexing large amount of data
Searching data
Very scalable
Creating a cluster is very easy
Ability to send commands via CURL to the API
Creating snapshots of your data

Nachteile: Managing indexes can be a bit of a pain
Sometimes issues with indexes becoming read only