Apache Solr vs Elasticsearch Differences | How to Choose Your Open Source Search Engine - Sematext

[@pragatirpatra]

Yes please make video on cross-collections joins. We are using it in Solr. I would like to know how good or bad it is.

[@tasdiqshaikh15]

Very neatly explained, nice work 👍🏼

[@pumpkinthestrange1]

Very useful video, short and to the point. Thanks a lot

Would definitely like to hear more about cross collection joins :)

[@LieberLois]

How do you feel about OpenSearch? Is it basically the best of both worlds? :) great video!

[@Sematext]

Good question. I just made a video on OpenSearch vs Elasticsearch here: kzbin.info/www/bejne/bGPYZGOKhJKCqaM In short: OpenSearch is not a bad tool. It has its reasons for existing and does open up a few doors to you because of the Apache 2.0 license. However, as long as you are not trying to sell Elasticsearch's code or be a competitor of Elastic Cloud (ES's Cloud management services) you can freely use Elasticsearch for your projects. And because Elasticsearch is more popular and has better documentation, you can typically troubleshoot ES easier. But OpenSearch is a great tool as well and is growing in popularity. In the near future, things may change. We will just have to wait and see.

[@LieberLois]

@@Sematext Thanks for the answer :) It's true that ElasticSearch is more popular and has more documentation - however I'm not sure about the licensing you mentioned. ElasticSearch isn't open for free use anymore and can AFAIK not be implemented for customers freely. Can u comment on that?

[@Sematext]

@LieberLois Elasticsearch switched to a dual license. The elastic 2.0 license and the SSPL (server-side public license). You will have to choose between one of those 2 licenses. Overly simplified, if you use elasticsearch on-premises or in your project, and you are not selling Elasticsearch's code itself as a service, you should be covered by the elasticsearch 2.0 license. it is free to use. You can use additional tools to get the most out of elasticsearch that cost money, but the core (elasticsearch and kibana) are free to use. According to Elastic, even if you "build a SaaS application using Elasticsearch as the backend...you can use [the] default distribution or develop applications on top of it for free, under the Elastic License." - Citation and info from: www.elastic.co/pricing/faq/licensing#dual-licensing BUT! If you are using elasticsearch products and selling the products as a service, you either have to do the SSPL or pay a licensing fee. (This was mainly done to stop Amazon from selling Elastics products without them getting a cut) So it can still be "free to use" depending on the use case. I recommend reading the licenses and the FAQ on Elastic's site to know if your use case is covered by one of those licenses. :D

[@MarkuBernardut]

@@LieberLois Elasticsearch does have a "Free" basic version. If you want more recent features that include GAI and other tools there is a license fee,, but many people still use the "basic" free version

[@walecharohit]

Need a video on that(cross-collections joins, the restricitive parent child structure in elasticsearch, and how elasticsearch is faster in that) too..Thanks in Advance !!

[@Sematext]

Great suggestion, thanks.

[@bigstones84]

Wow, super clear!

[@Tegna-Abächerli]

Elastic is the best product, super support and very easy and secure 👍👍👍

[@Sematext]

@CostaMariana1 Thanks for sharing your experience. How long have you been using Elasticsearch?

[@ranvijaymehta]

Nice information 🙏🤗🙏🤗

[@jaydevnath142]

Can you compare elastic and Amazon opensearch

[@Sematext]

Sure can! Hope this helps. kzbin.info/www/bejne/bGPYZGOKhJKCqaMsi=VXPlmL_Va1W3RwNi

[@adeolaojo]

Great explanation

[@Sematext]

Glad it was helpful!

[@sasithwikum2756]

Nice explanation

[@danielgianni-ror]

I would like some guidance, if I need to create a search engine to index multiple html files on another server, which option should I use?

[@Sematext]

Hey Daniel, Great question. When indexing dynamic content such as HTML files, both ES and Solr can ingest unstructured data. (which is probably what you will need for HTML files) In that context, both will do about the same performance-wise. The real concern I would have for that use case is scalability and how many files are "multiple HTML files." If you get really big, many of the necessary tools for elasticsearch to function well become paid-only. But as long as you're not trying to make your own Google v2.0, either one should work fine. Both do really a pretty good job schemaless. (unstructured data) Hope that helps!

[@MrSlundell]

Audiosync!