Reviewing this website, their information on TimescaleDB is either false or outdated:
"Time series databases like InfluxDB or TimescaleDB are unable to handle many IoT applications by themselves—if they require metadata operations, for example. They often need to work in combination with an RDBMS, duplicating maintenance efforts and costs. Besides, their performance is usually optimized for single-node processes."
This statement is true for InfluxDB (which doesn't store relational metadata, and where multi-node is an enterprise paid feature), but not for TimescaleDB. TimescaleDB is built on PostgreSQL, so it is in fact an RDBMS and allows you to store relational data alongside time-series, plus it is now distributed (multi-node), and this distributed feature is free [0].
Which is why we have some really large IoT deployments, including at Schneider Electric, Siemens, Bosch, Northvolt, and many others.
(I've met some folks from Crate and they've all been friendly, so I'm sure this is just an honest mistake, but it is incorrect nonetheless.)
I have not seen a proper TimescaleDB vs. Crate comparison, but at a high level, TimescaleDB is a relational database for time-series, ie "Postgres for time-series", so most developers find it really easy to use (and performant). Also, we just made all of our enterprise features free, so TimescaleDB is now completely free to use.
Hi there, thank you for pointing out the sections of the website that got outdated. We'll work on it asap.
However, we still need to defend our use cases... CrateDB has been distributed since the beginning, and it is a very mature technology in this regard. The page you mention is a high-level comparison indeed, but we do talk a little more about Timescale in this post:
We worked on this with your team actually, which in fact assisted us in configuring Timescale. They were great -- but CrateDB performs better here.
We've been also in touch with your team with regards to more comparisons with Timescale... As in order to be fair, we wanted to share everything in advance. This is not a Crate vs Timescale war! Still, we believe CrateDB is amazing and we invite everybody to try it: https://crate.io/download/
- Scalability
CrateDB is built for horizontal scale from the ground up on top of distributed technologies. We have customers using clusters with 80+ nodes in production for many years now.
Timescale just released their multi-node feature in beta and they follow a different concept then we do. While Timescale uses a leader (access node) - follower (data node) model with a single point of failure CrateDB is built on a shared-nothing architecture. Many features you would want to see in a distributed system are present in CrateDB and still missing in TS:
- cluster wide replication
- automatic rebalancing
- cluster wide backup
- shared nothing architecture / no single point of failure
- Full Text Search
CrateDB is built on Lucene and parts of ES and includes search capabilities you would typically need a separate product for when using PG/TS.
- Distributed Query Engine
Yes, PG/TS are fast if you query "small" amounts of data (e.g. last days data). But if you have distributed system, you might as well also want to run queries on larger data sets.
- Geospatial Queries
Powered with Lucenes BKD-Trees
---
Disclaimer:
I work for Crate.io and I also think Timescale are doing awesome stuff in many ways and give Influx the competition they deserve. I don't see us in direct competition (at least not yet), as the focus of Timescale is clearly more on smaller use cases.
13 comments
[ 4.0 ms ] story [ 22.9 ms ] thread(For those not getting it, Rust’s primary library distribution website is crates.io, one letter different)
[0] https://crate.io/cratedb-comparison/
"Time series databases like InfluxDB or TimescaleDB are unable to handle many IoT applications by themselves—if they require metadata operations, for example. They often need to work in combination with an RDBMS, duplicating maintenance efforts and costs. Besides, their performance is usually optimized for single-node processes."
This statement is true for InfluxDB (which doesn't store relational metadata, and where multi-node is an enterprise paid feature), but not for TimescaleDB. TimescaleDB is built on PostgreSQL, so it is in fact an RDBMS and allows you to store relational data alongside time-series, plus it is now distributed (multi-node), and this distributed feature is free [0].
Which is why we have some really large IoT deployments, including at Schneider Electric, Siemens, Bosch, Northvolt, and many others.
(I've met some folks from Crate and they've all been friendly, so I'm sure this is just an honest mistake, but it is incorrect nonetheless.)
I have not seen a proper TimescaleDB vs. Crate comparison, but at a high level, TimescaleDB is a relational database for time-series, ie "Postgres for time-series", so most developers find it really easy to use (and performant). Also, we just made all of our enterprise features free, so TimescaleDB is now completely free to use.
(Disclaimer, I work at TimescaleDB)
[0] https://blog.timescale.com/blog/timescaledb-2-0-a-multi-node...
However, we still need to defend our use cases... CrateDB has been distributed since the beginning, and it is a very mature technology in this regard. The page you mention is a high-level comparison indeed, but we do talk a little more about Timescale in this post:
https://crate.io/a/comparing-databases-industrial-iot-use-ca...
We worked on this with your team actually, which in fact assisted us in configuring Timescale. They were great -- but CrateDB performs better here.
We've been also in touch with your team with regards to more comparisons with Timescale... As in order to be fair, we wanted to share everything in advance. This is not a Crate vs Timescale war! Still, we believe CrateDB is amazing and we invite everybody to try it: https://crate.io/download/
(... I also work at Crate, hehe)
- Scalability CrateDB is built for horizontal scale from the ground up on top of distributed technologies. We have customers using clusters with 80+ nodes in production for many years now.
Timescale just released their multi-node feature in beta and they follow a different concept then we do. While Timescale uses a leader (access node) - follower (data node) model with a single point of failure CrateDB is built on a shared-nothing architecture. Many features you would want to see in a distributed system are present in CrateDB and still missing in TS:
- cluster wide replication - automatic rebalancing - cluster wide backup - shared nothing architecture / no single point of failure
- Full Text Search CrateDB is built on Lucene and parts of ES and includes search capabilities you would typically need a separate product for when using PG/TS.
- Distributed Query Engine Yes, PG/TS are fast if you query "small" amounts of data (e.g. last days data). But if you have distributed system, you might as well also want to run queries on larger data sets.
- Geospatial Queries Powered with Lucenes BKD-Trees
---
Disclaimer: I work for Crate.io and I also think Timescale are doing awesome stuff in many ways and give Influx the competition they deserve. I don't see us in direct competition (at least not yet), as the focus of Timescale is clearly more on smaller use cases.