Rendered at 07:38:35 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
jesol 1 days ago [-]
I've been working on a graph database in Rust this year actually! I'd love to hear anything you can talk about wrt the query planner and/or how you decided to do cardinality estimation. I decided to go with an EAV graph which makes CE pretty complex, and it's been an interesting challenge to balance quality and speed and expressiveness in the query language
thedreammachine 1 days ago [-]
What kinds of graph shapes or query patterns do you feel are the worst case for object storage?
GeorgeCurtis 18 hours ago [-]
OLAP queries, deep multi-hops where latency is a priority.
As long as the sub-graph you're trying to hop is cached, then there's no problem or latency issues. However, if you need to do a deep hop query, where all those nodes and edges are in cold storage, each hop costs ~50ms. So a 10-hop would take ~0.5 seconds.
Again though, we find most people are using us for agentic workloads, so even this worst case scenario the LLMs make up the majority of the latency.
mentioum 2 days ago [-]
We've been having some issues with intermittent performance on multi hop queries.
What's your p99 like for multi hops?
GeorgeCurtis 2 days ago [-]
In prod we see p99’s of <10ms ms for warm queries and around 50ms per hop for cold queries.
mentioum 2 days ago [-]
Hmmm... I'll get in touch. Got an email i can reach out to, there doesn't seem to be one listed on your website?
I'm more concerned about if the p99s stay consistent when things get spikey.
dgraph is fine otherwise...
GeorgeCurtis 2 days ago [-]
Sure! You can email me personally at george@helix-db.com
keynha 24 hours ago [-]
[dead]
zw17 2 days ago [-]
If your use case is OLAP based, please check it out PuppyGraph. It’s a graph query engine that sits on top of your Lakehouse (no ETL required). Our benchmark has shown consistently that 10-hop queries across billions of edges in <2 seconds. Our customers including some most data demanding companies like Coinbase, Datadog, Palo Alto Network, Netskope, AMD, etc.
mentioum 2 days ago [-]
It's not, its actually our prod db with direct user usage - we self host a large dgraph cluster. We have a very large number of people manage their car and car histories with us and host a full replica of the UK MOT Database.
We're fine with clickhouse and redshift for the OLAP work we do. I've been looking at ParaQuery lately if I really want to speed that up.
GeorgeCurtis 2 days ago [-]
This sounds like a perfect usecase. Would love to learn more and see if we can help!
email us: founders@helix-db.com
GeorgeCurtis 2 days ago [-]
PuppyGraph is a good fit for OLAP for sure.
We’re just two young founders sharing what we’ve been building, so I’ll take the drive-by competitor plug as a compliment :)
Definitely a different focus though. Helix is OLTP, built for operational graph + vector workloads, especially apps/agent memory where low-latency traversals and writes are concerned.
jauntywundrkind 2 days ago [-]
And is open source.
GeorgeCurtis 19 hours ago [-]
puppy graph is not open source
fouc 1 days ago [-]
where is puppygraph's source code?
nulltrace 1 days ago [-]
[dead]
rgbrgb 1 days ago [-]
congrats on the launch! site and docs look great.
can you host this yourself or do you need to use helix-cloud? the chat thing on the side seems to push me to helix-cloud but it looks like that starts at like $600/mo which is above my experimentation budget.
looking for a db for an agent memory application and i'd probably start with something that's just self-hosted / freeish. postgres is working ok but I want to start ingesting server and chat logs.
GeorgeCurtis 1 days ago [-]
You can definitely host it for free locally for now.
We aim to launch our GA cloud at the end of this month, which will be much more affordable.
Onawa 1 days ago [-]
Why do you say "for now"?
GeorgeCurtis 1 days ago [-]
Lapse of communication. For now as in you’ll be able to host it without reading the source code.
Soon you’ll be able to host it yourself AND have access to the source code
let_rec 23 hours ago [-]
This seems like a great idea.
What reassurance can you offer devs that are hesitant to try a new data-store?
GeorgeCurtis 19 hours ago [-]
Can't imagine why they'd be hesitant, Helix is awesome, we've never had any data loss issues, and are completely ACID.
I'd encourage them to start a local instance with claude/codex to build a mini project and see what it's like.
aitchnyu 13 hours ago [-]
Does it work only on S3?
cjlm 2 days ago [-]
Currently #5 on gdb-engines.com - definitely worth a look.
dig1 1 days ago [-]
How rankings are calculated? HelixDB has zero feature scores, unlike e.g. JanusGraph which is at #27 or Neo4j at #1.
cjlm 1 days ago [-]
The feature scores are from a paper cited on the site. The ranking features are proprietary to dissuade gaming but are a blend of open public data across a number of different sources.
GeorgeCurtis 2 days ago [-]
yooo this is awesome. Didn't even realise :)
caust1c 1 days ago [-]
Where's the source code for the database itself? Looks like the repo is just a client.
We’re 100% committed to going back to open-source on an Apache 2.0 license as soon as possible. In the meantime, you can continue to deploy us completely for free, however you like, using the compiled docker container.
tao_oat 23 hours ago [-]
Unfortunately it's not possible to read this without an X account.
GeorgeCurtis 19 hours ago [-]
[flagged]
ymir_e 1 days ago [-]
Congrats on the launch George!
Looking forward to looking into the generalised AI memory layer when it comes out.
lennertjansen 13 hours ago [-]
Are you ACID? And does this version have multi-tenancy?
maxrumpf 2 days ago [-]
does it support fts/vector on edges of the graph?
GeorgeCurtis 2 days ago [-]
Yes you can put vectors, full text data, secondary and range indexes on both nodes and edges.
brene 2 days ago [-]
How does this compare vs. Turbopuffer?
GeorgeCurtis 2 days ago [-]
We see comparable results for vectors and FTS.
For vector search we have warm and cold p99s of approx 20ms and 400ms respectively.
For FTS, warm and cold query p99s of approx 15ms and 250ms respectively.
Both of these benchmarks were run on 1m docs.
rajit 2 days ago [-]
when will the graph memory layer be available?
GeorgeCurtis 2 days ago [-]
We plan on launching end of month.
Bnjoroge 2 days ago [-]
congrats! how does this compare to turbopuffer, surreal or other multi-model ones built on object storage or not
GeorgeCurtis 2 days ago [-]
tpuffer is a vector/fts database. Surreal is a bit of an "everything database".
We're a graph database with vector and FTS capabilities. Our vector and FTS benchmarks are comparable with tpuffer, but you would primarily use us for building whole applications, knowledge graphs, or AI memory/retrieval. Anything that is relationship intense.
Let me know if this properly answers your question
Bnjoroge 18 hours ago [-]
it does, think I misunderstood your value prop. best of luck -definitely a real usecase.
raufakdemir 2 days ago [-]
what language does this support? cypher/gremlin?
GeorgeCurtis 2 days ago [-]
We don't support cypher or gremlin. We can
You can query HelixDB using JSON or directly in your programming language of choice by using our Rust, TypeScript, Go or Python SDKs.
We’ve found AI is very good at working with the SDKs and JSON itself to query, making the development experience much better than before: https://docs.helix-db.com/database/querying
As long as the sub-graph you're trying to hop is cached, then there's no problem or latency issues. However, if you need to do a deep hop query, where all those nodes and edges are in cold storage, each hop costs ~50ms. So a 10-hop would take ~0.5 seconds.
Again though, we find most people are using us for agentic workloads, so even this worst case scenario the LLMs make up the majority of the latency.
What's your p99 like for multi hops?
I'm more concerned about if the p99s stay consistent when things get spikey.
dgraph is fine otherwise...
We're fine with clickhouse and redshift for the OLAP work we do. I've been looking at ParaQuery lately if I really want to speed that up.
email us: founders@helix-db.com
We’re just two young founders sharing what we’ve been building, so I’ll take the drive-by competitor plug as a compliment :)
Definitely a different focus though. Helix is OLTP, built for operational graph + vector workloads, especially apps/agent memory where low-latency traversals and writes are concerned.
can you host this yourself or do you need to use helix-cloud? the chat thing on the side seems to push me to helix-cloud but it looks like that starts at like $600/mo which is above my experimentation budget.
looking for a db for an agent memory application and i'd probably start with something that's just self-hosted / freeish. postgres is working ok but I want to start ingesting server and chat logs.
We aim to launch our GA cloud at the end of this month, which will be much more affordable.
Soon you’ll be able to host it yourself AND have access to the source code
What reassurance can you offer devs that are hesitant to try a new data-store?
I'd encourage them to start a local instance with claude/codex to build a mini project and see what it's like.
Congrats on the launch!
We’re 100% committed to going back to open-source on an Apache 2.0 license as soon as possible. In the meantime, you can continue to deploy us completely for free, however you like, using the compiled docker container.
Looking forward to looking into the generalised AI memory layer when it comes out.
For vector search we have warm and cold p99s of approx 20ms and 400ms respectively. For FTS, warm and cold query p99s of approx 15ms and 250ms respectively.
Both of these benchmarks were run on 1m docs.
We're a graph database with vector and FTS capabilities. Our vector and FTS benchmarks are comparable with tpuffer, but you would primarily use us for building whole applications, knowledge graphs, or AI memory/retrieval. Anything that is relationship intense.
Let me know if this properly answers your question
You can query HelixDB using JSON or directly in your programming language of choice by using our Rust, TypeScript, Go or Python SDKs. We’ve found AI is very good at working with the SDKs and JSON itself to query, making the development experience much better than before: https://docs.helix-db.com/database/querying