https://github.com/DiceDB/dice/blob/0e241a9ca253f17b4d364cdf... defines func ExpandID, which reads from cycleMap without locking the package-global mutex; and func NextID, which writes to cycleMap under a lock of the package-global mutex. So writes are synchronized, but only between each other, and not with reads, so concurrent calls to ExpandID and NextID would race.
This is all fine as a hobby project or whatever, but very far from any kind of production-capable system.
Looking at the diceDB code base, I have few questions regarding its design, I'm asking this to understand the project's goals and design rationale. Anyone feel free to help me understand this.
I could be wrong but the primary in-memory storage appears to be a standard Go map with locking. Is this a temporary choice for iterative development, and is there a longer-term plan to adopt a more optimized or custom data structure ?
I find the DiceDB's reactivity mechanism very intriguing, particularly the "re-execution" of the entire watch command (i.e re-running GET.WATCH mykey on key modification), it's an intriguing design choice.
From what I understand is the Eval func executes client side commands this seem to be laying foundation for more complex watch command that can be evaluated before sending notifications to clients.
But I have the following question.
What is the primary motivation behind re-executing the entire command, as opposed to simply notifying clients of a key change (as in Redis Pub/Sub or streams)? Is the intent to simplify client-side logic by handling complex key dependencies on the server?
Given that re-execution seems computationally expensive, especially with multiple watchers or more complex (hypothetical) watch commands, how are potential performance bottlenecks addressed?
How does this "re-execution" approach compare in terms of scalability and consistency to more established methods like server-side logic (e.g., Lua scripts in Redis) or change data capture (CDC) ?
Are there plans to support more complex watch commands beyond GET.WATCH (e.g. JSON.GET.WATCH), and how would re-execution scale in those cases?
I'm curious about the trade-offs considered in choosing this design and how it aligns with the project's overall goals. Any insights into these design decisions would help me understand its use-cases.
I've seen this more and more with software landing pages, they are somehow so deep into developing/marketing that they totally forget to say what the thing actually is or does, that's why you show it to family and friends first to get some fresh eyes before publishing the site.
In a similar vein, lots of software is Mac-only, but omits to say this anywehere. You just get to the downloads page and see that there are only mac packages.
Looks like a Redis clone. The benchmarks compare it to Redis.
Description from GitHub:
> DiceDB is an open-source, fast, reactive, in-memory database optimized for modern hardware. Commonly used as a cache, it offers a familiar interface while enabling real-time data updates through query subscriptions. It delivers higher throughput and lower median latencies, making it ideal for modern workloads.
Not 100% a Redis clone, but the API appears to be very similar to Redis of 10 years ago, with some additions that Redis doesn't have. See the list of commands: https://dicedb.io/get-started/installation/
DiceDB is an in-memory database that is also reactive. So, instead of polling the database for changes, the database pushes the resultset if you subscribe to it.
We have a similar set of commands as Redis, but are not Redis-compliant.
Would "key-value" not have a place in the description?
This application may be very capable, but I agree with the person saying that its use-case isn't clear on the home page, you have to go deeper into the docs. "Smarter than a database" also seems kind of debatable.
When I ctrl+F the landing page for key and value, I find nothing. Reading it in full, I also come up empty handed. Which part of the landing page implies it's a key value store?
IMO, replace "More than a Cache.
Smarter than a Database." with an actual description.
The saying is cute but does not really convey information the reader is after. And that spot is where you want people to immediately understand what it is.
Still not clear to me what it is. Only the features it has, without knowing what it is.
Like, imagine a page that only said "SuperTransport -- 0 to 100 in 5 seconds", but it is not clear for the reader if it is a car or a horse or a plane or a parcel service...
... and the reader has to go and guess "hmm, guess due to the acceleration it is probably a car or a motorbike -- wonder of it is for sale or for rent?".
Just put "fast on premise key/value database" in the big font that was there -- if that is what it is. That is purely a guess from me, no idea if that is what it is.
Even clicking through to the Github, after reading the "What is DiceDB?", I'm still not very clear. It feels more like marketing than information.
"What is DiceDB?
DiceDB is an open-source, fast, reactive, in-memory database optimized for modern hardware. Commonly used as a cache, it offers a familiar interface while enabling real-time data updates through query subscriptions. It delivers higher throughput and lower median latencies, making it ideal for modern workloads."
From the benchmarks on 4vCPU and num_clients=4, the numbers doesn't look much different.
Reactive looks promising, doesn't look much useful in realworld for a cache.
For example, a client subscribes for something and the machines goes down, what happens to reactivity?
UPD Nevermind, I didn't have my eyes open. Sorry for the confusion.
Something I still fail to understand is where you can actually spend 20ms while answering a GET request in a RAM keyvalue storage (unless you implement it in Java).
I never gained much experience with existing opensource implementations, but when I was building proprietary solutions at my previous workplace, the in-memory response time was measured in tens-hundreds of microseconds. The lower bound of latency is mostly defined by syscalls so using io_uring should in theory result in even better timings, even though I never got to try it in production.
If you read from nvme AND also do the erasure-recovery across 6 nodes (lrc-12-2-2) then yes, you got into tens of milliseconds. But seeing these numbers for a single node RAM DB just doesn't make sense and I'm surprised everyone treats them as normal.
Does anyone has experience with low-latency high-throughput opensource keyvalue storages? Any specific implementation to recommend?
Different tool. I metrics I am optimizing for are different hence wrote a separate utility. May not be the most optimized one. But I am usign this to measure all things DiceDB and will be using this to optimize DiceDB further.
Based on this thread, I'm not sure you would want to use this over keyspace notifications, but I will also say that there comes a point in the maturity of a system when keyspace notifications become a complicated, unreliable, resource-heavy nightmare. They work fine is your needs and scale are limited, but it's definitely not what you want if handling lots of frequent chances across craploads of keys, with complicated logic for who needs them and how they get routed to them, and where it matters if the notification is successfully received.
But certainly you could build something to handle these and most other needs in this realm with mostly just redis, using streams for what needs to be more robust, in tandem with pub/sub, keyspace notifs, etc. in the areas they are suited to.
I didn't see it in the docs, but I'd want to know the delivery semantics of the pubsub before using this in production. I assume best effort / at most once? Any retries? In what scenarios will the messages be delivered or fail to be delivered?
From what I looked at in the past, they seem better on paper by comparing themselves to a very old version of Redis in a rigged scenario (no clustering or multithreading applied despite Drangonfly getting multithreading enabled), and they are a lot worse in terms of code updates. Maybe that's different today, but I'm more keen on using Valkey.
Does Redis support multithreading? Doesn't it use a single-threaded event loop, while DragonflyDB basic version is with multithreading enabled and shared-nothing architecture.
Also I found this latest comparison between Valkey and DragonflyDB : https://www.dragonflydb.io/blog/dragonfly-vs-valkey-benchmar...
Valkey/Redis support offloading of io processing to special I/O threads.
Their goal is to unload the "main" thread from performing i/o related tasks like socket reading and parsing, so it could only spend its precious time on datastore operations. This creates an asymmetrical architecture with I/O threads scaling to any number of CPUs, but the main thread is the only one that touches the hashtable and its entries. It helps a lot in cases where datastore operations are relatively lightweight, like SET/GET with short string values, but its impact will be insignificant for CPU heavy operations like lua EVALs, sorted sets, lists, MGET/MSET etc.
IO multithreading is still not fully there, there were significant improvements within the first couple of iterations, hopefully, it will improve further. I see that Dragonfly uses iouring, which is not recommended by Google due to security vulnerabilities.
Dragonfly supports both epoll and iouring, and polling engine choice is quite orthogonal to its shared nothing architecture. I do not think that Valkey or Redis will become fully multi-threaded any time soon - as such change will require building something like Dragonfly (or use locks that historically were a big NO for Redis).
I love the "Follow on twitter" link with the old logo and everything, they probably used a template that hasn't been updated recently but I'm choosing to believe it's actually a subtle sign of protest or resistance.
What are some example use cases where having the ability for the database to push updates to an application would be helpful (vs. the traditional polling approach)?
One example is when you want to display live data on a website. Could be a dashboard, a chat, or really the whole site. Polling is both slower and more resource hungry.
If it is built into your language/framework, you can completely ignore the problem of updating the client, as it happens automatically.
Snapshot functionality is WIP, which can be utilised to persist and replay data between reboots.
For now Golang SDK is only one, more SDKs are to be added soon.
In-memory caches (lacking persistence) shouldn't be called a database. It's not totally incorrect, but it's an abuse of terminology. Why is a Python dictionary not an in-memory key-value database?
The benchmark tool is different. I mentioned the same on my benchmark page.
We had to write a small benchmark utility (membench) ourselves because the long-term metrics that we are optimizing need to be evaluated in a different way.
Also, the scripts, utilities, and infra configurations are mentioned. Feel free to run it.
There are _so many_ bugs in this code.
One example among many:
https://github.com/DiceDB/dice/blob/0e241a9ca253f17b4d364cdf... defines func ExpandID, which reads from cycleMap without locking the package-global mutex; and func NextID, which writes to cycleMap under a lock of the package-global mutex. So writes are synchronized, but only between each other, and not with reads, so concurrent calls to ExpandID and NextID would race.
This is all fine as a hobby project or whatever, but very far from any kind of production-capable system.
Haven't looked at the code, but enforcing mutual exclusion between writers but not readers can make sense for a single-writer lock-free algorithm.
> single-writer lock-free algorithm
I understand the need for correct lock-free impls: Given OP's description, simply avoiding read mutexes can't be the way to go about it?
Looking at the diceDB code base, I have few questions regarding its design, I'm asking this to understand the project's goals and design rationale. Anyone feel free to help me understand this.
I could be wrong but the primary in-memory storage appears to be a standard Go map with locking. Is this a temporary choice for iterative development, and is there a longer-term plan to adopt a more optimized or custom data structure ?
I find the DiceDB's reactivity mechanism very intriguing, particularly the "re-execution" of the entire watch command (i.e re-running GET.WATCH mykey on key modification), it's an intriguing design choice.
From what I understand is the Eval func executes client side commands this seem to be laying foundation for more complex watch command that can be evaluated before sending notifications to clients.
But I have the following question.
What is the primary motivation behind re-executing the entire command, as opposed to simply notifying clients of a key change (as in Redis Pub/Sub or streams)? Is the intent to simplify client-side logic by handling complex key dependencies on the server?
Given that re-execution seems computationally expensive, especially with multiple watchers or more complex (hypothetical) watch commands, how are potential performance bottlenecks addressed?
How does this "re-execution" approach compare in terms of scalability and consistency to more established methods like server-side logic (e.g., Lua scripts in Redis) or change data capture (CDC) ?
Are there plans to support more complex watch commands beyond GET.WATCH (e.g. JSON.GET.WATCH), and how would re-execution scale in those cases?
I'm curious about the trade-offs considered in choosing this design and how it aligns with the project's overall goals. Any insights into these design decisions would help me understand its use-cases.
Thanks
Is there a single sentence anywhere that describes what it actually is?
I've seen this more and more with software landing pages, they are somehow so deep into developing/marketing that they totally forget to say what the thing actually is or does, that's why you show it to family and friends first to get some fresh eyes before publishing the site.
In a similar vein, lots of software is Mac-only, but omits to say this anywehere. You just get to the downloads page and see that there are only mac packages.
As if nobody ever uses anything else.
Why should they care about non-users. Offering our even mentioning choice only creates uncertainty and confusion in potential customers.
Looks like a Redis clone. The benchmarks compare it to Redis.
Description from GitHub:
> DiceDB is an open-source, fast, reactive, in-memory database optimized for modern hardware. Commonly used as a cache, it offers a familiar interface while enabling real-time data updates through query subscriptions. It delivers higher throughput and lower median latencies, making it ideal for modern workloads.
Not 100% a Redis clone, but the API appears to be very similar to Redis of 10 years ago, with some additions that Redis doesn't have. See the list of commands: https://dicedb.io/get-started/installation/
Arpit here.
DiceDB is an in-memory database that is also reactive. So, instead of polling the database for changes, the database pushes the resultset if you subscribe to it.
We have a similar set of commands as Redis, but are not Redis-compliant.
Would "key-value" not have a place in the description?
This application may be very capable, but I agree with the person saying that its use-case isn't clear on the home page, you have to go deeper into the docs. "Smarter than a database" also seems kind of debatable.
This is a lot clearer than any information I found anywhere else. There wasn't any room on your website, README, or docs for this summary?
This is a common enough pattern that it should have a name, where the submitted link isn't clear, but a single comment on HN is.
It is right there on the landing page. But, let me highlight it a bit.
When I ctrl+F the landing page for key and value, I find nothing. Reading it in full, I also come up empty handed. Which part of the landing page implies it's a key value store?
They did not say anything about key/value in their message.
You are absolutely right, my bad.
IMO, replace "More than a Cache. Smarter than a Database." with an actual description.
The saying is cute but does not really convey information the reader is after. And that spot is where you want people to immediately understand what it is.
I changed that :) now the value proposition is right at the top.
Still not clear to me what it is. Only the features it has, without knowing what it is.
Like, imagine a page that only said "SuperTransport -- 0 to 100 in 5 seconds", but it is not clear for the reader if it is a car or a horse or a plane or a parcel service...
... and the reader has to go and guess "hmm, guess due to the acceleration it is probably a car or a motorbike -- wonder of it is for sale or for rent?".
Just put "fast on premise key/value database" in the big font that was there -- if that is what it is. That is purely a guess from me, no idea if that is what it is.
So like RethinkDB? https://rethinkdb.com/
Not a month goes by where I don’t remember it at least once and realize that I still miss it.
This seems more like Redis though
It kinda surprising it was never really continued, but performance was just too bad even if the interface was fantastic.
Why don't you run the open source version?
I did for about a year and the issue is that ORMs have issues and maintainers don't feel the need to make changes.
In the list of things that DiceDB is at the top, you should add "an in-memory database". Pretty critical thing to leave out right at the top.
in-memory key-value store seems much more accurate
A Redis-inspired server in Go
Can't wait to feel the impact of garbage collection in my fast cache!
We had a similar thought, but it is not as bad as we think.
We have the benchmarks, and we will be sharing the numbers in subsequent releases.
But, there is still a chance that I may come to bite us and limit us to a smaller scale, and we are ready for it.
Vertical scaling this language also gets into painful territory quite often, I’ve had to workaround this problem before but never with a thing that felt like this: https://github.com/tailscale/tailscale/blob/main/syncs/shard...
it might help to add 99th percentile numbers to the landing page; would do a better job of showing GC impact.
Nope. it started as Redis clone. We are on a different trajectory now. Chasing different goals.
> Chasing different goals.
What are those goals? I was struggling to interpret a meaningful roadmap from the issue & commit history.
Secret goals are no selling point.
No. I had the exact same problem.
Feels arrogant. "Of course you already know what this is, how could you not?"
The docs do, the site is useless.
> DiceDB is an open-source, fast, reactive, in-memory database optimized for modern hardware.
A Redis-like database with a Redis-like interface. No info about drop-in compatibility, I assume no.
Even clicking through to the Github, after reading the "What is DiceDB?", I'm still not very clear. It feels more like marketing than information.
"What is DiceDB? DiceDB is an open-source, fast, reactive, in-memory database optimized for modern hardware. Commonly used as a cache, it offers a familiar interface while enabling real-time data updates through query subscriptions. It delivers higher throughput and lower median latencies, making it ideal for modern workloads."
Drop in replacement of Redis.
Nope. We are not redis compliant.
seems like a key store, with an ability to watch/subscribe to monitor for the change of values in real time
Yes. With DiceDB clients can "WATCH" the output of the commands and upon the change in data, the resultset are streamed to the subscribers.
"A key store, with an ability to watch/subscribe to monitor for the change of values in real time."
Should be the first sentence on their website and repo.
From the benchmarks on 4vCPU and num_clients=4, the numbers doesn't look much different.
Reactive looks promising, doesn't look much useful in realworld for a cache. For example, a client subscribes for something and the machines goes down, what happens to reactivity?
DiceDB sounds like the name of a joke database that returns random results.
No it doesn't.
Something I still fail to understand is where you can actually spend 20ms while answering a GET request in a RAM keyvalue storage (unless you implement it in Java).
I never gained much experience with existing opensource implementations, but when I was building proprietary solutions at my previous workplace, the in-memory response time was measured in tens-hundreds of microseconds. The lower bound of latency is mostly defined by syscalls so using io_uring should in theory result in even better timings, even though I never got to try it in production.
If you read from nvme AND also do the erasure-recovery across 6 nodes (lrc-12-2-2) then yes, you got into tens of milliseconds. But seeing these numbers for a single node RAM DB just doesn't make sense and I'm surprised everyone treats them as normal.
Does anyone has experience with low-latency high-throughput opensource keyvalue storages? Any specific implementation to recommend?
> Something I still fail to understand is where you can actually spend 20ms
Aren’t these numbers .2 ms, ie 200 microseconds?
They also sounded fishy to me. I'd expect closer to 10x as much throughput with Redis: https://redis.io/docs/latest/operate/oss_and_stack/managemen...
I think it is fishy based on this - https://dzone.com/articles/performance-and-scalability-analy...
Looks like your units are in ms, so 0.20 ms.
oh thank you, it's just me being blind
Using an instrument of chance to name a data store technology is pretty amusing to me.
This is essentially what all in-memory data stores have always been
Kinda refreshing to see someone own it and run with it
No chance if we live in a deterministic universe.
This seems orders of magnitude slower than Nubmq which was posted yesterday: https://news.ycombinator.com/item?id=43371097
Different tool. I metrics I am optimizing for are different hence wrote a separate utility. May not be the most optimized one. But I am usign this to measure all things DiceDB and will be using this to optimize DiceDB further.
ref: https://github.com/DiceDB/membench
Why would I use this over keyspace notifications in redis?
Based on this thread, I'm not sure you would want to use this over keyspace notifications, but I will also say that there comes a point in the maturity of a system when keyspace notifications become a complicated, unreliable, resource-heavy nightmare. They work fine is your needs and scale are limited, but it's definitely not what you want if handling lots of frequent chances across craploads of keys, with complicated logic for who needs them and how they get routed to them, and where it matters if the notification is successfully received.
But certainly you could build something to handle these and most other needs in this realm with mostly just redis, using streams for what needs to be more robust, in tandem with pub/sub, keyspace notifs, etc. in the areas they are suited to.
I didn't see it in the docs, but I'd want to know the delivery semantics of the pubsub before using this in production. I assume best effort / at most once? Any retries? In what scenarios will the messages be delivered or fail to be delivered?
I feel like this needs a ‘Why DiceDB instead of Redis or Valtio’ section prominently on the homepage.
Did you mean Valkey, or has the js community now managed to shoehorn an entire high-availability database server into a javascript object proxy?
Any reason to use this over Valkey, which is now faster than Redis and community driven? Genuinely interested.
DragonflyDB is also in that race, isn't it?
From what I looked at in the past, they seem better on paper by comparing themselves to a very old version of Redis in a rigged scenario (no clustering or multithreading applied despite Drangonfly getting multithreading enabled), and they are a lot worse in terms of code updates. Maybe that's different today, but I'm more keen on using Valkey.
Does Redis support multithreading? Doesn't it use a single-threaded event loop, while DragonflyDB basic version is with multithreading enabled and shared-nothing architecture. Also I found this latest comparison between Valkey and DragonflyDB : https://www.dragonflydb.io/blog/dragonfly-vs-valkey-benchmar...
Valkey/Redis support offloading of io processing to special I/O threads.
Their goal is to unload the "main" thread from performing i/o related tasks like socket reading and parsing, so it could only spend its precious time on datastore operations. This creates an asymmetrical architecture with I/O threads scaling to any number of CPUs, but the main thread is the only one that touches the hashtable and its entries. It helps a lot in cases where datastore operations are relatively lightweight, like SET/GET with short string values, but its impact will be insignificant for CPU heavy operations like lua EVALs, sorted sets, lists, MGET/MSET etc.
IO multithreading is still not fully there, there were significant improvements within the first couple of iterations, hopefully, it will improve further. I see that Dragonfly uses iouring, which is not recommended by Google due to security vulnerabilities.
Dragonfly supports both epoll and iouring, and polling engine choice is quite orthogonal to its shared nothing architecture. I do not think that Valkey or Redis will become fully multi-threaded any time soon - as such change will require building something like Dragonfly (or use locks that historically were a big NO for Redis).
(Author of Dragonfly here)
I read Google is limitting the use of io_uring, but I have seen io_uring being used in other Databases, TigerBeetle is another DB which uses io_uring.
I love the "Follow on twitter" link with the old logo and everything, they probably used a template that hasn't been updated recently but I'm choosing to believe it's actually a subtle sign of protest or resistance.
Just use Bluesky. It’s the better middle finger.
I prefer that over X icon.
What are some example use cases where having the ability for the database to push updates to an application would be helpful (vs. the traditional polling approach)?
One example is when you want to display live data on a website. Could be a dashboard, a chat, or really the whole site. Polling is both slower and more resource hungry.
If it is built into your language/framework, you can completely ignore the problem of updating the client, as it happens automatically.
Hope that makes sense.
Interesting -- is that normally done with database updates + polling vs. something purpose-built?
Not sure how many such solutions there are out there so no idea about the norm. I doubt polling is a real option.
You may want to search for realtime databases.
Is Arpit is the system design course guy?
Yes. I do run a sys design course on weekends.
Is this suffering from the same problems like Redis when trying to horizontally scale?
I guess yes.
I like it!
Anyway to persist data in case of reboots?
That's the only thing missing here.
Is Go the only SDK ?
Snapshot functionality is WIP, which can be utilised to persist and replay data between reboots. For now Golang SDK is only one, more SDKs are to be added soon.
- proudly open source. cool! - join discord. YAY :(
Who is this for? Can you help me explain why and when I'd want to use this in place of redis/dragonfly
In-memory caches (lacking persistence) shouldn't be called a database. It's not totally incorrect, but it's an abuse of terminology. Why is a Python dictionary not an in-memory key-value database?
I think Postgres can do everything this does and better if you use LISTEN/NOTIFY.
DiceDB is an in-memory, multi-threaded key-value DBMS that supports the Redis protocol.
It’s written in Go.
nope. We do not support Redis protocol :)
Did you remove support? Cause Google found mentions of it on your website.
Heh. Redis protocol support is still listed on their Linkedin. https://www.google.com/search?q=%22DiceDB%22%20%22supports%2...
I think performance benchmark you have done for DiceDB is fake.
These are the real numbers - https://dzone.com/articles/performance-and-scalability-analy...
Does not match with your benchmarks.
The benchmark tool is different. I mentioned the same on my benchmark page.
We had to write a small benchmark utility (membench) ourselves because the long-term metrics that we are optimizing need to be evaluated in a different way.
Also, the scripts, utilities, and infra configurations are mentioned. Feel free to run it.