This would be a great bounty for someone to pursue while exploring these patterns at the library level. Pipelineify already handles a lot of this queuing action, and it would likely need to use some of the libraries that would help with managing input output queues.
I remember reading about throughput limitations of BTC transactions, but can’t find it. Something like one transaction per block per subnet. Does anyone know the actual number and whether there is ongoing work to increase it?
To clarify: the limit you point to is the current throughput of tECDSA signatures per second that a subnet can produce (you need those in order to submit BTC transactions so of course the limit on those sort of implies a limit on the BTC transactions themselves).
AFAIK, the current implementation of tECDSA is not super optimized, so I would expect that this limit can be increased eventually. Likely there is going to be some limit on what a single subnet can do (the protocol is quite complex) but there is always the possibility to allow more subnets to sign using tECDSA which means we can “scale out” the throughput of these operations on the IC as whole.
we’ll be moving this week’s session to next week same time:
Thursday, February 23rd 5:30pm CET (GMT+1).
I’ve already updated the calendar.
We’ll have another session on using storage on the IC. We’ll prepare a survey that we’ll share before the session and want to discuss the results in the session.
Sorry for the late notice and I hope many of you can join nevertheless.
Hey everybody ,
In preparation for the WG session this week, we’d like to invite you to take a quick survey on how you use memory on the IC:
Please participate and join the WG on Thursday where we’d like to discuss the results.
Hey everybody ,
The Working Group is happening in about 40 mins (5:30pm GMT+1).
See you soon!
Some topics from the discussion:
- Discussion on alternative encodings for serializing to stable memory (bincode and MessagePack instead of Candid)
- Sqlite on the IC. See also ICSQLite:Sqlite backed by stable memory - #17 by FroghubMan
- Discussion on data structures vs. databases on the IC
- Discussion on cycle limits
- Update on ic-stable-memory which allows dynamic key/value lengths and a certified map compatible with certified assets protocol.
If you watch the recording and have thoughts please feel free to add them here.
We have also a request to move the working group from Thursdays at 5:30 pm CET (GMT+1) to Wednesdays at 6:30 pm CET (GMT+1). Are there any objections to that? Note that we’d still use the cadance of every 3rd Wednesday of a month.
Hey @domwoe, will the move to 18:30 permanent ?
Sorry for being late on the mater but I feel it is kinda late as it is the time dedicated to the kids for me.
I really enjoy this working group and I don’t want to sounds selfish though. This is why I’m asking if this is the permanent schedule from the on.
we wanted to change the date because @abk will take over the DFINITY lead of the working group, and the time didn’t really fit for him. However, he read your response and made it work .
We won’t change the date/time of the Working Group. This means we will have the next session, this Thursday at 5:30pm CET!
For the WG tomorrow I’ll be presenting some benchmark results comparing the
ic-stable-memory libraries, as well as sharing some tips on efficiently using stable memory. This should be useful for those of you planning to store significant data in stable memory.
I will do my best to join the next meeting on Thursday, April 20th.
Have a wonderful Easter everyone, and a good Passover too.
P.S. Thanks so much for sharing the slides and videos!
I remember that we already had interactions on Twitter. Good to have you here!
Thanks, the feeling is mutual.
Have a great Holy Week, and Happy Passover too
Join us tomorrow, Thursday, April 20th at 5:30 pm CEST (GMT+2) for the next session of the Scalability & Performance WG.
We’ll have presentations from OpenChat, DSCVR, and Dragginz about their ideas and approaches toward sharding data across canisters.
Pretty exciting, right? Hope to see many of you there for a lively discussion.
Are any of these sessions recorded?
yep, the check the links in the meeting notes
we’ll move this month’s working group session by one week to May 25th because there’s a holiday in Switzerland this Thursday.
See you next week!
For the session this Thursday @senior.joinu will discuss the design of a new multi-canister storage library and we’ll give an overview of the Storage+Scaling ICP.Lab. Hope to see you there!
Details: Thursday May 25th at 5:30 pm CEST (GMT+2) Zoom link
In case somebody needs my presentation on union-db: