TL (@tlefky) profile picture
Joined June 2011
Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)Photo by TL (@tlefky)
All charts like these should come with the y-axis
The best mornings start with a latency going down. Rolling out changes to our runtime to significantly reduce tail latency. There are *so* many opportunities for solving difficult technical problems at Convex. Come work here. It's hard but fun.
1
1
5
736
RT @spacetimedb: Damn it feels good to drop that understore.
2
11
Thanks to everyone who attended the first SpacetimeDB Launchpad hackathon in NYC! Congrats to our winners! Come see what they built! 1/
2
5
3
28
6,619
This is what people tell themselves to feel better about their AWS bill. You can offer your customers better services at lower prices if you tackle the hard problems. Just ask @dhh.
Infra is hard to design and build, which is why almost everyone outsources it. Even Convex uses AWS despite having built billion-dollar bare metal infra before. It's not currently vibe-codable and would be a waste of time for most businesses to do so anyway.
1
11
855
This is a dangerous and fundamental misunderstanding of how SpacetimeDB works. By default, SpacetimeDB waits for all data to be durably persisted to disk BEFORE exposing it to clients. The window in which you can lose data not "very small", it's absolutely 0. SpacetimeDB can be configured to expose data to clients before data is persisted, but so can Postgres (synchronous_commit = off). The genius of SpacetimeDB is that although we fsync asynchronously, we buffer all client reads (and write acks) until we've guaranteed they are persisted to disk. Not only does this guarantee durability of transactions from the perspective of clients, it also allows us to decouple fsync latency and fsync throughput. Pipelining is a powerful optimization. medium.com/@mihir20/understa… That being said, it's understandable how people who spent their careers just operating Postgres instead of designing their own database engine could be so confused.
"The WAL is asynchronous, and is flushed periodically to disk on the background (by default, every 50ms)." these kinds optimizations are appealing because the "window" at which you could lose data is very small. usually, though, production workloads are continuously writing data. when continuously writing, you're actually *guaranteed* to lose *some* data each time your host loses power. this is generally not desirable.
11
4
6
178
73,119
Refs carrying Caitlin Clark to the championship game
549
7,058
859
31,927
2,068,457
Check out our Official Gameplay Trailer for BitCraft! youtube.com/watch?v=lnwPaLml…
490
SpacetimeDB Elevator Pitch!
8
21
1
53
5,233
I was quoted this fare online and requested to be rebooked. This fare was then not honored and the agent tried to tell me it was my fault for not answering my phone (when no call was ever made to my phone). I haven’t been able to get in touch with a manager to discuss. @FlySWISS
1
1
442
I did, but nobody responded
Replying to @tlefky @SinemiaApp
Hi, if you could please send us a DM including your registered e-mail address, we would be happy to provide assistance right away.
2
I don't need to date, the Mets consistently break my heart. That's enough for me.
1
1
The worst thing about California officially is that Game of Thrones isn't on TV until 9 PM pacific time #cantwait#brutal
2
I guess I should be happy that the only days it's cold out in LA are the ones during finals #shouldbestudying
My Game of Thrones match is Jon Snow. Who's Yours? spotify-gameofthrones.com/sh… #SpotifyGameOfThrones
THIS JUST IN: Mets and Yoenis Cespedes reach agreement on a 3-year deal worth $75 million (via multiple reports).
135
3,100
4,062