- DirectX12 and Vulkan benchmark applications
- Hair rendering example (roving capsules)
- Improved included threadpool
- Bring your own threadpool. :)
...And lots more. API is still stable. Come check it out!
- DirectX12 and Vulkan benchmark applications
- Hair rendering example (roving capsules)
- Improved included threadpool
- Bring your own threadpool. :)
...And lots more. API is still stable. Come check it out!
Node caps the threadpool at 4. Nub lets this float to the number of machine cores (with conservative nicing beyond 4), increasing max throughput for crypto, zlib, DNS, and native addons like bcrypt and sharp
Node caps the threadpool at 4. Nub lets this float to the number of machine cores (with conservative nicing beyond 4), increasing max throughput for crypto, zlib, DNS, and native addons like bcrypt and sharp
github.com/rasta-mouse/...
github.com/rasta-mouse/...
Replace the built-in job system (adapted from Wicked Engine) by an alternative, using three hooks in BVHContext.
TinyBVH remains single-header and dependency-free of course.
github.com/jbikker/tiny...
Replace the built-in job system (adapted from Wicked Engine) by an alternative, using three hooks in BVHContext.
TinyBVH remains single-header and dependency-free of course.
github.com/jbikker/tiny...
- Primary / readonly replica (within the same process), with versioning
- Probabilistic sets for quick membership/follow/block lookup
- Separate threadpool for firehose indexing and web serving
- Direct IO with sector-sized reads
- Primary / readonly replica (within the same process), with versioning
- Probabilistic sets for quick membership/follow/block lookup
- Separate threadpool for firehose indexing and web serving
- Direct IO with sector-sized reads
1. You never have to configure threads/concurrency again.
2. It responds dynamically, even after the app has booted.
1. You never have to configure threads/concurrency again.
2. It responds dynamically, even after the app has booted.
Nothing fancy really, just a way of shoving arbitrary funcs (including capturing lambdas) into ordinary func pointers without needing any inheritance or allocation.
E.g. ThreadPool:
Nothing fancy really, just a way of shoving arbitrary funcs (including capturing lambdas) into ordinary func pointers without needing any inheritance or allocation.
E.g. ThreadPool:
1. Framework/lib authors can move threads into a special "high IO" pool
2. single/cluster DSL hooks for easier config
3. API for adjusting threadpool size at runtime (who would want to that? stay tuned...)
1. Framework/lib authors can move threads into a special "high IO" pool
2. single/cluster DSL hooks for easier config
3. API for adjusting threadpool size at runtime (who would want to that? stay tuned...)
Mostrei como usar o HttpClient e os Executors do Java para criar ThreadPool
Segue os links:
1. openjdk.org/groups/net/h...
2. docs.oracle.com/javase/8/doc...
Mostrei como usar o HttpClient e os Executors do Java para criar ThreadPool
Segue os links:
1. openjdk.org/groups/net/h...
2. docs.oracle.com/javase/8/doc...
batman: man.bat
deadpool: threadpool
cyclops: apple vision pro max
batman: man.bat
deadpool: threadpool
cyclops: apple vision pro max
ᅠmy amazing and wonderful sister gave me the idea to swap the tag around and I like it so much better 😭😭😭 (bsky edit feature when)
ᅠfrom now on #threadpool will be the tag for threads :)))
— DP SOLO THREAD —
[ people can still interact with their response to the thread, but the thread is centered around Wade so do with that as you will ]
TW : suicidal thoughts / tendencies, self-harm thoughts / tendencies, gore, heartbreak, and a WHOOOOLE lot of loneliness <3
ᅠmy amazing and wonderful sister gave me the idea to swap the tag around and I like it so much better 😭😭😭 (bsky edit feature when)
ᅠfrom now on #threadpool will be the tag for threads :)))
- Adicionar índices tem que ter cuidado porque pode atrasar os inserts e updates
- O paralelismo tem que ter um Threadpool bem dimensionado pra não dar OOM
- Isso se aplica mais a processos com quantidades razoáveis de dados
…mas no geral é isso aí 💨
- Adicionar índices tem que ter cuidado porque pode atrasar os inserts e updates
- O paralelismo tem que ter um Threadpool bem dimensionado pra não dar OOM
- Isso se aplica mais a processos com quantidades razoáveis de dados
…mas no geral é isso aí 💨
github.com/sarah-quinon...
command is `cargo bench --bench barrier -- --min-time=2`
github.com/sarah-quinon...
command is `cargo bench --bench barrier -- --min-time=2`
- various O(n) footguns
- stale ClickHouse keepalive sockets
- libuv DNS threadpool starvation
- silly and wrong constants causing trouble
- plc mirroring stalled for 9 hours
- various O(n) footguns
- stale ClickHouse keepalive sockets
- libuv DNS threadpool starvation
- silly and wrong constants causing trouble
- plc mirroring stalled for 9 hours
The test works but it'll be fun messing with a mutex (or semaphore). The .NET source uses a (the?) threadpool.
The test works but it'll be fun messing with a mutex (or semaphore). The .NET source uses a (the?) threadpool.