snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2025-11-22 04:36:32 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #5990 from meilisearch/default-max-batch-size

Browse Source 
Make the limit batched tasks total size defaults to half of the max indexing memory
This commit is contained in:
Clément Renault
2025-11-19 16:56:34 +00:00
committed by GitHub
parent 7597b1049f c9726674a0
commit 5ed9be0789
3 changed files with 20 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
crates/meilisearch/src/analytics/segment_analytics.rs
View File

@@ -195,7 +195,7 @@ struct Infos {
experimental_enable_logs_route: bool,
experimental_reduce_indexing_memory_usage: bool,
experimental_max_number_of_batched_tasks: usize,
experimental_limit_batched_tasks_total_size: u64,
experimental_limit_batched_tasks_total_size: Option<u64>,
experimental_network: bool,
experimental_multimodal: bool,
experimental_chat_completions: bool,
@@ -359,7 +359,7 @@ impl Infos {
http_payload_size_limit,
experimental_max_number_of_batched_tasks,
experimental_limit_batched_tasks_total_size:
experimental_limit_batched_tasks_total_size.into(),
experimental_limit_batched_tasks_total_size.map(|size| size.as_u64()),
task_queue_webhook: task_webhook_url.is_some(),
task_webhook_authorization_header: task_webhook_authorization_header.is_some(),
log_level: log_level.to_string(),

6
crates/meilisearch/src/lib.rs
View File

@@ -230,7 +230,11 @@ pub fn setup_meilisearch(
cleanup_enabled: !opt.experimental_replication_parameters,
max_number_of_tasks: 1_000_000,
max_number_of_batched_tasks: opt.experimental_max_number_of_batched_tasks,
batched_tasks_size_limit: opt.experimental_limit_batched_tasks_total_size.into(),
batched_tasks_size_limit: opt.experimental_limit_batched_tasks_total_size.map_or_else(
// By default, we use half of the available memory to determine the size of batched tasks
|| opt.indexer_options.max_indexing_memory.map_or(u64::MAX, |mem| mem.as_u64() / 2),
|size| size.as_u64(),
),
index_growth_amount: byte_unit::Byte::from_str("10GiB").unwrap().as_u64() as usize,
index_count: DEFAULT_INDEX_COUNT,
instance_features: opt.to_instance_features(),

20
crates/meilisearch/src/option.rs
View File

@@ -473,11 +473,13 @@ pub struct Opt {
#[serde(default = "default_limit_batched_tasks")]
pub experimental_max_number_of_batched_tasks: usize,
/// Experimentally reduces the maximum total size, in bytes, of tasks that will be processed at once,
/// see: <https://github.com/orgs/meilisearch/discussions/801>
#[clap(long, env = MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE, default_value_t = default_limit_batched_tasks_total_size())]
#[serde(default = "default_limit_batched_tasks_total_size")]
pub experimental_limit_batched_tasks_total_size: Byte,
/// Experimentally controls the maximum total size, in bytes, of tasks that will be processed
/// simultaneously. When unspecified, defaults to half of the maximum indexing memory.
///
/// See: <https://github.com/orgs/meilisearch/discussions/801>
#[clap(long, env = MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE)]
#[serde(default)]
pub experimental_limit_batched_tasks_total_size: Option<Byte>,
/// Enables experimental caching of search query embeddings. The value represents the maximal number of entries in the cache of each
/// distinct embedder.
@@ -701,10 +703,12 @@ impl Opt {
MEILI_EXPERIMENTAL_MAX_NUMBER_OF_BATCHED_TASKS,
experimental_max_number_of_batched_tasks.to_string(),
);
if let Some(limit) = experimental_limit_batched_tasks_total_size {
export_to_env_if_not_present(
MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZE,
experimental_limit_batched_tasks_total_size.to_string(),
limit.to_string(),
);
}
export_to_env_if_not_present(
MEILI_EXPERIMENTAL_EMBEDDING_CACHE_ENTRIES,
experimental_embedding_cache_entries.to_string(),
@@ -1273,10 +1277,6 @@ fn default_limit_batched_tasks() -> usize {
usize::MAX
}
fn default_limit_batched_tasks_total_size() -> Byte {
Byte::from_u64(u64::MAX)
}
fn default_embedding_cache_entries() -> usize {
0
}