Struct GcSchedulerOptions 

pub struct GcSchedulerOptions {

    pub enable: bool,
    pub max_concurrent_tables: usize,
    pub max_retries_per_region: usize,
    pub region_gc_concurrency: usize,
    pub retry_backoff_duration: Duration,
    pub min_region_size_threshold: u64,
    pub sst_count_weight: f64,
    pub file_removed_count_weight: f64,
    pub gc_cooldown_period: Duration,
    pub regions_per_table_threshold: usize,
    pub mailbox_timeout: Duration,
    pub full_file_listing_interval: Duration,
    pub tracker_cleanup_interval: Duration,
}

Configuration for GC operations.

TODO(discord9): not expose most config to users for now, until GC scheduler is fully stable.

Fields

enable: bool

Whether GC is enabled. Default to false. If set to false, no GC will be performed, and potentially some files from datanodes will never be deleted.

max_concurrent_tables: usize

Maximum number of tables to process concurrently.

max_retries_per_region: usize

Maximum number of retries per region when GC fails.

region_gc_concurrency: usize

Concurrency for region GC within a table.

retry_backoff_duration: Duration

Backoff duration between retries.

min_region_size_threshold: u64

Minimum region size threshold for GC (in bytes).

sst_count_weight: f64

Weight for SST file count in GC scoring.

file_removed_count_weight: f64

Weight for file removal rate in GC scoring.

gc_cooldown_period: Duration

Cooldown period between GC operations on the same region.

regions_per_table_threshold: usize

Maximum number of regions to select for GC per table.

mailbox_timeout: Duration

Timeout duration for mailbox communication with datanodes.

full_file_listing_interval: Duration

Interval for performing full file listing during GC to find orphan files. Full file listing is expensive but necessary to clean up orphan files. Set to a larger value (e.g., 24 hours) to balance performance and cleanup. Every Nth GC cycle will use full file listing, where N = full_file_listing_interval / TICKER_INTERVAL.

tracker_cleanup_interval: Duration

Interval for cleaning up stale region entries from the GC tracker. This removes entries for regions that no longer exist (e.g., after table drops). Set to a larger value (e.g., 6 hours) since this is just for memory cleanup.

Implementations

impl GcSchedulerOptions

pub fn validate(&self) -> Result<()>

Validates the configuration options.

Trait Implementations

impl Clone for GcSchedulerOptions

fn clone(&self) -> GcSchedulerOptions

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for GcSchedulerOptions

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for GcSchedulerOptions

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for GcSchedulerOptions

where GcSchedulerOptions: Default,

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for GcSchedulerOptions

fn eq(&self, other: &GcSchedulerOptions) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for GcSchedulerOptions

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for GcSchedulerOptions