`module Smooth`

Enforces even task distribution to prevent clustering.

Smooth strategies prevent clustering by requiring tasks to wait for frame boundaries, ensuring smooth and predictable task execution timing. This creates consistent load patterns and prevents resource spikes.

Smooth Behavior

Smooth behavior with 3 tasks per 15-second window:

Window 1: [Task1] -- [Task2] -- [Task3] ---- (evenly spaced) 0s 5s 10s 15s Window 2: [Task4] -- [Task5] -- [Task6] ---- (evenly spaced) 15s 20s 25s 30s

Perfect for: API rate limiting, smooth resource usage

Definitions

`def self.can_acquire?(window_count, limit, frame_changed)`

Check if a task can be acquired in continuous mode.

Signature

parameter　window_count　Integer: Ignored in continuous mode.
parameter　limit　Integer: Ignored in continuous mode.
parameter　frame_changed　Boolean: Whether the frame boundary has been crossed.
returns　Boolean: True only if the frame boundary has been crossed.

Implementation

def self.can_acquire?(window_count, limit, frame_changed)
	frame_changed
end

`def self.next_acquire_time(window_start_time, window_duration, frame_start_time, frame_duration)`

Calculate the next time a task can be acquired.

Signature

parameter　window_start_time　Numeric: Ignored in continuous mode.
parameter　window_duration　Numeric: Ignored in continuous mode.
parameter　frame_start_time　Numeric: When the current frame started.
parameter　frame_duration　Numeric: Duration of each frame.
returns　Numeric: The next frame start time.

Implementation

def self.next_acquire_time(window_start_time, window_duration, frame_start_time, frame_duration)
	frame_start_time + frame_duration
end

`def self.window_blocking?(window_count, limit, window_changed)`

Check if window constraints are blocking new tasks.

Signature

parameter　window_count　Integer: Ignored in continuous mode.
parameter　limit　Integer: Ignored in continuous mode.
parameter　window_changed　Boolean: Ignored in continuous mode.
returns　Boolean: Always false for continuous mode.

Implementation

def self.window_blocking?(window_count, limit, window_changed)
	false  # Continuous mode doesn't use window blocking
end

`def self.frame_blocking?(frame_changed)`

Check if frame constraints are blocking new tasks.

Signature

parameter　frame_changed　Boolean: Whether the frame boundary has been crossed.
returns　Boolean: True if frame hasn't changed (blocking until next frame).

Implementation

def self.frame_blocking?(frame_changed)
	!frame_changed
end

`def self.statistics`

Get current burst strategy statistics.

Signature

returns　Hash: Statistics hash with current state.

Implementation

def self.statistics
	{
		name: "Smooth",
	}
end