Files
sglang/sgl-router/src/policies/cache_aware.rs

562 lines
20 KiB
Rust

/*
Cache-Aware Load Balancing Router
This router combines two strategies to optimize both cache utilization and request distribution:
1. Cache-Aware Routing (Approximate Tree)
2. Load Balancing (Shortest Queue with Balance Thresholds)
The router dynamically switches between these strategies based on load conditions:
- Uses load balancing when the system is imbalanced
- Uses cache-aware routing when the system is balanced
A system is considered imbalanced if both conditions are met:
1. (max - min) > abs_threshold
2. max > rel_threshold * min
Strategy Details:
1. Cache-Aware Routing (Approximate Tree)
-------------------------------------------
This strategy maintains an approximate radix tree for each worker based on request history,
eliminating the need for direct cache state queries. The tree stores raw text characters
instead of token IDs to avoid tokenization overhead.
Process:
a. For each request, find the worker with the highest prefix match
b. If match rate > cache_threshold:
Route to the worker with highest match (likely has relevant data cached)
c. If match rate ≤ cache_threshold:
Route to the worker with smallest tree size (most available cache capacity)
d. Background maintenance:
Periodically evict least recently used leaf nodes to prevent memory overflow
2. Load Balancing (Shortest Queue)
-------------------------------------------
This strategy tracks pending request counts per worker and routes new requests
to the least busy worker when the system is detected to be imbalanced.
Configuration Parameters:
------------------------
1. cache_threshold: (float, 0.0 to 1.0)
Minimum prefix match ratio to use highest-match routing.
Below this threshold, routes to worker with most available cache space.
2. balance_abs_threshold: (integer)
Absolute difference threshold for load imbalance detection.
System is potentially imbalanced if (max_load - min_load) > abs_threshold
3. balance_rel_threshold: (float)
Relative ratio threshold for load imbalance detection.
System is potentially imbalanced if max_load > min_load * rel_threshold
Used in conjunction with abs_threshold to determine final imbalance state.
4. eviction_interval_secs: (integer)
Interval between LRU eviction cycles for the approximate trees.
5. max_tree_size: (integer)
Maximum nodes per tree. When exceeded, LRU leaf nodes are evicted
during the next eviction cycle.
*/
use super::{get_healthy_worker_indices, CacheAwareConfig, LoadBalancingPolicy};
use crate::core::Worker;
use crate::metrics::RouterMetrics;
use crate::tree::Tree;
use std::collections::HashMap;
use std::sync::{Arc, Mutex};
use std::thread;
use std::time::Duration;
use tracing::debug;
/// Cache-aware routing policy
///
/// Routes requests based on cache affinity when load is balanced,
/// switches to shortest-queue routing when load is imbalanced.
/// Maintains separate trees per model for multi-model support.
#[derive(Debug)]
pub struct CacheAwarePolicy {
config: CacheAwareConfig,
trees: Arc<Mutex<HashMap<String, Tree>>>, // model_id -> Tree
eviction_handle: Option<thread::JoinHandle<()>>,
}
impl CacheAwarePolicy {
pub fn new() -> Self {
Self::with_config(CacheAwareConfig::default())
}
pub fn with_config(config: CacheAwareConfig) -> Self {
let trees = Arc::new(Mutex::new(HashMap::<String, Tree>::new()));
// Start background eviction thread if configured
let eviction_handle = if config.eviction_interval_secs > 0 {
let trees_clone = Arc::clone(&trees);
let max_tree_size = config.max_tree_size;
let interval = config.eviction_interval_secs;
Some(thread::spawn(move || loop {
thread::sleep(Duration::from_secs(interval));
if let Ok(mut trees_guard) = trees_clone.lock() {
// Evict for all model trees
for (model_id, tree) in trees_guard.iter_mut() {
tree.evict_tenant_by_size(max_tree_size);
debug!(
"Cache eviction completed for model {}, max_size: {}",
model_id, max_tree_size
);
}
}
}))
} else {
None
};
Self {
config,
trees,
eviction_handle,
}
}
/// Initialize the tree with worker URLs (used only during initial setup)
pub fn init_workers(&self, workers: &[Arc<dyn Worker>]) {
if let Ok(mut trees) = self.trees.lock() {
// Group workers by model
let mut model_workers: HashMap<String, Vec<&Arc<dyn Worker>>> = HashMap::new();
for worker in workers {
// Use "default" for unknown/empty model_ids for backward compatibility
let model_id = worker.model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
model_workers.entry(tree_key).or_default().push(worker);
}
// Initialize tree for each model
for (tree_key, model_workers) in model_workers {
let tree = trees.entry(tree_key).or_insert_with(Tree::new);
for worker in model_workers {
tree.insert("", worker.url());
}
}
}
}
/// Add a single worker to the tree (incremental update)
pub fn add_worker(&self, worker: &dyn Worker) {
if let Ok(mut trees) = self.trees.lock() {
// For backward compatibility: if model_id is "unknown" or empty,
// use a default tree. This preserves existing behavior for single-model routers.
let model_id = worker.model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
let tree = trees.entry(tree_key).or_insert_with(Tree::new);
tree.insert("", worker.url());
}
}
/// Add a worker by URL and model (for backward compatibility)
pub fn add_worker_by_url(&self, url: &str, model_id: &str) {
if let Ok(mut trees) = self.trees.lock() {
let tree = trees.entry(model_id.to_string()).or_insert_with(Tree::new);
tree.insert("", url);
}
}
/// Remove a worker from the tree
pub fn remove_worker(&self, worker: &dyn Worker) {
if let Ok(mut trees) = self.trees.lock() {
// Use same logic as add_worker for consistency
let model_id = worker.model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
if let Some(tree) = trees.get_mut(&tree_key) {
tree.remove_tenant(worker.url());
}
}
}
/// Remove a worker by URL (removes from all model trees for backward compatibility)
pub fn remove_worker_by_url(&self, url: &str) {
if let Ok(mut trees) = self.trees.lock() {
// Remove from all trees since we don't know which model it belongs to
for (_model_id, tree) in trees.iter_mut() {
tree.remove_tenant(url);
}
}
}
/// Run cache eviction to prevent unbounded growth
pub fn evict_cache(&self, max_size: usize) {
if let Ok(mut trees) = self.trees.lock() {
for (model_id, tree) in trees.iter_mut() {
tree.evict_tenant_by_size(max_size);
debug!(
"Cache eviction for model {}, max_size: {}",
model_id, max_size
);
}
}
}
}
impl LoadBalancingPolicy for CacheAwarePolicy {
fn select_worker(
&self,
workers: &[Arc<dyn Worker>],
request_text: Option<&str>,
) -> Option<usize> {
let healthy_indices = get_healthy_worker_indices(workers);
if healthy_indices.is_empty() {
return None;
}
// Group workers by model (using "default" for unknown/empty model_ids)
let mut model_workers: HashMap<String, Vec<usize>> = HashMap::new();
for idx in &healthy_indices {
let model_id = workers[*idx].model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
model_workers.entry(tree_key).or_default().push(*idx);
}
// Get current load statistics
let loads: Vec<usize> = workers.iter().map(|w| w.load()).collect();
let max_load = *loads.iter().max().unwrap_or(&0);
let min_load = *loads.iter().min().unwrap_or(&0);
// Check if load is imbalanced
let is_imbalanced = max_load.saturating_sub(min_load) > self.config.balance_abs_threshold
&& (max_load as f32) > (min_load as f32 * self.config.balance_rel_threshold);
if is_imbalanced {
// Log load balancing trigger
let worker_loads: Vec<(String, usize)> = workers
.iter()
.map(|w| (w.url().to_string(), w.load()))
.collect();
debug!(
"Load balancing triggered | max: {} | min: {} | workers: {:?}",
max_load, min_load, worker_loads
);
RouterMetrics::record_load_balancing_event();
RouterMetrics::set_load_range(max_load, min_load);
// Use shortest queue when imbalanced
let min_load_idx = healthy_indices
.iter()
.min_by_key(|&&idx| workers[idx].load())
.copied()?;
// Even in imbalanced mode, update the tree to maintain cache state
if let Some(text) = request_text {
if let Ok(mut trees) = self.trees.lock() {
let model_id = workers[min_load_idx].model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
let tree = trees.entry(tree_key).or_insert_with(Tree::new);
tree.insert(text, workers[min_load_idx].url());
}
}
// Increment processed counter
workers[min_load_idx].increment_processed();
RouterMetrics::record_processed_request(workers[min_load_idx].url());
RouterMetrics::record_policy_decision(self.name(), workers[min_load_idx].url());
return Some(min_load_idx);
}
// Use cache-aware routing when balanced
let text = request_text.unwrap_or("");
if let Ok(mut trees) = self.trees.lock() {
let mut best_match_idx: Option<usize> = None;
let mut best_match_rate: f32 = 0.0;
// Find best match across all models
for (model_id, worker_indices) in &model_workers {
let tree = trees.entry(model_id.clone()).or_insert_with(Tree::new);
let (matched_text, matched_worker) = tree.prefix_match(text);
let match_rate = if text.is_empty() {
0.0
} else {
matched_text.chars().count() as f32 / text.chars().count() as f32
};
// Check if this model has the best match
if match_rate > best_match_rate {
// Find the worker index for this URL
if let Some(idx) = worker_indices
.iter()
.find(|&&idx| workers[idx].url() == matched_worker)
{
best_match_idx = Some(*idx);
best_match_rate = match_rate;
}
}
}
// Select worker based on cache threshold
let selected_idx = if let (Some(idx), true) = (
best_match_idx,
best_match_rate > self.config.cache_threshold,
) {
RouterMetrics::record_cache_hit();
idx
} else {
RouterMetrics::record_cache_miss();
// Find model with smallest tree (most cache capacity)
let mut smallest_tree_model = String::new();
let mut smallest_tree_size = usize::MAX;
for model_id in model_workers.keys() {
let tree = trees.entry(model_id.clone()).or_insert_with(Tree::new);
let size = tree.get_used_size_per_tenant().values().sum::<usize>();
if size < smallest_tree_size {
smallest_tree_size = size;
smallest_tree_model = model_id.clone();
}
}
// Select least loaded worker from model with most cache capacity
if let Some(worker_indices) = model_workers.get(&smallest_tree_model) {
worker_indices
.iter()
.min_by_key(|&&idx| workers[idx].load())
.copied()
.unwrap_or(healthy_indices[0])
} else {
healthy_indices[0]
}
};
// Update the tree with this request
let model_id = workers[selected_idx].model_id();
let tree_key = if model_id.is_empty() || model_id == "unknown" {
"default".to_string()
} else {
model_id.to_string()
};
let tree = trees.entry(tree_key).or_insert_with(Tree::new);
tree.insert(text, workers[selected_idx].url());
// Increment processed counter
workers[selected_idx].increment_processed();
RouterMetrics::record_processed_request(workers[selected_idx].url());
RouterMetrics::record_policy_decision(self.name(), workers[selected_idx].url());
return Some(selected_idx);
}
// Fallback to first healthy worker if tree operations fail
healthy_indices.first().copied()
}
fn name(&self) -> &'static str {
"cache_aware"
}
fn needs_request_text(&self) -> bool {
true // Cache-aware policy needs request text for cache affinity
}
fn on_request_complete(&self, worker_url: &str, success: bool) {
// Could track success rates per worker for more intelligent routing
if !success {
// Optionally reduce affinity for failed requests
tracing::debug!(
"Request to {} completed with success={}",
worker_url,
success
);
}
}
fn as_any(&self) -> &dyn std::any::Any {
self
}
fn select_worker_pair(
&self,
prefill_workers: &[Arc<dyn Worker>],
decode_workers: &[Arc<dyn Worker>],
request_text: Option<&str>,
) -> Option<(usize, usize)> {
// DEPRECATED: This method is no longer used when separate policies are configured.
// The PD router now uses separate policies for prefill and decode selection.
// This implementation remains for backward compatibility when a single policy is used.
// In PD mode with single policy:
// - Prefill: Use cache-aware routing for better cache utilization
// - Decode: Use least-load routing for better load distribution
// Select prefill worker using cache-aware logic
let prefill_idx = self.select_worker(prefill_workers, request_text)?;
// Select decode worker using least-load logic
let healthy_decode = get_healthy_worker_indices(decode_workers);
if healthy_decode.is_empty() {
return None;
}
let decode_idx = healthy_decode
.iter()
.min_by_key(|&&idx| decode_workers[idx].load())
.copied()?;
Some((prefill_idx, decode_idx))
}
}
impl Default for CacheAwarePolicy {
fn default() -> Self {
Self::new()
}
}
impl Drop for CacheAwarePolicy {
fn drop(&mut self) {
// Note: We can't properly stop the eviction thread since it's in an infinite loop
// In a production system, we'd use a channel or atomic flag to signal shutdown
if let Some(handle) = self.eviction_handle.take() {
// The thread will continue running until the program exits
// This is acceptable for now since the router typically runs for the lifetime of the program
drop(handle);
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use crate::core::{BasicWorkerBuilder, WorkerType};
#[test]
fn test_cache_aware_with_balanced_load() {
// Create policy without eviction thread for testing
let config = CacheAwareConfig {
eviction_interval_secs: 0, // Disable eviction thread
..Default::default()
};
let policy = CacheAwarePolicy::with_config(config);
let workers: Vec<Arc<dyn Worker>> = vec![
Arc::new(
BasicWorkerBuilder::new("http://w1:8000")
.worker_type(WorkerType::Regular)
.api_key("test_api_key")
.build(),
),
Arc::new(
BasicWorkerBuilder::new("http://w2:8000")
.worker_type(WorkerType::Regular)
.api_key("test_api_key")
.build(),
),
];
// Initialize the policy with workers
policy.init_workers(&workers);
// First request should be distributed
let idx1 = policy.select_worker(&workers, Some("hello world")).unwrap();
// Same request should go to same worker (cache hit)
let idx2 = policy.select_worker(&workers, Some("hello world")).unwrap();
assert_eq!(idx1, idx2);
// Similar request should also go to same worker
let idx3 = policy.select_worker(&workers, Some("hello")).unwrap();
assert_eq!(idx1, idx3);
}
#[test]
fn test_cache_aware_with_imbalanced_load() {
let policy = CacheAwarePolicy::with_config(CacheAwareConfig {
cache_threshold: 0.5,
balance_abs_threshold: 5,
balance_rel_threshold: 2.0,
eviction_interval_secs: 0, // Disable eviction thread
max_tree_size: 10000,
});
let worker1 = BasicWorkerBuilder::new("http://w1:8000")
.worker_type(WorkerType::Regular)
.build();
let worker2 = BasicWorkerBuilder::new("http://w2:8000")
.worker_type(WorkerType::Regular)
.build();
// Create significant load imbalance
for _ in 0..20 {
worker1.increment_load();
}
// worker2 has load 0
let workers: Vec<Arc<dyn Worker>> = vec![Arc::new(worker1), Arc::new(worker2)];
policy.init_workers(&workers);
// Should select worker2 (lower load) despite cache affinity
for _ in 0..5 {
let idx = policy.select_worker(&workers, Some("test")).unwrap();
assert_eq!(idx, 1); // Should always pick worker2
}
}
#[test]
fn test_cache_aware_worker_removal() {
let config = CacheAwareConfig {
eviction_interval_secs: 0, // Disable eviction thread
..Default::default()
};
let policy = CacheAwarePolicy::with_config(config);
let workers: Vec<Arc<dyn Worker>> = vec![
Arc::new(
BasicWorkerBuilder::new("http://w1:8000")
.worker_type(WorkerType::Regular)
.build(),
),
Arc::new(
BasicWorkerBuilder::new("http://w2:8000")
.worker_type(WorkerType::Regular)
.build(),
),
];
policy.init_workers(&workers);
// Route some requests
policy.select_worker(&workers, Some("test1"));
policy.select_worker(&workers, Some("test2"));
// Remove a worker
policy.remove_worker_by_url("http://w1:8000");
workers[0].set_healthy(false);
// All requests should now go to worker2
let idx = policy.select_worker(&workers, Some("test1")).unwrap();
assert_eq!(idx, 1);
}
}