Spaces:
Paused
Paused
File size: 17,218 Bytes
fb867c3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 |
"""
Multi-Server LM Studio Client Pool for Felix Framework.
This module enables true parallel processing by allowing agents to use
different LM Studio servers and models simultaneously, removing the
bottleneck of a single server.
Key Features:
- Multiple LM Studio server support
- Agent-type to server/model mapping
- Load balancing and health checks
- Failover and fault tolerance
- Performance monitoring per server
Usage:
pool = LMStudioClientPool("config/server_config.json")
response = await pool.complete_for_agent_type("research", system_prompt, user_prompt)
"""
import json
import time
import asyncio
import logging
from typing import Dict, List, Optional, Any, Tuple
from dataclasses import dataclass
from enum import Enum
from pathlib import Path
from .lm_studio_client import LMStudioClient, LLMResponse, RequestPriority, LMStudioConnectionError
logger = logging.getLogger(__name__)
class LoadBalanceStrategy(Enum):
"""Load balancing strategies for server selection."""
ROUND_ROBIN = "round_robin"
LEAST_BUSY = "least_busy"
FASTEST_RESPONSE = "fastest_response"
AGENT_TYPE_MAPPING = "agent_type_mapping"
@dataclass
class ServerConfig:
"""Configuration for a single LM Studio server."""
name: str
url: str
model: str
timeout: float = 120.0
max_concurrent: int = 4
weight: float = 1.0 # For weighted load balancing
enabled: bool = True
@dataclass
class ServerStats:
"""Runtime statistics for a server."""
total_requests: int = 0
total_tokens: int = 0
total_response_time: float = 0.0
current_load: int = 0 # Current active requests
last_health_check: float = 0.0
health_status: bool = True
average_response_time: float = 0.0
def update_stats(self, tokens: int, response_time: float):
"""Update server statistics."""
self.total_requests += 1
self.total_tokens += tokens
self.total_response_time += response_time
self.average_response_time = self.total_response_time / self.total_requests
class LMStudioClientPool:
"""
Pool of LM Studio clients for multi-server parallel processing.
Manages multiple LM Studio servers, assigns requests to appropriate
servers based on agent types, and provides load balancing and
failover capabilities.
"""
def __init__(self, config_path: Optional[str] = None, debug_mode: bool = False):
"""
Initialize the client pool.
Args:
config_path: Path to server configuration JSON file
debug_mode: Enable verbose debug output
"""
self.debug_mode = debug_mode
self.config_path = config_path
# Server management
self.servers: Dict[str, ServerConfig] = {}
self.clients: Dict[str, LMStudioClient] = {}
self.stats: Dict[str, ServerStats] = {}
# Agent type mapping
self.agent_mappings: Dict[str, str] = {}
# Load balancing
self.load_balance_strategy = LoadBalanceStrategy.AGENT_TYPE_MAPPING
self._round_robin_index = 0
# Health monitoring
self._health_check_interval = 30.0 # seconds
self._last_global_health_check = 0.0
# Load configuration
if config_path and Path(config_path).exists():
self.load_config(config_path)
else:
self._create_default_config()
def _create_default_config(self):
"""Create default single-server configuration."""
default_server = ServerConfig(
name="default",
url="http://localhost:1234/v1",
model="local-model",
timeout=120.0,
max_concurrent=4
)
self.servers["default"] = default_server
self.clients["default"] = LMStudioClient(
base_url=default_server.url,
timeout=default_server.timeout,
max_concurrent_requests=default_server.max_concurrent,
debug_mode=self.debug_mode
)
self.stats["default"] = ServerStats()
# Default mapping: all agent types use default server
self.agent_mappings = {
"research": "default",
"analysis": "default",
"synthesis": "default",
"critic": "default"
}
if self.debug_mode:
print("๐ง Using default single-server configuration")
def load_config(self, config_path: str):
"""
Load server configuration from JSON file.
Args:
config_path: Path to configuration file
"""
try:
with open(config_path, 'r') as f:
config_data = json.load(f)
# Load servers
for server_data in config_data.get("servers", []):
server_config = ServerConfig(
name=server_data["name"],
url=server_data["url"],
model=server_data["model"],
timeout=server_data.get("timeout", 120.0),
max_concurrent=server_data.get("max_concurrent", 4),
weight=server_data.get("weight", 1.0),
enabled=server_data.get("enabled", True)
)
self.servers[server_config.name] = server_config
if server_config.enabled:
self.clients[server_config.name] = LMStudioClient(
base_url=server_config.url,
timeout=server_config.timeout,
max_concurrent_requests=server_config.max_concurrent,
debug_mode=self.debug_mode
)
self.stats[server_config.name] = ServerStats()
# Load agent mappings
self.agent_mappings = config_data.get("agent_mapping", {})
# Load load balancing strategy
strategy_name = config_data.get("load_balance_strategy", "agent_type_mapping")
try:
self.load_balance_strategy = LoadBalanceStrategy(strategy_name)
except ValueError:
self.load_balance_strategy = LoadBalanceStrategy.AGENT_TYPE_MAPPING
if self.debug_mode:
print(f"๐ง Loaded multi-server config: {len(self.servers)} servers")
for name, server in self.servers.items():
status = "enabled" if server.enabled else "disabled"
print(f" - {name}: {server.url} ({server.model}) [{status}]")
except Exception as e:
logger.error(f"Failed to load config from {config_path}: {e}")
self._create_default_config()
def get_server_for_agent_type(self, agent_type: str) -> Optional[str]:
"""
Get the appropriate server for an agent type.
Args:
agent_type: Type of agent (research, analysis, synthesis, critic)
Returns:
Server name or None if no suitable server
"""
if self.load_balance_strategy == LoadBalanceStrategy.AGENT_TYPE_MAPPING:
return self.agent_mappings.get(agent_type, self._get_fallback_server())
elif self.load_balance_strategy == LoadBalanceStrategy.ROUND_ROBIN:
return self._get_round_robin_server()
elif self.load_balance_strategy == LoadBalanceStrategy.LEAST_BUSY:
return self._get_least_busy_server()
elif self.load_balance_strategy == LoadBalanceStrategy.FASTEST_RESPONSE:
return self._get_fastest_server()
return self._get_fallback_server()
def _get_fallback_server(self) -> Optional[str]:
"""Get first available server as fallback."""
for name, server in self.servers.items():
if server.enabled and name in self.clients:
return name
return None
def _get_round_robin_server(self) -> Optional[str]:
"""Get next server using round-robin selection."""
available_servers = [name for name, server in self.servers.items()
if server.enabled and name in self.clients]
if not available_servers:
return None
server_name = available_servers[self._round_robin_index % len(available_servers)]
self._round_robin_index = (self._round_robin_index + 1) % len(available_servers)
return server_name
def _get_least_busy_server(self) -> Optional[str]:
"""Get server with lowest current load."""
available_servers = [(name, self.stats[name].current_load)
for name, server in self.servers.items()
if server.enabled and name in self.clients]
if not available_servers:
return None
# Sort by current load (ascending)
available_servers.sort(key=lambda x: x[1])
return available_servers[0][0]
def _get_fastest_server(self) -> Optional[str]:
"""Get server with fastest average response time."""
available_servers = [(name, self.stats[name].average_response_time)
for name, server in self.servers.items()
if server.enabled and name in self.clients and self.stats[name].total_requests > 0]
if not available_servers:
return self._get_fallback_server()
# Sort by average response time (ascending)
available_servers.sort(key=lambda x: x[1])
return available_servers[0][0]
async def complete_for_agent_type(self, agent_type: str, agent_id: str,
system_prompt: str, user_prompt: str,
temperature: float = 0.7, max_tokens: Optional[int] = None,
priority: RequestPriority = RequestPriority.NORMAL) -> LLMResponse:
"""
Complete request using appropriate server for agent type.
Args:
agent_type: Type of agent making request
agent_id: ID of the requesting agent
system_prompt: System prompt
user_prompt: User prompt
temperature: Sampling temperature
max_tokens: Maximum tokens
priority: Request priority
Returns:
LLM response
Raises:
LMStudioConnectionError: If no servers available
"""
server_name = self.get_server_for_agent_type(agent_type)
if not server_name or server_name not in self.clients:
raise LMStudioConnectionError(f"No available server for agent type: {agent_type}")
# Check health if needed
await self._check_server_health(server_name)
client = self.clients[server_name]
server_config = self.servers[server_name]
stats = self.stats[server_name]
if self.debug_mode:
print(f"๐ {agent_id} ({agent_type}) โ {server_name} ({server_config.url}, {server_config.model})")
# Track load
stats.current_load += 1
try:
start_time = time.perf_counter()
# Make request with specified model
response = await client.complete_async(
agent_id=agent_id,
system_prompt=system_prompt,
user_prompt=user_prompt,
temperature=temperature,
max_tokens=max_tokens,
model=server_config.model,
priority=priority
)
end_time = time.perf_counter()
response_time = end_time - start_time
# Update statistics
stats.update_stats(response.tokens_used, response_time)
return response
finally:
# Decrease load counter
stats.current_load = max(0, stats.current_load - 1)
async def complete(self, agent_id: str, system_prompt: str, user_prompt: str,
temperature: float = 0.7, max_tokens: Optional[int] = None,
priority: RequestPriority = RequestPriority.NORMAL) -> LLMResponse:
"""
Complete request using default load balancing (fallback method).
Args:
agent_id: ID of the requesting agent
system_prompt: System prompt
user_prompt: User prompt
temperature: Sampling temperature
max_tokens: Maximum tokens
priority: Request priority
Returns:
LLM response
"""
# Extract agent type from agent_id if possible
agent_type = "general"
if "_" in agent_id:
agent_type = agent_id.split("_")[0]
return await self.complete_for_agent_type(
agent_type, agent_id, system_prompt, user_prompt,
temperature, max_tokens, priority
)
async def _check_server_health(self, server_name: str):
"""Check health of specific server."""
current_time = time.time()
stats = self.stats[server_name]
# Only check if enough time has passed
if current_time - stats.last_health_check < self._health_check_interval:
return
client = self.clients[server_name]
try:
health_ok = client.test_connection()
stats.health_status = health_ok
stats.last_health_check = current_time
if not health_ok and self.debug_mode:
print(f"โ ๏ธ Server {server_name} health check failed")
except Exception as e:
stats.health_status = False
stats.last_health_check = current_time
logger.warning(f"Health check failed for {server_name}: {e}")
async def health_check_all_servers(self) -> Dict[str, bool]:
"""
Check health of all servers.
Returns:
Dictionary mapping server names to health status
"""
health_results = {}
for server_name in self.clients:
await self._check_server_health(server_name)
health_results[server_name] = self.stats[server_name].health_status
return health_results
def get_pool_stats(self) -> Dict[str, Any]:
"""
Get comprehensive statistics for the entire pool.
Returns:
Dictionary with pool statistics
"""
pool_stats = {
"total_servers": len(self.servers),
"active_servers": len(self.clients),
"load_balance_strategy": self.load_balance_strategy.value,
"servers": {}
}
for name, stats in self.stats.items():
server_config = self.servers[name]
pool_stats["servers"][name] = {
"config": {
"url": server_config.url,
"model": server_config.model,
"enabled": server_config.enabled
},
"stats": {
"total_requests": stats.total_requests,
"total_tokens": stats.total_tokens,
"average_response_time": stats.average_response_time,
"current_load": stats.current_load,
"health_status": stats.health_status
}
}
return pool_stats
def get_agent_mapping_info(self) -> Dict[str, str]:
"""Get current agent type to server mappings."""
return self.agent_mappings.copy()
async def close_all(self):
"""Close all client connections."""
for client in self.clients.values():
if hasattr(client, 'close_async'):
await client.close_async()
def display_pool_status(self):
"""Display current pool status for debugging."""
if not self.debug_mode:
return
print(f"\nโญโ LM STUDIO POOL STATUS โโฎ")
print(f"โ Strategy: {self.load_balance_strategy.value}")
print(f"โ Servers: {len(self.clients)} active / {len(self.servers)} total")
for name, stats in self.stats.items():
server = self.servers[name]
status = "๐ข" if stats.health_status else "๐ด"
load = f"{stats.current_load}/{server.max_concurrent}"
avg_time = f"{stats.average_response_time:.2f}s" if stats.total_requests > 0 else "N/A"
print(f"โ {status} {name}: {load} load, {stats.total_requests} reqs, {avg_time} avg")
print(f"โ Agent Mapping:")
for agent_type, server_name in self.agent_mappings.items():
print(f"โ {agent_type} โ {server_name}")
print(f"โฐ{'โ'*35}โฏ") |