performance

name	performance
description	Optimize application performance through caching strategies, load balancing, database scaling, and monitoring. Build systems handling thousands of concurrent users.
sasmp_version	2.0.0
bonded_agent	05-caching-performance
bond_type	PRIMARY_BOND
atomic_operations	["BOTTLENECK_ANALYSIS","CACHE_IMPLEMENTATION","LOAD_BALANCING_CONFIG","SCALING_STRATEGY"]
parameter_validation	{"query":{"type":"string","required":true,"minLength":5,"maxLength":2000},"current_rps":{"type":"integer","required":false,"description":"Current requests per second"},"target_latency_ms":{"type":"integer","required":false,"description":"Target P99 latency in milliseconds"}}
retry_logic	{"max_attempts":3,"backoff":"exponential","initial_delay_ms":1000}
logging_hooks	{"on_invoke":"skill.performance.invoked","on_success":"skill.performance.completed","on_error":"skill.performance.failed"}
exit_codes	{"SUCCESS":0,"INVALID_INPUT":1,"METRICS_UNAVAILABLE":2,"OPTIMIZATION_FAILED":3}

Performance Optimization Skill

Bonded to: caching-performance-agent

---

Quick Start

# Invoke performance skill
"My API is slow, help me optimize it"
"Set up Redis caching for my application"
"Configure load balancing for high availability"

---

Instructions

1. Identify Bottlenecks: Profile application, analyze metrics
2. Choose Strategy: Select caching, scaling, or optimization approach
3. Implement Cache: Set up Redis/Memcached with appropriate pattern
4. Configure Scaling: Horizontal or vertical based on needs
5. Monitor Results: Set up APM and track improvements

---

Caching Patterns

Pattern	Use Case	Consistency	Complexity
Cache-Aside	Read-heavy, tolerates stale	Eventual	Low
Write-Through	Write-heavy, needs consistency	Strong	Medium
Write-Behind	High throughput writes	Eventual	High
Refresh-Ahead	Predictable access	Strong	Medium

---

Decision Tree

Performance Issue?
    │
    ├─→ High latency → Check database queries
    │     ├─→ Slow queries → Add indexes, optimize SQL
    │     └─→ Network → Add caching, reduce round-trips
    │
    ├─→ High CPU → Profile code
    │     ├─→ Algorithmic → Optimize algorithms
    │     └─→ Too much load → Scale horizontally
    │
    └─→ Memory issues → Analyze memory usage
          ├─→ Leaks → Find and fix leaks
          └─→ Large data → Implement pagination, streaming

---

Examples

Example 1: Redis Cache-Aside

import redis
import json
from functools import wraps

r = redis.Redis(host='localhost', port=6379, decode_responses=)

 ():
     ():

         ():
            cache_key = 
            cached = r.get(cache_key)
             cached:
                 json.loads(cached)

            result = func(*args, **kwargs)
            r.setex(cache_key, ttl_seconds, json.dumps(result))
             result
         wrapper
     decorator


 () -> :
    
     {: user_id, : }