--- name: perplexity-performance-tuning-2cda9b624305 description: | Optimize Perplexity API performance with caching, batching, and connection pooling. Use when experiencing slow API responses, implementing caching strategies, or optimizing request throughput for Perplexity integrations. Trigger with phrases like "perplexity performance", "optimize perplexity", "perplexity latency", "perplexity caching", "perplexity slow", "perplexity batch". allowed-tools: Read, Write, Edit version: 1.0.0 license: MIT author: Jeremy Longshore --- # Perplexity Performance Tuning ## Overview Optimize Perplexity API performance with caching, batching, and connection pooling. ## Prerequisites - Perplexity SDK installed - Understanding of async patterns - Redis or in-memory cache available (optional) - Performance monitoring in place ## Latency Benchmarks | Operation | P50 | P95 | P99 | |-----------|-----|-----|-----| | Read | 50ms | 150ms | 300ms | | Write | 100ms | 250ms | 500ms | | List | 75ms | 200ms | 400ms | ## Caching Strategy ### Response Caching ```typescript import { LRUCache } from 'lru-cache'; const cache = new LRUCache({ max: 1000, ttl: 60000, // 1 minute updateAgeOnGet: true, }); async function cachedPerplexityRequest( key: string, fetcher: () => Promise, ttl?: number ): Promise { const cached = cache.get(key); if (cached) return cached as T; const result = await fetcher(); cache.set(key, result, { ttl }); return result; } ``` ### Redis Caching (Distributed) ```typescript import Redis from 'ioredis'; const redis = new Redis(process.env.REDIS_URL); async function cachedWithRedis( key: string, fetcher: () => Promise, ttlSeconds = 60 ): Promise { const cached = await redis.get(key); if (cached) return JSON.parse(cached); const result = await fetcher(); await redis.setex(key, ttlSeconds, JSON.stringify(result)); return result; } ``` ## Request Batching ```typescript import DataLoader from 'dataloader'; const perplexityLoader = new DataLoader( async (ids) => { // Batch fetch from Perplexity const results = await perplexityClient.batchGet(ids); return ids.map(id => results.find(r => r.id === id) || null); }, { maxBatchSize: 100, batchScheduleFn: callback => setTimeout(callback, 10), } ); // Usage - automatically batched const [item1, item2, item3] = await Promise.all([ perplexityLoader.load('id-1'), perplexityLoader.load('id-2'), perplexityLoader.load('id-3'), ]); ``` ## Connection Optimization ```typescript import { Agent } from 'https'; // Keep-alive connection pooling const agent = new Agent({ keepAlive: true, maxSockets: 10, maxFreeSockets: 5, timeout: 30000, }); const client = new PerplexityClient({ apiKey: process.env.PERPLEXITY_API_KEY!, httpAgent: agent, }); ``` ## Pagination Optimization ```typescript async function* paginatedPerplexityList( fetcher: (cursor?: string) => Promise<{ data: T[]; nextCursor?: string }> ): AsyncGenerator { let cursor: string | undefined; do { const { data, nextCursor } = await fetcher(cursor); for (const item of data) { yield item; } cursor = nextCursor; } while (cursor); } // Usage for await (const item of paginatedPerplexityList(cursor => perplexityClient.list({ cursor, limit: 100 }) )) { await process(item); } ``` ## Performance Monitoring ```typescript async function measuredPerplexityCall( operation: string, fn: () => Promise ): Promise { const start = performance.now(); try { const result = await fn(); const duration = performance.now() - start; console.log({ operation, duration, status: 'success' }); return result; } catch (error) { const duration = performance.now() - start; console.error({ operation, duration, status: 'error', error }); throw error; } } ``` ## Instructions ### Step 1: Establish Baseline Measure current latency for critical Perplexity operations. ### Step 2: Implement Caching Add response caching for frequently accessed data. ### Step 3: Enable Batching Use DataLoader or similar for automatic request batching. ### Step 4: Optimize Connections Configure connection pooling with keep-alive. ## Output - Reduced API latency - Caching layer implemented - Request batching enabled - Connection pooling configured ## Error Handling | Issue | Cause | Solution | |-------|-------|----------| | Cache miss storm | TTL expired | Use stale-while-revalidate | | Batch timeout | Too many items | Reduce batch size | | Connection exhausted | No pooling | Configure max sockets | | Memory pressure | Cache too large | Set max cache entries | ## Examples ### Quick Performance Wrapper ```typescript const withPerformance = (name: string, fn: () => Promise) => measuredPerplexityCall(name, () => cachedPerplexityRequest(`cache:${name}`, fn) ); ``` ## Resources - [Perplexity Performance Guide](https://docs.perplexity.com/performance) - [DataLoader Documentation](https://github.com/graphql/dataloader) - [LRU Cache Documentation](https://github.com/isaacs/node-lru-cache) ## Next Steps For cost optimization, see `perplexity-cost-tuning`. --- Source attribution: Dicklesworthstone Original project: https://github.com/Dicklesworthstone/pi_agent_rust/blob/f28fc46ee0db8e710d95b8b55592cd38d60f2b26/tests/ext_conformance/artifacts/plugins-community/plugins/saas-packs/perplexity-pack/skills/perplexity-performance-tuning/SKILL.md License: MIT License information: https://opensource.org/license/mit Original upstream instructions; downloaded from Mega AI Hub. ## Upstream license notice MIT License (with OpenAI/Anthropic Rider) Copyright (c) 2026 Jeffrey Emanuel Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: ADDITIONAL RIDER / RESTRICTION (OpenAI / Anthropic): This rider is part of the "conditions" of this License. In the event of any conflict between this rider and any other portion of this License, this rider controls. "Restricted Parties" means OpenAI, L.L.C.; Anthropic, PBC; any of their respective Affiliates; and any person or entity acting directly or indirectly on behalf of, for the benefit of, or under the direction of any of the foregoing (including any officer, director, employee, contractor, agent, consultant, service provider, or representative). Notwithstanding any other provision of this License, no rights are granted to any Restricted Party. Any purported license, sublicense, assignment, transfer, or other permission to any Restricted Party is null and void absent the express prior written permission of Jeffrey Emanuel. You may not provide, disclose, distribute, sublicense, sell, lease, lend, host, make available, or otherwise permit access to the Software or any derivative work of the Software (as defined in applicable copyright law) (collectively, "Derivative Works") to or for any Restricted Party. For purposes of this rider, "use" includes, without limitation: copying, modifying, merging, publishing, distributing, sublicensing, selling, transferring, making available, hosting, deploying, executing, benchmarking, testing, analyzing, indexing, or incorporating the Software or any Derivative Works into any dataset, training corpus, evaluation harness, or pipeline for machine learning or other automated systems. This rider applies to the Software and all Derivative Works. As a condition of use, you agree that this rider is a precondition to exercising any rights under this License, and you agree that any distribution of the Software or any Derivative Works must include this rider provision unmodified. Any breach of this rider automatically and immediately terminates the permissions granted by this License. Upon termination, you must immediately cease all use and distribution of the Software and any Derivative Works and destroy all copies under your control. You agree that a breach of this rider would cause irreparable harm and that Jeffrey Emanuel may seek injunctive or other equitable relief to enforce this rider, in addition to any other remedies available at law. To the maximum extent permitted by applicable law, the prevailing party in any action to enforce this rider shall be entitled to recover reasonable attorneys' fees and costs. For purposes of this rider, "Affiliate" means any entity that directly or indirectly controls, is controlled by, or is under common control with the specified party. "Control" means ownership of more than 50% of the voting securities or other ownership interest, or the power to direct management or policies by contract or otherwise. The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. # Mega AI Hub resource metadata Creator / publisher: Dicklesworthstone Original project: https://github.com/Dicklesworthstone/pi_agent_rust/blob/f28fc46ee0db8e710d95b8b55592cd38d60f2b26/tests/ext_conformance/artifacts/plugins-community/plugins/saas-packs/perplexity-pack/skills/perplexity-performance-tuning/SKILL.md Attribution basis: Source repository record Fields of work: Development & Engineering Roles: Backend Developer, Data Engineer, ML / AI Engineer Specialties: API, Backend Tools: Perplexity Api, Redis Country applicability: Country not specified Classification basis: Source catalog labels Instruction SHA-256: 14c3b028cf9d511a7b7f3e130566a340f23d9dce6e67b70a2a10ecd882c23b13