Weave Code
S3 Vectors Laravel Package
async-aws/s3-vectors
Async AWS S3 Vectors client for PHP. Provides lightweight, non-blocking access to Amazon S3 vector features with request/response models, retries, and signing—ideal for apps that need fast, async integration without the full AWS SDK.
[READ ONLY] S3 Vectors client
Frequently asked questions about S3 Vectors
- How do I integrate async-aws/s3-vectors with Laravel’s queue system for background vector storage?
- Use Laravel’s `ShouldQueue` interface to dispatch async jobs. For example, create a `StoreVectorJob` that extends `ShouldQueue` and inject the `S3VectorsClient` via Laravel’s service container. Queue jobs with `Queue::push(new StoreVectorJob($vectorData))` for non-blocking operations.
- Does this package support S3-compatible storage like MinIO for on-prem or edge deployments?
- Yes, the package works with any S3-compatible storage, including MinIO. Configure the client with your endpoint URL and credentials, making it ideal for GDPR-compliant or low-latency edge AI deployments where AWS S3 isn’t an option.
- What Laravel versions are officially supported by async-aws/s3-vectors?
- The package is designed for Laravel 9+ and PHP 8.1+. Check the [GitHub repository](https://github.com/async-aws/s3-vectors) for the latest compatibility notes, as it aligns with the async-aws ecosystem’s support policy for modern Laravel releases.
- Can I use this for approximate nearest neighbor (ANN) search, or is it just for raw vector storage?
- This package handles raw vector storage in S3 efficiently, but ANN search requires a separate layer. Pair it with client-side libraries like FAISS or serverless solutions like OpenSearch for semantic search capabilities.
- How do I handle eventual consistency in S3 when retrieving vectors in Laravel?
- Leverage Laravel’s retry logic (e.g., `retry_after` in jobs) and idempotency keys to mitigate stale reads. For critical applications, implement a cache layer (e.g., Redis) with short TTLs to serve fresh data while waiting for S3 consistency.
- What vector formats are supported, and how do I pre-process data before storage?
- The package supports raw binary vectors (e.g., float32 arrays) and structured formats like Parquet via `spatie/array-to-parquet`. Pre-process vectors in Laravel’s service container—normalize dimensions, reduce redundancy, or validate checksums before storage.
- How do I test async-aws/s3-vectors in a Laravel application without hitting real AWS S3?
- Mock the `S3VectorsClient` using tools like Mockery or PestPHP. For example, stub `putVector` and `getVector` methods in tests to simulate S3 responses, ensuring your logic works without external dependencies.
- Is there a performance impact when storing large vectors (>1MB) in S3 with this client?
- For vectors >1MB, use S3’s multipart uploads (supported by the client) to improve throughput. Benchmark PUT/GET operations in your environment, and consider S3 Select for partial queries to reduce latency.
- How do I secure API keys or credentials when using this package in production?
- Store AWS credentials in Laravel’s `.env` file (e.g., `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`) and use Laravel’s built-in encryption for sensitive data. For additional security, integrate PSR-15 middleware (e.g., rate-limiting) via Laravel’s HTTP client.
- What are the alternatives to async-aws/s3-vectors for Laravel vector storage, and when should I choose them?
- Consider PostgreSQL with `pgvector` for strong consistency or Pinecone for managed ANN search. Use this package if you need cost-efficient, scalable storage with eventual consistency, async Laravel integration, or S3-compatible multi-cloud support.
Weaver
How can I help you explore Laravel packages today?