# Intel IAA Plugin

Intel IAA Plugin is a RocksDB storage plugin that offloads compression and decompression to Intel In-Memory Analytics Accelerator (IAA) hardware, letting server operators speed up data-heavy workloads without. It is open source and self-hosted, installable on Docker Hub.

Canonical page: https://ourlocalhost.com/app/5929/intel-iaa-plugin

## Description

Intel IAA Plugin is a RocksDB storage plugin that offloads compression and decompression to Intel In-Memory Analytics Accelerator (IAA) hardware, letting server operators speed up data-heavy workloads without overloading the main CPU. It hooks into RocksDB's pluggable compression framework and uses the Intel Query Processing Library (QPL) to accelerate compression, so applications that read and write large datasets run faster and more smoothly. It is aimed at homelab and production operators running RocksDB-based services on Intel IAA-capable servers who want hardware-accelerated compression. Reach for it when your database is compression-bound and you have IAA hardware available to move that work off the CPU cores.

## Platform availability

Available on 0 of 10 platforms.

## Health score

39/100 (Low), based on 4 of 5 factors.
- Maintained: 42 (last commit 305 days ago)
- Popular: 25 (16 GitHub stars)
- Easy to install: 25 (0 install platforms, install notes, Docker image)
- Security check: 65 (no serious findings, tests, OpenSSF Scorecard 3.6/10)

## Links

- GitHub: https://github.com/intel/iaa-plugin-rocksdb
- Docker Hub: https://hub.docker.com/r/intel/intel-iaa-plugin

## Common questions

### What must be done before installing the Kubernetes IAA plugin?

Deploy Node Feature Discovery and NodeFeatureRules so Kubernetes labels nodes that have Intel devices. Replace the release version placeholders in the install commands with an actual release tag.

### Can IAA devices that work on bare metal be passed through to QEMU?

There are reported problems with passthrough and configuration of IAA devices in QEMU after they work on the bare-metal host in dedicated mode.

---
Source: Our Local Host (https://ourlocalhost.com/app/5929/intel-iaa-plugin), the self-hosted and open-source app directory.
