matthewlam721/octopus-parallel
Octopus: Block-Level GPU Scheduling for Variable-Length Batches
This project helps operations engineers and data scientists efficiently process large batches of images or video frames on edge devices. It takes a collection of images, often of different sizes, and outputs processed results without wasting resources on padding. It's designed for scenarios where every millisecond and joule counts, especially for applications on drones, satellites, or other embedded systems.
Use this if you need to perform image preprocessing (like cropping, resizing, or simple filters) on a large, variable-sized batch of images or video frames on a power-constrained edge device.
Not ideal if your image processing tasks are purely compute-heavy operations (like complex blurring) where memory access isn't the bottleneck, or if you're working with uniformly sized inputs on powerful cloud GPUs.
Stars
26
Forks
—
Language
Python
License
MIT
Category
Last pushed
Feb 23, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/matthewlam721/octopus-parallel"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
tensorflow/tensorflow
An Open Source Machine Learning Framework for Everyone
microsoft/tensorwatch
Debugging, monitoring and visualization for Python Machine Learning and Data Science
KomputeProject/kompute
General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics...
hailo-ai/hailort-drivers
The Hailo PCIe driver is required for interacting with a Hailo device over the PCIe interface
NVIDIA/nvshmem
NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM...