Introducing ROCm-LS: Accelerating Life Science Workloads with AMD Instinct™ GPUs#
AMD is thrilled to announce the early access release of ROCm-LS (ROCm Life Science), a new cutting-edge software toolkit designed to accelerate life science computational workloads on AMD Instinct™ GPUs. ROCm-LS joins ROCm-DS as a part of AMD’s new family of toolkits aimed at providing powerful solutions to real world problems. Similar to ROCm-DS, ROCm-LS is built upon the established ROCm software ecosystem, offering a collection of components and libraries that address the pressing needs of the life science community. The early access release of ROCm-LS enables you to experiment with accelerating your life science workloads, such as digital pathology, automated medical image analysis, and feature extraction and enhancement in large TIFF files on AMD Instinct GPUs. Join us in exploring this tantalizing glimpse into the future capabilities of ROCm-LS, setting the stage for the next evolution in life science computing.
This early access release serves as a software technology preview and is not recommended for production workloads.
GPU Accelerated Life Science#
ROCm-LS utilizes AMD Instinct GPUs to accelerate life science and healthcare workloads. The early access release serves as the first step in an ongoing effort towards creating a fully fleshed toolkit which aims to address the needs of the life science and healthcare communities. This release introduces the hipCIM library which enhances scientific and medical imaging applications by leveraging all of the advantages offered by AMD Instinct GPUs.
What is hipCIM?#
hipCIM is an open-source, GPU-accelerated software library which enables computer vision and image processing operations for multidimensional image datasets. It empowers users from a diverse array of scientific fields to accelerate their workloads on AMD Instinct GPUs. Some of these fields include biomedical imaging, geospatial analytics, material sciences, life sciences, remote sensing, and more.
The early access release, based on the open-source cuCIM library, offers an impressive array of features:
GPU-accelerated I/O and Processing Primitives: hipCIM enhances tasks such as color conversion, feature extraction, filters, morphology, segmentation, and transformations for N-dimensional images.
API Compatibility: hipCIM maintains API compatibility with the NVIDIA cuCIM library, enabling effortless integration and transition of existing workloads to AMD GPUs without the need for “hipification.”
Multifaceted APIs: Supporting both Python and C++ APIs, hipCIM facilitates a wide variety of applications and development environments.
Mirroring Established Libraries: The familiar hipCIM API design is similar to the scikit-image and OpenSlide libraries, enabling ease of adoption for users already familiar with these staples.
Open Source Collaboration: Fully open-sourced under the Apache-2.0 license, hipCIM welcomes contributions and fosters a collaborative development environment.
hipCIM provides a powerful solution to accelerate computational vision and image processing aimed at multidimensional image analysis. It contains a comprehensive set of tools which enable developers to craft advanced image processing applications by leveraging GPU acceleration. This not only accelerates important scientific research, but also enables larger datasets to be processed faster for critical real world applications.
For a better understanding of hipCIM and all it has to offer, be sure to have a look at the Announcing hipCIM: A Cutting-Edge Solution for Accelerated Multidimensional Image Processing blog.
ROCm-LS as a Revolutionary Platform#
The ROCm Life Science toolkit builds upon the established core ROCm platform, providing a GPU accelerated alternative to existing life science toolkits. It empowers users to expedite their life science processing and analysis workloads on AMD GPUs, providing the capability to run intensive applications involving vast datasets swiftly.
ROCm-LS enables users to:
Accelerate New and Existing Workloads: Boost the performance of life science applications, building scalable solutions tailored to modern, data-centric challenges.
Create Advanced Processing Pipelines: Develop intricate pre- and post-processing applications for AI models and expedite data science workflows.
Dive Into ROCm-LS: Unlocking Accelerated Computational Vision and Image Processing for Life Sciences#
With the early access release of ROCm-LS, AMD invites you to explore and experiment with this transformative opportunity to delve into enhanced life science workflows. As users engage with ROCm-LS, they can experience firsthand the remarkable performance improvements that come with utilizing the full power of AMD Instinct GPUs to accelerate life science workloads. This release marks the beginning of a promising journey, with future versions set to deliver additional optimizations, functionalities, and robust support to fully harness the power of AMD Instinct GPUs across a variety of diverse life science applications.
ROCm-LS features versatile APIs in both Python and C++, catering to professionals and researchers across various sectors. This dual-language support ensures that users can seamlessly maximize the toolkit’s vast potential, regardless of their field, expertise, or experience. To explore these capabilities, we recommend consulting the hipCIM documentation, which provides comprehensive installation instructions and a complete overview of the functionalities available in this early access release.
Embrace this chance to innovate and accelerate your data-driven life science endeavors with ROCm-LS, laying the groundwork for more powerful and efficient applications in the future.
Summary#
The early access release of ROCm-LS marks a significant leap forward in the realm of life science computing on AMD GPUs. This release allows you to accelerate some of your real world life science workloads such as digital pathology, image preparation for deep learning pipelines, feature extraction, and tissue segmentation with ROCm-LS and AMDs Instinct GPUs. We invite developers and researchers to engage with ROCm-LS, explore its capabilities, and contribute to its evolution. Dive into the future of accelerated scientific computing today with ROCm-LS!
Be sure to check out the latest developments and community contributions hosted on the ROCm-LS GitHub. For more information on contributing or exploring related content, visit the ROCm-LS documentation page, Instinct Docs page, and have a look at the Announcing hipCIM: A Cutting-Edge Solution for Accelerated Multidimensional Image Processing blog.
Disclaimers#
Third-party content is licensed to you directly by the third party that owns the content and is not licensed to you by AMD. ALL LINKED THIRD-PARTY CONTENT IS PROVIDED “AS IS” WITHOUT A WARRANTY OF ANY KIND. USE OF SUCH THIRD-PARTY CONTENT IS DONE AT YOUR SOLE DISCRETION AND UNDER NO CIRCUMSTANCES WILL AMD BE LIABLE TO YOU FOR ANY THIRD-PARTY CONTENT. YOU ASSUME ALL RISK AND ARE SOLELY RESPONSIBLE FOR ANY DAMAGES THAT MAY ARISE FROM YOUR USE OF THIRD-PARTY CONTENT.