2 min read

PAIGE and Igneous Build Industry-Leading Compute Cluster for Healthcare AI

By Catherine Chiang on January 16, 2019

PAIGE’s mission to revolutionize the diagnosis and treatment of cancer through machine learning requires an extremely large dataset of high resolution slide images. To do so, they are building an industry-leading compute cluster for healthcare AI. The team needed to not only protect all of this unstructured data, but also programmatically move and process small subsets of the overall dataset on demand for high performance computations.

PAIGE chose Igneous to act as their unstructured data management system of record for its intelligent indexing, scalability, and as-a-service model, as well as its data movement capabilities. Igneous is integrated tightly within PAIGE’s machine-learning-based data workflow along with Pure FlashBlade and NVIDIA for compute and image processing.

Some of the benefits PAIGE has realized by implementing Igneous include:

  • Enabling data owners and AI engineers to quickly organize, find, and move specific subsets of the overall dataset, accelerating research workflows

  • Providing scalable data protection for petabytes of unstructured data in any location

  • High performance iterative data movement via Igneous’ API-based integration with the Pure FlashBlade/AIRI and NVIDIA compute cluster  

  • Enabling data analysis collaboration between pathologists and AI engineers for model training via centralized system of record and dataset classification

  • Reducing organizational costs, allowing the team to focus resources on computational pathology and AI development, not systems management, via Igneous’ managed as-a-Service delivery model.



Full UDM benefits delivered in a single solution

Igneous DataProtect provides a scalable solution for protecting PAIGE’s research-critical datasets, which are used to train their algorithms that assist in diagnosing cancer. In addition, Igneous DataDiscover and Igneous DataFlow enable PAIGE data owners, including pathologists, researchers, and AI engineers, to see, access, and move data within training and modeling workflows. This capability to easily find and move specific subsets of data for high performance computing and image processing ties PAIGE’s ML workflow together and makes it possible.

To learn more about how Igneous supports PAIGE in developing AI algorithms for diagnosing cancer, check out the case study.

Read case study

Catherine Chiang

Written by Catherine Chiang

Subscribe for Updates

Get the latest Igneous blog posts delivered to your inbox.