Run:AI Case Studies London Medical Imaging & AI Centre Speeds Up Research with Run:ai

Edit This Case Study Record

	London Medical Imaging & AI Centre Speeds Up Research with Run:ai Run:AI

London Medical Imaging & AI Centre Speeds Up Research with Run:ai

Run:AI

Technology Category	Analytics & Modeling - Machine Learning Application Infrastructure & Middleware - API Integration & Management
Applicable Industries	Healthcare & Hospitals
Applicable Functions	Product Research & Development
Use Cases	Computer Vision Predictive Maintenance
Services	Data Science Services System Integration
Challenge	The London Medical Imaging & AI Centre for Value Based Healthcare was facing several challenges with its AI hardware. The total GPU utilization was below 30%, with significant idle periods for some GPUs despite demand from researchers. The system was overloaded on multiple occasions where more GPUs were needed for running jobs than were available. Poor visibility and scheduling led to delays and waste, with bigger experiments requiring a large number of GPUs sometimes unable to begin because smaller jobs using only a few GPUs were blocking them out of their resource requirements. Read More
About Customer	The London Medical Imaging & Artificial Intelligence Centre for Value Based Healthcare is a consortium of academic, healthcare and industry partners, led by King’s College London and based at St. Thomas’ Hospital. It uses medical images and electronic healthcare data held by the UK National Health Service to train sophisticated deep learning algorithms for computer vision and natural-language processing. These algorithms are used to create new tools for effective screening, faster diagnosis and personalized therapies, to improve patients’ health. Read More
Solution	The AI Centre implemented Run:ai's Platform to address these challenges. The platform increased GPU utilization by 110%, with resultant increases in experiment speed. Researchers ran more than 300 experiments in a 40-day period, compared to just 162 experiments that were run in a simulation of the same environment without Run:ai. By dynamically allocating pooled GPU to workloads, hardware resources were shared more efficiently. The platform also improved visibility with advanced monitoring and cluster management tools, allowing data scientists to see which GPU resources were not being used and dynamically adjust the size of their job to run on available capacity. The platform also enabled fair scheduling and guaranteed resources, allowing large ongoing workloads to use the optimal amount of GPU during low-demand times, and automatically allowing shorter, higher-priority workloads to run alongside. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Machine Learning

Application Infrastructure & Middleware - API Integration & Management

Applicable Industries

Healthcare & Hospitals

Applicable Functions

Product Research & Development

Use Cases

Computer Vision

Predictive Maintenance

Services

Data Science Services

System Integration

Challenge

The London Medical Imaging & AI Centre for Value Based Healthcare was facing several challenges with its AI hardware. The total GPU utilization was below 30%, with significant idle periods for some GPUs despite demand from researchers. The system was overloaded on multiple occasions where more GPUs were needed for running jobs than were available. Poor visibility and scheduling led to delays and waste, with bigger experiments requiring a large number of GPUs sometimes unable to begin because smaller jobs using only a few GPUs were blocking them out of their resource requirements.

About Customer

The London Medical Imaging & Artificial Intelligence Centre for Value Based Healthcare is a consortium of academic, healthcare and industry partners, led by King’s College London and based at St. Thomas’ Hospital. It uses medical images and electronic healthcare data held by the UK National Health Service to train sophisticated deep learning algorithms for computer vision and natural-language processing. These algorithms are used to create new tools for effective screening, faster diagnosis and personalized therapies, to improve patients’ health.

Solution

The AI Centre implemented Run:ai's Platform to address these challenges. The platform increased GPU utilization by 110%, with resultant increases in experiment speed. Researchers ran more than 300 experiments in a 40-day period, compared to just 162 experiments that were run in a simulation of the same environment without Run:ai. By dynamically allocating pooled GPU to workloads, hardware resources were shared more efficiently. The platform also improved visibility with advanced monitoring and cluster management tools, allowing data scientists to see which GPU resources were not being used and dynamically adjust the size of their job to run on available capacity. The platform also enabled fair scheduling and guaranteed resources, allowing large ongoing workloads to use the optimal amount of GPU during low-demand times, and automatically allowing shorter, higher-priority workloads to run alongside.

Impact #1	Increased GPU utilization by 110%, with resultant increases in experiment speed.
Impact #2	Researchers ran more than 300 experiments in a 40-day period, compared to just 162 experiments that were run in a simulation of the same environment without Run:ai.
Impact #3	Improved visibility with advanced monitoring and cluster management tools.

Impact #1

Increased GPU utilization by 110%, with resultant increases in experiment speed.

Impact #2

Researchers ran more than 300 experiments in a 40-day period, compared to just 162 experiments that were run in a simulation of the same environment without Run:ai.

Impact #3

Improved visibility with advanced monitoring and cluster management tools.

Benefit #1	2.1X Higher GPU Utilization
Benefit #2	31X Faster Experiments
Benefit #3	1.85X More Experiments

Benefit #1

2.1X Higher GPU Utilization

Benefit #2

31X Faster Experiments

Benefit #3

1.85X More Experiments

Download PDF Version

Overview

London Medical Imaging & AI Centre Speeds Up Research with Run:ai

Operational Impact

Quantitative Benefit