YellowDog Platform

Scaling applications across the cloud can be challenging. Without aligning computing resources with your workload’s fluctuating requirements, you could end up exceeding your budgets and missing your deadlines.

With the YellowDog Platform, you can benefit from all the cloud has to offer, without any limitations. Our powerful scheduler works seamlessly alongside our intelligent provisioning tools, so you can manage your workloads across different geographical regions, machine shapes and instance lifecycles.

Book a Demo

What Does the YellowDog Platform Deliver?

Rapid Scaling

Access unparalleled scale and easily manage computing resources from multiple providers, regions, and machine shapes.

Dynamic Clusters

Respond to the specific needs of your workload with computing clusters that are composable and expand and contract, as required.

User-Friendly Solution

Deliver workloads on time and on budget with a solution that is cloud native, fault tolerant and accessible via an easy-to-use User Portal.

Cloud Native Scheduling

The YellowDog Scheduler has been built with the cloud in mind. It is free from the restrictions of traditional legacy tooling and enables users to scale far beyond existing technologies, across multiple regions, data centres and instance shapes.

It can utilise Spot instances, where other tools cannot, acting as both a scheduler and meta scheduler, to work with loosely and tightly coupled workloads. It also removes complexity with a single cluster view – with no need to consider multi-data centre/provider configurations.

The addition of YellowDog’s Object Store Service, gives the Scheduler access to a coherent data surface for workloads, regardless of the cloud provider and location.

Best Source of Compute

With a deep understanding of infrastructure and performance, YellowDog help businesses make smarter decisions about what infrastructure is used to deliver compute-intensive workloads. Utilising data from over 9,000 cloud instances globally, the YellowDog Platform can optimise your compute infrastructure.

Simply set your constraints and preferences and allow our intelligent provisioning strategies to automatically select the Best Source of Compute for your workload. The technology finds the computing resources that most closely match your workload’s unique requirements, across a varied list of attributes, such as cost, performance, and environmental impact.

Below is a short demonstration of YellowDog Insights, the data which feeds Best Source of Compute.

Intelligent Provisioning

Intelligent: Select the exact instances you want, or use an automated strategy to provision compute based on your broader constraints and preferences
Intuitive: Choose whether you want to split your resources across the instances you choose, or if you wish to rank each resource based on your preferences
Resilient: Automatically maintain your instance count if you should encounter an issue or failure, for minimal interruption (important for pre-emptible instances)
Scalable: Create clusters of different machine types, instance lifecycles, data centres, and regions to dynamically scale your workloads across the cloud

Key Features

YellowDog executes workloads in the cloud end-to-end. Workload management, resource provisioning, and data capture from the provider’s object store, all occur without ever leaving the YellowDog environment. Key features include:

A cloud native scheduler providing automated task management, allocation, and resource coordination.
Cloud computing and Object Store connectors for all major cloud providers.
Direct API, Java and Python SDKs to integrate with our cloud native scheduler.
Meta scheduling to coordinate multiple third-party clusters (e.g. Slurm, LSF) using YellowDog as a single workload submission platform.
An intuitive and easy-to-use User Portal for configuration, monitoring, reporting and analytics.
Keyrings for secure and audited Identity Access and Management across all public clouds.
Cloud native design to ensure optimum fault tolerance, scalability and performance.

Use Cases

Workload Management

Coordinate and monitor workload execution across all compute types and automatically manage compute resource levels.

Data-Led Compute Provisioning

Ensure compliance with provisioning strategies that are influenced by user-defined attributes and insight from the YellowDog Index.

Best Source of Compute

Improve performance by automatically selecting the most effective resources available for your workload's specific needs.

Resource Management

Manage computing resources regardless of provider, location, or type through our intuitive User Portal, API or Java and Python SDKs.

Multi-Cloud Abstraction

Overcome the complexity of multiple providers, regions, and resource types by viewing the cloud through a single management pane.

Increase On-Premise Efficiency

Lower total cloud costs with more efficient utilisation of your existing
on-premise infrastructure.

“The professional and timely manner of this collaboration has enabled OMass to rapidly screen a novel target and help the company in its’ ambition to build a pipeline of small molecule therapeutics.”

Giles Brown VP, Head of Medicinal Chemistry, OMass

“YellowDog have transformed how we use our existing hardware and the cloud, making all the behind the scenes workload management invisible.”

Jago Westmacott CTO, Fulcrum Asset Management

“Having the YellowDog Platform there for us on demand really was priceless.”

Alex Sherwood CEO, Giant Animation

“We know exactly how many machines we get, how many hours we use, and how powerful every machine is. So we have total clarity on our available compute.”

Rob Walsh Head of Visualisation, PriestmanGoode

User Videos

Learn how to use the YellowDog Platform with our library of user videos.

Documentation

Access useful information about the YellowDog Platform here.