25th June 2024
Core Modules
Build, deploy, and scale AI applications quickly and efficiently on Peak’s fully-managed AI application development platform. The following core modules make up the current out of the box capabilities.
Workspaces
Features: Cloud-based data and machine learning development environments allowing for instant connectivity to Peak data sources and secure development of AI applications on Peak via a supported browser. Supports Jupyter Lab and RStudio Workspaces as standard. Remote tunneling access may be available on request.
Workflows
Features: Cloud-native orchestration platform for creating and managing data and machine learning learning pipelines and other automation tasks. Provides elastic compute options to manage workload scaling, notifications and alerting, and conditional logic and parameterisation for workload orchestration.
Services
Features: Deploy, scale and manage web applications, data visualizations and model APIs. Provides elastic compute options for managing data and memory intensive applications, notifications and alerting, scaling rules and advanced connectivity and session management tools.
Models
Features: Track model experiments and performance over time. Annotate experiments and model training pipeline outputs with performance metadata and plots. This is a Peak-managed version of MLFlow.
Images
Features: Create and manage your development environments to ensure the stability and security of your data and machine learning workloads on Peak. Choose from preset images, or define your own. Create and manage environment versions and secrets, and instantly deploy them across Peak.
Press
Features: Create, manage and share AI applications using Press - Peak’s AI Application Development Toolkit. Automate and manage complex deployments and update rollouts. Leverage Press’ built-in composability capabilities to quickly hot-swap data and machine learning components, and share useful components across your projects.
Core Module Requirements:
This list of requirements is not exhaustive and should be reviewed with a member of the Peak team alongside specific business requirements.
A customer-managed Redshift or Snowflake instance that Factory can connect to. Alternatively, Peak can deploy a Peak-managed Redshift or Snowflake instance as an Additional Module (see below).
Factory Additional Modules
Optional extras are the additional capabilities that can be added to the Subscription for an additional cost. The Customer may elect to receive the following Optional Extras. In the event the Customer elects to receive any Optional Extras, the parties shall discuss and negotiate relevant terms in good faith.
Snowflake Data Warehouse (Peak Managed)
Features: Peak-provisioned and managed Snowflake Data Warehouse. Warehouse size and number of Warehouses will be subject to discussion with Peak technical sales representatives.
Redshift Data Warehouse (Peak Managed)
Features: Peak-provisioned and managed Redshift Data Warehouse. Instance size and number of nodes will be subject to discussion with Peak technical sales representatives.