FEATURES
Jobs compute control plane
Gain unprecedented visibility and control over your Databricks Workspace with Gradient’s single pane of glass for compute cost and performance
ML-powered insights
Actionable insights into your data infrastructure and customized optimization
Centralized dashboard
for your jobs
Get a birds-eye view of your Databricks Workspace. Track total spend, total savings with Gradient, engineering hours saved, rate of SLAs met, and ROI all in one place.
Job-specific
return on investment
See how Gradient is doing. Access a detailed report on the initial state of your clusters and their current state, covering parameters such as data size, costs, costs per gigabyte, and spot market.
Control SLAs,
not cluster settings
Gradient allows users to simply set their desired runtime SLAs and walk away. Gradient does all of the heavy lifting involved in finding the right cluster configuration to consistently hit your SLAs without at the lowest price point.
Surfacing
low-hanging fruit
Identify your most expensive jobs and key optimization opportunities. Gradient’s machine learning models are fine-tuned using historical cluster metrics to ensure recommendations are tailored to your workloads.
Detect inefficiencies
and misconfigurations
Gradient helps you pinpoint inefficiencies, misconfigurations, and misuses of compute resources. See misuses of APC, Photon, and autoscaling, across your workspace. Discover compute cluster optimization opportunities with Gradient.
Granular
cost analysis
Understand your Databricks spend at a granular level. Analyze costs by job, customer, app ID, or other tags within the same job. Easily identify which customers are driving up costs and which are optimizing their usage.
Multi-dimensional reports
See job costs by customer, app-id, or any other custom tag. Analyze detailed Spark metrics per job and cross-reference them with job cost with performance indicators to get a holist view of the performance of your data pipelines.
Transparent optimization
See exactly how Gradient tunes every cluster configuration and the impact of each optimization. Every change and recommendation is automatically logged for simplified compliance and auditability.
Works with your data tools
Track cluster performance even when using data orchestrators like Airflow, Azure Data Factory, or Dagster. Connect. Assess job performance across several runs, even if your DAGs have multiple tasks.
Databricks health check
Get instant insights into your compute infrastructure with our free Databricks Workspace health check. Identify your top 10 expensive jobs, optimization opportunities, and understand your current cost and performance landscape.
Key
benefits:
All of the benefits of serverless without the drawbacks
Complete visibility
and control
Cost and time
savings
100% customized models
Self-improving constantly
Secure and compliant
I’d be surprised if there was any data team on the planet that wouldn’t save money and time from using Gradient.
Jesse Lancaster, CTO, Forma.ai
“Gradient is an excellent tool that delivers real results that you can see in your monthly [data infrastructure] bill.”
Deniz Parmaksız, Sr. Data Engineer, Insider
“With automated optimization and scalable solutions, Gradient guarantees seamless pipeline operation.”
Blaine Elliot, Data Engineer, Abnormal
Explore additional capabilities
Get started today
Leverage customized machine-learning models to take your compute to the next level