Customer Case Study: Quby - Databricks

Quby

Customer Case Study

Quby

Quby is the technology company behind Toon, the smart energy management device that gives people control over their energy usage, their comfort, the security of their homes, and much more. Quby's smart devices are in hundreds of thousands of homes across Europe. As such, they maintain Europe’s largest energy dataset, consisting of petabytes of IoT data, collected from sensors on appliances throughout the home. With this data they are on a mission to help their customers live more comfortable lives while reducing energy consumption through personalized energy usage recommendations.

Industry

Consumer Technology

Vertical Use Case

Personalized Energy Usage Recommendations – Leverage machine learning and IOT data to power their Waste Checker app which provides personalized recommendations to reduce in-home energy consumption.

Technical Use Case

  • Data Ingest and ETL
  • Streaming
  • Machine Learning
  • Deep Learning

Customer Testimonial: Stephen Galsworthy, Head of Data Science at Quby

The Challenges

Through their connected home platform Quby is on a mission to improve the lives of their customers while helping reduce their energy footprint. Core this strategy is analyzing the 3+ petabytes of appliance and energy usage data they’ve collected from hundreds of thousands of homes in Europe. On top of that is managing over 1 million unsupervised machine learning models to power the personalized recommendations delivered through their Waste Checker app. As Quby built out these capabilities they ran into a number of challenges:

  • On-premise Hadoop cluster was very difficult and costly to scale to meet their big data collection needs.
  • The amount of time to manage infrastructure was too high – often times a data engineer would spend 1-2 days setting up clusters to ensure a new package was installed correctly and dependencies well managed.
  • Data pipelines were brittle and struggled to reliably ingest terabytes of streaming and batch data each day.
  • Data scientists were struggling to track the performance of their 1+ million models, making it hard to iterate and improve their in-app services.

The Solution

Databricks provides Quby with a Unified Data Analytics Platform that has fostered a scalable and collaborative environment across data science and engineering, allowing data teams to more quickly innovate and deliver ML-powered services to Quby’s customers.

  • Fully managed platform with automated cluster management simplifies the infrastructure and operations at any scale.
  • Collaborative notebook environment with support for multiple languages (SQL, Scala, Python, R) enables a diverse team of users to work together in their preferred language.
  • Native support for MLflow enables data science teams to easily track model performance and rapidly iterate across their millions of models in a systematic fashion.
  • Able to easily leverage the latest deep learning frameworks and technologies such as TensorFlow and Keras with built-in ML libraries.
  • Native support for Delta Lake allows their data engineering team to reliably run and scale both batch and streaming pipelines on the same data. Features such as Schema Enforcement help prevent the errors that they previously experienced.

The Results

With Databricks, Quby has been able to deliver on their mission: leverage machine learning  to improve the comfort and lives of their customers while helping reduce energy consumption.

  • Lowered Costs – cost saving features provided by Databricks (such as auto-scaling clusters and Spot instances) has helped Quby significantly reduce the operational costs of managing infrastructure, while still being able to process large amounts of data.
  • Reduced Energy Consumption – through their Waste Checker app, Quby has identified over 67 million kilowatt hours of energy that can be saved by leveraging their personalized recommendations.
  • Faster Innovation – with their legacy architecture, moving from proof of concept to production took over 12 months. Now with Databricks, the same process takes less than eight weeks. This enables Quby’s data teams to develop new ML-powered features for their customers much faster.

Databricks, through the power of Delta Lake and Structured Streaming, allows us to deliver alerts and recommendations to our customers with a very limited latency, so they’re able to react to problems or make adjustments within their home before it affects their comfort levels.

Steven Galsworthy: Head of Data Science at Quby

Discover more

TryDatabricks for free. Learnmore about our product. Talkto an expert.

Databricks Inc.
160 Spear Street, 13th Floor
San Francisco, CA 94105
1-866-330-0121

Contact Us

© Databricks . All rights reserved. Apache, Apache Spark, Spark and the Spark logo are trademarks of the Apache Software Foundation.
Privacy Policy | Terms of Use