Case Study
How Parity Achieved Streamlined Data Excellence
A Case Study in HVAC Optimization and Modern Data Infrastructure
How Parity Achieved Streamlined Data Excellence
A Case Study in HVAC Optimization and Modern Data Infrastructure
About Parity
Parity is an HVAC Optimization as a Service company. They use software to deliver automatic, guaranteed energy and cost savings to multifamily residential and hospitality clients.
Parity's Optimizer service slashes energy use and CO2 emissions while reducing on-site staff workloads by combining real-time data, automation, remote monitoring and control, and expert advice. Parity micromanages HVAC systems, monitoring performance and controlling set points remotely, with minimal time and attention required from facility staff.
Parity's energy experts also provide extra protection for your building's HVAC system by tracking patterns and identifying potential issues. In the face of tightening energy efficiency and greenhouse gas emissions regulations, Parity provides low-cost opportunities for compliance and access to government incentive programs.
About Parity
Parity is an HVAC Optimization as a Service company. They use software to deliver automatic, guaranteed energy and cost savings to multifamily residential and hospitality clients.
Parity's Optimizer service slashes energy use and CO2 emissions while reducing on-site staff workloads by combining real-time data, automation, remote monitoring and control, and expert advice. Parity micromanages HVAC systems, monitoring performance and controlling set points remotely, with minimal time and attention required from facility staff.
Parity's energy experts also provide extra protection for your building's HVAC system by tracking patterns and identifying potential issues. In the face of tightening energy efficiency and greenhouse gas emissions regulations, Parity provides low-cost opportunities for compliance and access to government incentive programs.
Delivering a streamlined data pipeline
The data pipeline we created for Parity serves multiple purposes, one of which is simplifying access to information. It also ensures scalability for current and future needs and improves data quality, management, transparency, and security.
Additionally, it automates the process of storing data in the cloud. It integrates data across all company assets to create automated business critical reports such as their Energy Savings Report.
The pipeline detects duplicated or incorrect records, aggregates monthly/daily/hourly energy consumption for reporting, and archives real-time data for easy access. It establishes a data warehouse for historical data, enabling mechanical and chemical engineers to analyze data without requiring software development teams. Furthermore, it facilitates the development of new internal products for business growth while simultaneously creating a single, centralized source of truth for the entire company.
With the data pipeline in place, Parity enabled their teams with more efficient internal Research and Development, allowing them to explore aggregated data to quote additional properties with similar conditions.
Delivering a streamlined data pipeline
The data pipeline we created for Parity serves multiple purposes, one of which is simplifying access to information. It also ensures scalability for current and future needs and improves data quality, management, transparency, and security.
Additionally, it automates the process of storing data in the cloud. It integrates data across all company assets to create automated business critical reports such as their Energy Savings Report.
The pipeline detects duplicated or incorrect records, aggregates monthly/daily/hourly energy consumption for reporting, and archives real-time data for easy access. It establishes a data warehouse for historical data, enabling mechanical and chemical engineers to analyze data without requiring software development teams. Furthermore, it facilitates the development of new internal products for business growth while simultaneously creating a single, centralized source of truth for the entire company.
With the data pipeline in place, Parity enabled their teams with more efficient internal Research and Development, allowing them to explore aggregated data to quote additional properties with similar conditions.
The main challenge
Parity's data was siloed in multiple locations, formats, and data structures (AWS RDS Postgresql, TimescaleDB, One-time ingest CSVs, client PDF utility bills, and complex Excel workbooks).
They used this data to evaluate energy consumption and savings in buildings. This process was highly manual and labor-intensive. Parity wanted to automate the data integration pipeline to enable them to scale and add additional facilities.
By pulling all the data into Snowflake, they wanted to create a unique source of truth. Previously, Parity used Excel to integrate data from multiple sources and manually create reporting data for each Building. In some cases, a Django app was used as well. However, it had challenges and required manual intervention 50% of the time.
How we helped Parity
The Future Architecture of the Parity's data platform, suggested by Infostrux, follows best practices, well-known and proven architecture, and design patterns. The platform's success is based on choosing the right tools and technologies that are the best fit, easy to implement, highly available, extensible, and scalable.
The team started by quickly defining and implementing the necessary configuration to properly integrate Snowflake with Parity's AWS-based infrastructure, leveraging Infostrux's accelerators to avoid wasting time during the setup phase and sidestepping common pitfalls.
This solution allowed us more time to focus on working with the business to properly model data and provide a production-ready data model for the team. Our team improved their platform stability by leveraging managed services for most of the stack, as Parity had no internal team to manage and support.
Additionally, provided business and R & D users with access to raw and curated data in Sigma so users are able to explore and self-serve reports and dashboards. Even taking it a step further by using Sigma Input Tables so users can add data manually to create varying benchmarks and data simulations.
Sigma’s functionality gave Parity enough flexibility without requiring them to be data experts.
The main challenge
Parity's data was siloed in multiple locations, formats, and data structures (AWS RDS Postgresql, TimescaleDB, One-time ingest CSVs, client PDF utility bills, and complex Excel workbooks).
They used this data to evaluate energy consumption and savings in buildings. This process was highly manual and labor-intensive. Parity wanted to automate the data integration pipeline to enable them to scale and add additional facilities.
By pulling all the data into Snowflake, they wanted to create a unique source of truth. Previously, Parity used Excel to integrate data from multiple sources and manually create reporting data for each Building. In some cases, a Django app was used as well. However, it had challenges and required manual intervention 50% of the time.
How we helped Parity
The Future Architecture of the Parity's data platform, suggested by Infostrux, follows best practices, well-known and proven architecture, and design patterns. The platform's success is based on choosing the right tools and technologies that are the best fit, easy to implement, highly available, extensible, and scalable.
The team started by quickly defining and implementing the necessary configuration to properly integrate Snowflake with Parity's AWS-based infrastructure, leveraging Infostrux's accelerators to avoid wasting time during the setup phase and sidestepping common pitfalls.
This solution allowed us more time to focus on working with the business to properly model data and provide a production-ready data model for the team. Our team improved their platform stability by leveraging managed services for most of the stack, as Parity had no internal team to manage and support.
Additionally, provided business and R & D users with access to raw and curated data in Sigma so users are able to explore and self-serve reports and dashboards. Even taking it a step further by using Sigma Input Tables so users can add data manually to create varying benchmarks and data simulations.
Sigma’s functionality gave Parity enough flexibility without requiring them to be data experts.
Key metrics and achieved goals resulting from their data project
- Reports & Dashboards
- Energy Savings Report - Aggregated report mainly in monthly granularity but can be accessed in more granular time frames such as hour, daily, monthly, quarterly, and yearly (Performance vs. cost dependent).
- Research and Development - Ability to query raw sensor data aggregated by the hour
- Implementation of Sigma Input tables to provide the ability to extend data for specific use cases.
- Parity now has increased trust in their data as it's centralized.
- Energy Savings Report - Aggregated report mainly in monthly granularity but can be accessed in more granular time frames such as hour, daily, monthly, quarterly, and yearly (Performance vs. cost dependent).
- Modern data stack approach
- ELT (as opposed to ETL) - delivering fast data ingestion and the ability to move all raw data from a multitude of sources into a single repository (Snowflake) and well governed have unfettered access to that data
- Cloud-based extract and load platform -Rivery
- Cloud data platform - Snowflake, Sigma, AWS, TimescaleDB, Arcadia
- DevOps and DataOps friendly transformations - dbt
- A high degree of automation with GitHub Actions-based data pipelines
- ELT (as opposed to ETL) - delivering fast data ingestion and the ability to move all raw data from a multitude of sources into a single repository (Snowflake) and well governed have unfettered access to that data
- Proven data architecture
- Layered data processing offering clear separation of concerns
- Business domain-focused components
- Wide table support for data analysis and visualization, enabling ad-hoc reporting
- Extensibility for data-science workloads
- Layered data processing offering clear separation of concerns
Key metrics and achieved goals resulting from their data project
- Reports & Dashboards
- Energy Savings Report - Aggregated report mainly in monthly granularity but can be accessed in more granular time frames such as hour, daily, monthly, quarterly, and yearly (Performance vs. cost dependent).
- Research and Development - Ability to query raw sensor data aggregated by the hour
- Implementation of Sigma Input tables to provide the ability to extend data for specific use cases.
- Parity now has increased trust in their data as it's centralized.
- Energy Savings Report - Aggregated report mainly in monthly granularity but can be accessed in more granular time frames such as hour, daily, monthly, quarterly, and yearly (Performance vs. cost dependent).
- Modern data stack approach
- ELT (as opposed to ETL) - delivering fast data ingestion and the ability to move all raw data from a multitude of sources into a single repository (Snowflake) and well governed have unfettered access to that data
- Cloud-based extract and load platform -Rivery
- Cloud data platform - Snowflake, Sigma, AWS, TimescaleDB, Arcadia
- DevOps and DataOps friendly transformations - dbt
- A high degree of automation with GitHub Actions-based data pipelines
- ELT (as opposed to ETL) - delivering fast data ingestion and the ability to move all raw data from a multitude of sources into a single repository (Snowflake) and well governed have unfettered access to that data
- Proven data architecture
- Layered data processing offering clear separation of concerns
- Business domain-focused components
- Wide table support for data analysis and visualization, enabling ad-hoc reporting
- Extensibility for data-science workloads
- Layered data processing offering clear separation of concerns
How Parity is using the solution
- Parity has two significant use cases:
- Monthly Energy Report and Weekly Operational Report. Before the deployment of this tool, reporting generation would take hours per report; now, they are down to minutes and seconds, respectively.
-
- Parity is seeing approximately a 75% decrease in time spent per report, in some cases more.
- As the project progresses into full automation for obtaining utility data, Parity expects a 50% reduction in time once this part of the tool is fully built out with their customers.
- Parity is seeing approximately a 75% decrease in time spent per report, in some cases more.
What Parity has to say
"We had the pleasure of working with the exceptional team at Infostrux (Mustafa/Alex/Negar) as data consultants for our project. Their expertise and responsiveness were evident from the start, and they quickly grasped the complexities of our product and data architecture challenge. Collaborating closely with us throughout the project, they developed a tailored rollout plan that perfectly aligned with our goals.
Thanks to Infostrux's excellent work, we've seen remarkable improvements in our time-to-report and overall data quality. The solutions they provided have had a lasting impact on our approach to data. We continue to build upon their work delivered in Spring 2023; the benefits are already evident."
Are you ready to leap forward with your data?
No matter where you are in your data cloud journey or what industry you come from, our team of experts is ready to embed themselves into your existing structure, pinpoint the value in your data, and help you achieve your business goals.
True innovation with your data awaits. Are you ready?
