Azure Time Series Insights Gen2: Leading the next generation of industrial IoT analytics platforms

2020年7月16日 に投稿済み

Principal PM Manager, Azure IoT

The Internet of Things (IoT) is well-established for helping businesses find real-time insights from their industrial assets opening the path towards Industry 4.0. Answering questions like “how are all of my assets performing right now?” or “how can I improve my manufacturing process and attainment?” and “when will my assets need servicing?” used to be impossible to know or required manual data collection that was always out of date.

Today, business leaders are taking advantage of IoT to see this information with the click of a button. Yet as larger volumes of data are collected from industrial assets, finding insights can become more and more difficult. It can start to require costly and time-consuming data wrangling and data analytics techniques performed by highly specialized staff.

This is where Azure Time Series Insights Gen2 comes in. This fully managed IoT analytics platform—generally available today—enables you to uncover hidden trends, spot anomalies, and conduct root-cause analysis in large volumes of industrial data with an intuitive and straightforward user experience. Simple yet powerful, Azure Time Series Insights Gen2 allows you to explore and analyze billions of contextualized events across millions of sensors.

Since Azure Times Series Insights Gen2 is a serverless offering, you don’t have to worry about managing complicated compute clusters yourself. Additionally, Azure Time Series Insights Gen2 provides a scalable, pay-as-you-go pricing model enabling you to tune your usage to your business demands.

Azure Time Series Insights Gen2 is both a web experience and a platform. Knowledge workers can use the Time Series Explorer web experience to find insights from petabytes of IoT data in seconds through the simple, intuitive user interface. Developers can use the open and scalable platform to build solutions and custom user experiences with our rich APIs and JavaScript SDKs.

The Azure Time Series Explorer user interface displays three different contextualized.

Azure Time Series Insights Gen2 is tailored for industrial IoT applications.

Driven by feedback from customers around the globe, here are key features that are now generally available and how they benefit industrial IoT customers.

Azure Time Series Insights Gen2 offers multi-layered storage

IoT customers work with IoT data in a variety of ways. The two most common scenarios we see are:

  • Highly interactive analytics over a short time span.
  • Advanced analysis of decades worth of historical data.

Azure Time Series Insights Gen2 covers both scenarios with retention-based data routing between managed warm and bring your own cold stores, including Azure Data Lake Storage. Warm store can be configured to retain up to 31 days of IoT data allowing you to perform highly interactive asset-centric analytics with low latency to monitor, trend, and troubleshoot your assets. Cold store, with its near-infinite, retention can be used to store decades worth of historical IoT data, ready to be used for operational intelligence and improved efficiencies.

Azure Time Series Insights Gen2 provides retention-based data routing between managed warm and “bring your own” cold stores.

Multi-layered storage.

Enterprise scale to power the analytics needs of industrial customers

Azure Time Series Insights Gen2 powers the analytics needs of many industrial customers across all major segments, including manufacturing, power and utilities, oil and gas, automotive, smart buildings, and mining. These customers generate billions of events across millions of data points, with most struggling to keep pace with the vast amounts of data generated by their assets. Azure Time Series Insights Gen2 scales to accommodate high volumes of data quickly and efficiently. Alongside our scalable storage options, Azure Time Series Insights Gen2 supports one-million-time series instances (or tags) per environment with rich semantic modeling. This allows you to seamlessly explore highly contextualized data and correlate trends across your industrial assets to unlock insights and achieve operational excellence.

Azure Time Series Insights Gen2 supports one-million-time series instances (or tags) per environment with rich semantic modeling.

Azure Time Series Gen2 supports one million tag instances.

Microsoft Power BI connecter helps bring your data silos together

The ability to bring your data silos together is important to make data driven decisions and drive digital transformation. Azure Time Series Insights Gen2 provides an out of the box Power BI connector which connects your Azure Time Series Insights Gen2 queries to a Power BI workspace. You can easily view your time series and business intelligence data in a single pane of glass to make better decisions with a holistic view of your business posture.

Azure Time Series provides an out of the box Power BI connector which connects your Azure Time Series queries to a Power BI workspace.

Azure Time Series Gen2 integrates with Power BI.

Contextualize raw telemetry with the Time Series Model

Traditionally, the data that's collected from IoT devices lacks contextual information, which makes it difficult to use for business purposes. The Time Series Model, within Azure Time Series Insights Gen2, allows you to contextualize raw telemetry by defining hierarchies, instance properties, and types. This makes your analysis of asset-centric data simple and more valuable to your organization.

It’s easy to get started with Time Series Model using Time Series Explorer to both author and curate your model. Alternatively, the Time Series Model can also be managed through our rich API surface.

The Time Series Model, within Azure Time Series Insights Gen2, allows you to contextualize raw telemetry.

The Time Series Model, within Azure Time Series Insights Gen2, allows you to contextualize raw telemetry.

Gain insights using Azure Time Series Insights Gen2 with Azure Digital Twins

Achieve even greater insights by integrating Azure Time Series Insights Gen2 and Azure Digital Twins. Azure Digital Twins allows you to fully model your physical environment and stream live IoT data for a complete view of your connected assets and environments. Understand how your assets, customers, and processes interact in both real and simulated environments.

 

Gain greater insights by coming Azure Time Series Insights Gen2 with Azure Digital Twins.

Gain greater insights using Azure Time Series Insights Gen2 with Azure Digital Twins.

Open and flexible integration

Azure Time Series Insights Gen2 can be used with tools you know and love. Our cold store is backed by a customer-owned Azure Data Lake. Combining Azure Data Lake storage with our native support for the open source, highly efficient Apache Parquet lets you dive into decades of historical IoT data.

In addition, Azure Time Series Insights Gen2 ships with a Power BI connector allowing customers to export the time series queries they create in Azure Time Series Insights Gen2 into Power BI and view their time series data alongside other business data. Other highly sought-after connectors for popular analytics platforms such as Apache Spark™, Databricks, and Synapse will become available over time.

Time Series Explorer—analytics tool for knowledge workers and developers

The first-class user experience of the Time Series Explorer lets you use interpolation, scalar and aggregate functions, categorical variables, scatter plots, and time shifting of time series signals to analyze the data.

Time Series Explorer features the following user experience capabilities:

  • Automatically refresh charts.
  • Reverse lookup instance placement within the hierarchy.
  • Select and chart multiple variables through a single operation.
  • View chart statics.
  • Create marker annotations.
  • Duplicate time series instances in the well and change variables.
  • Change the line colors through the new color picker tool.
  • Use swim lanes to group related time series together.

New rich query APIs now give you the ability to use interpolation, new scalar and aggregate functions and categorical variables outside of the Time Series Explorer.

Time Series Explorer features the following API capabilities:

  • Interpolate patterns from existing data to reconstruct time series signals.
  • Process discrete signals using categorial variables.
  • Apply trigonometric functions to identify patterns.
  • Calculate time weighted averages.
  • Leverage new APIs for hierarchy traversal, time series search, auto-complete, paths, and facets.
  • Query data at scale with improved search and navigation efficiency.
  • Leverage new conditional logic, such as IFF, which allows you to determine if an expression is true or false when selecting what data should be considered for computation. When used with categorical variables, you can create threshold monitors and map ranges of values to their categories.

Customers are using Azure Time Series Insights to gain business insights in manufacturing, power and utilities, oil and gas, automotive, smart buildings, and mining.

Fonterra empowers employees with data

Founded in 2001, Fonterra is the world’s second largest dairy processor, responsible for approximately 30 percent of global dairy exports. Owned by over 10,000 New Zealand farmers, the co-operative operates in over 100 countries and processes approximately 22 billion liters of milk each year.

In 2018, Fonterra made a decision to fast-forward their digital transformation. After a lengthy review, Microsoft was chosen to upgrade their old system with a new, cutting-edge, cloud-based platform. Renamed the “New Historian,” the updated system promises to deliver on their goal of becoming a data driven organization by giving their operators, leaders, data scientists, and business intelligence teams the power to use data more intelligently.

"Fonterra is embracing advanced technologies to transform into a data-driven organization. We selected Azure Time Series Insights to provide storage, contextualization, and analysis capabilities and replace our legacy on-premises historian. This will allow us to effectively consolidate our data to empower operators, leaders, data scientists, and business intelligence teams." —Tristan Hunter, General Manager of Automation and Operational Technology, Fonterra

ENGIE Digital supports thousands of assets

ENGIE Digital, a provider of renewable energy, delivers energy and provides energy-related services to millions of consumers in more than 50 countries. ENGIE Digital designs, builds, and runs unique solutions that help other ENGIE Digital business units by supporting their development and operations. ENGIE Digital uses an in-house operational platform to collect and process millions of IoT signals every second from thousands of wind, solar, biogas, and hydroelectric energy assets around the globe—often in real-time.

ENGIE Digital selected Azure Time Series Insights and Microsoft Azure IoT Edge to modernize its platform. With these updates, the platform now supports ENGIE Digital teams across hundreds of renewable energy sites worldwide.

Azure Time Series Insights is a foolproof solution. Its scalability, resilience, performance, and cost-effectiveness mean we always have the latest data at hand.” —Sebastien Gauthier, Head of Darwin Delivery, ENGIE Digital, energy and energy-related service provider

ShookIOT leverages Azure Time Series Insights to deliver customer insights

Oil and gas industry veterans, Dr. Dave Shook and Leanna Chan, have spent twenty years consulting with clients in the oil and gas industry. Time and time again, they see oil and gas companies struggling to leverage the full value of their data.

Traditionally companies store data in on-premises time-series database applications called historians; legacy operational technology (OT) tools that keep data siloed. This makes it difficult to connect with powerful information technology (IT) tools, such as cloud-based analytics. Additionally, collecting process data can be prohibitively expensive. Some process manufacturers store less than 75 percent of their data.

To address these challenges, the two entrepreneurs had a vision to fuse OT data with IT. They founded ShookIOT in Edmonton, Alberta, Canada in 2017. Their philosophy was to free data siloed on-premises and migrate it to the cloud—specifically the ShookIOT Fusion Cloud Historian running on Microsoft Azure. Once in the cloud, customers, such as Chevron, could harness the full value of their data leverage tools like Azure Time Series Insights.

“After our customer’s data and contextual information is stored in Azure, we leverage tools like Azure Time Series Insights to view data trends and Power BI to create data visualizations.” —Dave Shook, Co-Founder and CEO, ShookIOT

“ShookIOT Fusion improves upon the traditional long-term data storage found at most sites, leverages the Microsoft Azure cloud platform and accelerates all Azure analytics tools by providing operational and business data with context to users. —Leanna Chan, Co-Founder and Chief Revenue Officer, ShookIOT

Gain insights from large volumes of data easily

Explore and analyze billions of contextualized events across millions of industrial sensors. Uncover hidden trends, spot anomalies, and conduct root-cause analysis in large volumes of data with an intuitive and straightforward user experience. We’re excited to see how you use Azure Time Series Insights Gen2 to drive your digital transformation.

See the following resources to learn more: