What is telemetry?

Telemetry is the automated remote collection and analysis of data through either wired or wireless communications. Telemetry is used throughout many industries, including information technology (IT), healthcare, aerospace, and more. 

Telemetry is a key part of making observability possible. Instruments, such as hardware and software, measure characteristics and produce the data that stands as the output of telemetry.

How does telemetry work?

For telemetry to work properly, the process requires input from multiple disciplines, including data science, communications, and storage science. Each of these disciplines work together in harmony to make telemetry possible. Consider the following:

• Identify telemetry requirements: To get clear answers from data, you have to start with a specific question. Based on this question and key metrics that can answer it, data scientists and IT personnel decide what data needs to be collected from the remote system.
• Data collection: Engineers and developers set up a remote system of instruments to collect the necessary data, including software agents or hardware.
• Data formatting and transmission: The data is formatted into an appropriate stream for transmission through wired or wireless communications. This formatting includes protecting the data from “bad actors” (i.e. a person or group that exploits data with malicious intent). The data stream is sent to a central receiver, where it’s deciphered and formatted into the wanted data elements, and then stored for analysis.
• Data storage and analysis: Telemetry data is stored in a data warehouse or data lake, capable of securely holding large quantities of data. Once data has been stored, the observability system analyzes it for insights using ML, structured data analytics, and other modern data analysis techniques.

What is telemetry data?

Telemetry data includes various characteristics IT and data that scientists deem necessary to understand a system. These can include: 

• Environmental data: Characteristics that a system uses to operate. For instance, in the field of aeronautics and spacecraft, environmental data could include temperature, pressure, and velocity.
• Performance data: Examines how a system is performing, giving insights into what’s working efficiently and what needs improvement. Common metrics for performance data might include CPU cycles, throughput, and power usage. 
• Operational data: Essentially a health report for a system, showing if it’s running properly. Data may include errors, uptime or downtime, user data, or page loads, among others. 

This data is collected into the telemetry data stream for later analysis.

What's the difference between telemetry, monitoring, and observability?

Observability, monitoring, and telemetry are all related. However, they differ in their own ways, forming part of a larger system of data collection, testing, and analysis.

• Telemetry: Provides the data that is actively monitored by personnel or systems and also deeply analyzed within an observability practice.
• Monitoring: Involves actively watching and testing specific types of data for critical information. Monitored data can be used in observability analysis, but all data collected for observability may not be monitored.
• Observability: The larger practice of deeper analysis from the telemetry data that can lead to insight.

Benefits of telemetry

Telemetry is the process of gathering, measuring, and transmitting critical data from remote sources that can be used in business, science, research, defense, and other endeavors. The use of telemetry in IT to gather data for observability impacts developers, data scientists and analysts, business managers, and end users. Different teams rely on telemetry data in various ways:

• DevOps teams get quick feedback on how their applications are running as soon as they are deployed. They can quickly identify issues and optimize software and systems using data instead of guesswork. They can also see what features customers like and don’t like.
• Business managers and marketers rely on telemetry data to understand how their products and services are impacting their markets and customers. They can see results of real-time transactions and trends to understand what motivates buyers, what they are buying, and revenue impact.
• Customer service agents use telemetry data from user monitoring and other application performance monitoring (APM) tools to quickly solve customer problems.
• IT personnel can quickly locate objects, such as mobile phones or laptops, using telemetry tracking. Telemetry tracking uses GPS, radio signals, or even satellites to locate a device providing a “ping” to a locating system.

Challenges of telemetry

All technology practices have their challenges. Telemetry is no different. To create an effective telemetry strategy, it’s important to understand what factors can pose a roadblock. 

• Too much data: The trend in today’s data-driven business environment is to instrument and monitor everything, because more data means more possibilities to gain insights. However, too much data can lead to “analysis paralysis.” Be purposeful in choosing metrics and data to analyze. The communications technologies used and the time-dependency for your analytics will determine how much data you can send and receive.
• Bandwidth and latency: These go hand in hand with the volume of data you collect. Fortunately, wired and wireless communications technologies are getting faster with dramatically lower latency. However, the more frequently you need additional data, the greater the demand will be for faster—and potentially more expensive—technologies.
• Data protection: As bad actors get more sophisticated with their attacks and ability to steal data, privacy and protection is a chief concern. Encryption is a foregone conclusion, but communications and storage systems must be designed to limit access and protect against breaches. There’s also the concern of the sensitivity of any personal data you collect and how you protect it or anonymize it before transmitting it through your telemetry system.
• Interoperability: When a telemetry system feeds multiple clients and downstream systems, it’s important to make sure data is correctly formatted and deciphered so that each of these systems receive accurate, error-free data. 
• Cost: While telemetry is critical for optimizing operations, it’s just one of many tasks IT teams must oversee. Maintaining a constantly evolving suite of communications systems, software, and hardware needed to gather telemetry data is an intricate and time-consuming process. Budgets must accommodate these services appropriately. 

Telemetry is all about continuously and safely getting the right data to your observability system. From there, the challenges become observability system challenges, including timeliness of analysis and types of analytics.

How industries leverage telemetry

Nearly every industry relies on telemetry, because business today is data-driven, and data is supplied through some kind of telemetry. Here are just a few of the industries using telemetry:

• IT:  Business infrastructures are driven by their IT practice. IT relies on data to understand and optimize infrastructure behavior. From office computers to the Internet of Things (IoT) and security systems on a manufacturing floor, various industries use different instruments to supply telemetry data to IT, helping personnel to work smarter and more efficiently.
• Healthcare: Healthcare around the world is experiencing a caregiver shortage alongside rising populations and more patients in hospitals. Telemetry enables those caregivers to maintain a high quality of care with fewer resources in a more timely manner. Telemetry also provides the backbone for remote care, such as in-home monitoring and remote surgery.
• Automotive: Advanced driver assistance systems (ADAS) depend on telemetry throughout the vehicle. The future of widespread use of autonomous vehicles will depend on telemetry—within the vehicle, between vehicles, and between vehicles and municipal data systems. Vehicle networks, Bluetooth, and ultra-wideband (UWB) are all technologies in-use today to transmit telemetry data in the auto industry.
• Weather: Modern weather science couldn’t exist without telemetry. Telemetry data is provided by geosynchronous satellites, surface and submerged ocean buoys, and sensors atop the highest mountains, lowest valleys, and countless homes. Weather monitoring, observance, and analytics systems have access to massive amounts of telemetry data.
• Astrophysics: We know more about the universe today than ever before because of telemetry data from satellites and space telescopes. We can now  identify new objects throughout the universe due to instruments in near and deep space that send data back to Earth scientists.
• Defense: From space vehicles and unmanned air vehicles to the soldier on the battlefield, defense strategies depend on immense amounts of telemetry data.