What observability can do for data security

Inside the vehicle, in the cloud, on mobile devices – the flow of data and the software environment in cars are becoming larger, more complex, and more fragmented. And more vulnerable to attacks. "The automotive industry is increasingly evolving into a software industry, selling cars online, maintaining vehicles, updating them via over-the-air updates, and monetizing data," says Ewald Munz, Head of Manufacturing, Automotive, and Sustainability EMEA at the software company Splunk. "Digitalization, autonomous driving, and increased connectivity offer many benefits but also introduce new challenges for cybersecurity."

New threats, more sophisticated attacks 

Cyberattacks are becoming increasingly sophisticated: "Attackers are constantly developing new methods to compromise vehicles. For example, zero-day exploits are used to take advantage of vulnerabilities in vehicle software," Munz reports. As a result of growing cyber risks, the automotive industry must address the need for a Vehicle Security Operations Center (VSOC). A major driving force behind the establishment of a VSOC, according to Munz, is UNECE WP.29, a global regulatory forum for harmonizing vehicle regulations.

This makes an observability strategy necessary, just like in any other software company, to maintain performance, availability, and cybersecurity. The key component for this is seamless real-time monitoring of the entire IT infrastructure. Analysts at Gartner predict that by 2026, around 70 percent of organizations that successfully implement observability will gain significant competitive advantages – ranging from fewer and shorter downtimes to more resilient IT and faster decision-making.

What experts mean by observability 

But what exactly does observability mean? "Observability is the ability to understand what is happening within a system based on external data published by that system," defines David Groombridge, an analyst in the IT Leaders and Technical Professionals team at Gartner Research & Advisory. "Observability requires actionable data from various sources to be appropriately connected, optimized, and contextually enriched." This opens up a broad field, as it involves analyzing everything that can be observed and measured in terms of data.

Suppliers Enhance Traffic Safety with V2X

In 2023, the number of traffic accidents in Germany increased again. Technologies like Bosch's Road Hazard Service and ZF's Rescue Connect could help reduce accidents and make emergency services more efficient.

Ronja Schmiedchen

Nevertheless, there is some confusion about the term: "There is no clear and precise definition," says Munz. "We are experiencing a phase of observability-washing, where every vendor or IT expert has their own definition." The core idea, however, is straightforward: "The key to observability is the availability of data." This highlights a fundamental requirement for a solid observability strategy. The essential question is: Do I have the data I need to make my system 'observable' so that it can be properly monitored? Munz concludes, "The more observable the system is, the better the monitoring." An observability team must ensure that all software and IT services function smoothly and transparently – inside the vehicle, in the cloud, on mobile devices, and with third-party providers.

What to consider regarding observability solutions 

"With regard to the necessary infrastructure monitoring, the backend systems of the industry resemble those of a typical corporate IT environment," says Munz. "Given this, automotive companies must consider where and how an attacker might attempt to infiltrate them." Attack vectors could include the mobile app, dealer connections, or even compromised vehicle backend connections. Munz adds, "End-to-end visibility is essential, whether for production processes in manufacturing, customer-facing service processes, or mission-critical business processes tailored to the company."

BMW employs a multi-tiered approach to implement observability solutions, ensuring vertical, horizontal, global, and predictive end-to-end visibility. These solutions capture all telemetry data in real-time in the form of logs, events, metrics, and traces, providing information on software transactions, errors, performance, customer interactions, and IT system health. Observability platforms aggregate and correlate telemetry data, transforming it into insights with dashboards, alerts, and analysis and troubleshooting features.

How to implement an observability strategy 

Without available data, there can be no observability – that much is clear. "The first step is to establish an 'industrialized' way of making data available," Munz explains. A good method for collecting observability data, according to the expert, is using OpenTelemetry (OTel). OTel is an open-source project under the Cloud Native Computing Foundation. This community-driven project aims to use a single open-source agent for collecting all relevant telemetry data, with full control over how the data is collected, transformed, and filtered before being sent to an observability solution.

"Once the data collection strategy is defined, the next step is to determine where the data should be sent and which observability platform to use," Munz explains. He also observes a washing effect in this area, as all observability solutions can collect logs, metrics, and traces, but not in the same way. "We recommend a streaming-based solution to achieve real-time results as closely as possible and to avoid using sample data, ensuring that the root cause of problems is always identified."

Modern monitoring solutions have evolved significantly, but they are often still inadequate. "Therefore, consistent observability is essential," Munz emphasizes. "A proper solution provides important context, identifies the most significant problem areas, and helps teams optimize service performance and customer experience."

This article was first published at automotiveit.eu

How Generative AI is Changing the Automotive Industry

Generative AI has long reached the peak of its own hype. There are many promising applications in the automotive industry, but the reality check is yet to come. What can be expected from ChatGPT and others in development, production, or sales?

Daniela Hoffmann-Topp