5 reasons why observability and security work well together

Rowing_team.jpg

Site reliability engineers (SREs) and security analysts — despite having very different roles — share a lot of the same goals. They both employ proactive monitoring and incident response strategies to identify and address potential issues before they become service impacting. They also both prioritize organizational stability and resilience, aiming to minimize downtime and disruptions. 

Yet it’s when they both emphasize the importance of collaboration and communication not only within, but across their respective teams, that they achieve a higher level of operational resilience and proactively respond to potential threats to the business than if they were operating independently. 

Here are five ways security and observability teams accomplish more when working together:

1. Improved team collaboration

1. Improved team collaboration

When it comes to vendor and technology partners, the math is simple: More tools = more swivel chairing and more time spent resolving problems. One-off, siloed technologies increase the burden of collaboration and insights on teams, which increases toil, making it error-prone. The additional tools also increase the burden of managing, updating, and maintaining software. Unifying technologies not only reduces this manual workload but also reduces an organization’s attack surface by limiting the access tools have to the organization's systems. 

Collaboration in action

DISH Media's ad revenue business ingests and processes 10 billion records a day from 25 million device endpoints — including operational, business, and security data. With a unified solution and single agent through Elastic, dashboards and data are now available in a single pane of glass across teams for quick analysis.

This has significantly reduced incident detection and MTTR, improving customer experiences. Because of Elastic’s single agent, anomalies across millions of systems and customer devices are spotted much more quickly, accelerating root cause analysis and remediation. And with the team operating from a single agent, there were no additional implementation costs.

“With Elastic, we now have a unified view of data that we can correlate to detect patterns and anomalies,” says John Haskell, head of engineering at DISH Media. “In the past, root cause analysis and remediation could take weeks. Now it takes hours.”

View the full DISH Media story.

2. Complete visibility with a unified data platform

2. Complete visibility with a unified data platform

Observability and security teams are inundated with data as the complexity of the infrastructure and applications continues to increase. Often, this data is the same but used by the two teams differently. The separation of data into siloed tools creates artificial boundaries, which slows down problem detection and resolution from a performance as well as threat detection perspective. In addition, data generated from different systems is likely to be in different formats, which creates additional challenges with visibility across the organization. The ability to have a unified platform that relies on a common schema of ingested and stored data makes it easier to search and correlate relevant information, improving visibility across the organization.

Collaboration in action

OpenTelemetry, one of the highest velocity projects in the Cloud Native Computing Foundation (CNCF) ecosystem, is considered the de facto standard for telemetry data and is a widely adopted framework for both SRE and security teams. OTel Semantic Conventions framework helps users reduce the time and effort required for querying and correlating diverse data, building visualizations, and analyzing features for machine learning applications. 

Normalizing security and observability data with the OTel Semantic Conventions is a powerful tool that drastically reduces the complexity that so often hinders the efficient analysis of software, performance, and security issues. SRE and security teams along with technology vendors are embracing open data standards to enable holistic analysis of diverse and heterogeneous data.

3. Anomaly and threat detection

3. Anomaly and threat detection

The exponential growth in data along with the rapid pace of code and infrastructure deployment creates the challenge of finding anomalies and detecting threats before they become service-impacting. Using out-of-the-box and customizable machine learning (ML) models, AIOps capabilities help automatically detect anomalies providing root cause analysis and remediation support. The ability of observability solutions to reduce noise is dependent on the telemetry data including metrics, logs, traces, and profiling data. 

Logs, distributed tracing, and metrics provide a view into the request flow, amount, and type of requests along with other performance characteristics. This correlated and contextual data for distributed systems provides a comprehensive view of the application behavior, which can also be leveraged for investigating security incidents. The ability to analyze the data and identify deviations based on established historical baselines accelerates security investigations.

The evolution of generative AI and retrieval augmented generation (RAG) capabilities enables SRE and security teams to investigate and analyze further with interactive assistants that understand natural language and can provide quick answers to all levels of operations and security teams, reducing the time to resolution.

Collaboration in action

SIEM solutions and other security technologies that integrate with observability platforms leverage insights from logs, metrics, and traces. This unified approach enables the proactive identification of abnormal patterns, suspicious activities, and potential security incidents. 

By correlating abnormal log spikes in network traffic with server performance metrics, organizations can quickly distinguish between legitimate traffic surges and potential DDoS attacks. Unusual patterns, such as repeated login failures or access from unusual locations, are quickly surfaced — significantly reducing the likelihood of successful attacks.

4. Tool consolidation and reduced costs

4. Tool consolidation and reduced costs

Besides the increased visibility and proactive identification of issues, consolidating observability and security capabilities on a unified platform also leads to tool consolidation that delivers the added benefit of cost savings. A unified platform means reducing the total cost of ownership by bundling the associated operational fees, services, data storage, and personnel required to manage both practices. 

Collaboration in action

Enterprise cloud data management leader Informatica replaced its complex observability and SIEM solutions with Elastic’s unified platform. It boosted application performance while protecting systems from external threats — saving big on budget in the process.

"With Elastic, we have a single vendor for observability and SIEM. This represents a cost saving of 50% compared to other solutions for an organization of our size," says Amreth Chandrasehar, director of ML engineering, observability, and site reliability engineering at Informatica.

And performance need not be compromised by consolidation. In fact, Informatica found the opposite to be true. "Elastic's search functionality is incredibly fast,” explains Chandrasehar. “We store trillions of documents, but a search query returns accurate results in little more than 10 seconds."

View the full Informatica story.

5. Regulatory compliance for data handling

5. Regulatory compliance for data handling

Strengthening security practices helps organizations comply with industry regulations  that govern observability data handling. By aligning observability initiatives with strict compliance requirements, organizations not only avoid legal repercussions, but also instill trust in stakeholders.

This alignment facilitates the seamless integration of observability tools within a regulated environment. It also showcases the potential for a symbiotic relationship between security and observability in meeting those compliance standards.

Collaboration in action

No industry knows compliance requirements quite like the financial sector. Emirates NBD, one of the largest banking groups in the Middle East by assets, built a centralized logging system that crunches multiple terabytes of data a day from a multitude of data sources. With Elastic at its core, this new environment amounts to what Vice President of Cloud and Data Platforms at Emirates NBD Ali Rey characterizes as the foundation of a single source of truth. 

Centralized logging provides the bank with an avenue to beef up security and to store and retrieve audit logs required by governance stakeholders. "If there are any disputes, or any questions, queries or anything that happens from both an internal or external point of view, we've got these audit logs that haven't been tampered with," Rey says.

Thanks to its migration to centralized logging with Elastic, the bank has expanded from its original observability investment into security, which helps it detect both external and internal threats.

View the full Emirates NBD story.

Take the first step toward unified data visibility

When observability and security functions work in harmony, they ensure a more secure and reliable operational environment. A fortified security practice is not only an essential defensive measure for business and reputational health, but also a catalyst for the efficacy of observability tools. And, in a self-perpetuating cycle, security posture is further bolstered by discrepancies surfaced from observability monitoring. 

Relying on a unified data platform based on open standards for both security and observability practices may seem like a distant goal, but taking the initial steps today will prepare your organization for the long term. 

Read the SANS report, Shining a light in the dark: Observability + security, or watch the webinar to learn more about this emerging strategy and how you can take steps to unify your organization’s observability and security functions.

The release and timing of any features or functionality described in this post remain at Elastic's sole discretion. Any features or functionality not currently available may not be delivered on time or at all.