How to Centralize Proctoring Across 6,000 Workstations

A large organization with a distributed network of testing centers implemented TrustExam.ai to centralize control over exam sessions. The solution brought together around 6,000 workstations, biometric identity verification, video streams, and automated alerts in a single situation center. The project also included integration with the existing examination platform and a specialized ArtLinux-based client environment for thin clients.

The starting point

The organization runs large-scale computer-based testing across a distributed network of examination centers. The project covered around 6,000 workstations with different configurations, including legacy thin clients with limited hardware resources.

Video alone was not enough. The examination environment had to detect remote-access attempts, prohibited applications, signs of virtualization, and hardware changes. Identity was another critical requirement: the system had to confirm that the registered participant was taking the exam and continue verifying that identity throughout the session.

Before the unified control layer was introduced, these signals were difficult to collect and process centrally. Local duty teams and the central situation center needed one interface with a clear status for every workstation, live video, and a complete incident history.

Project objectives

• Unify all proctoring services and the existing examination platform in one workflow.

• Create a situation center for real-time monitoring across the entire distributed network.

• Control participant identity, the software environment, connected devices, and behavior during the exam.

• Run the solution on the existing fleet of thin clients without a large-scale hardware replacement.

• Centralize workstation updates and technical support.

Solution architecture

The system was divided into three infrastructure segments. Testing sites handle local video processing and the client application. The central data center hosts the core services, database, and interaction routing. Office infrastructure supports monitoring, administration, and the situation center.

This architecture connected workstations, server-side components, video services, alerts, and the existing testing platform. Operators gained a single interface, while technical teams received a centralized support and administration layer.

How an exam session works

Participant identification

Before the exam, the participant selects the test type and language, enters a unique identifier, and completes biometric identification with a liveness check. Once verification succeeds, the system grants access to the exam. Background biometric verification continues during the session to match the participant with the registered account.

Workstation verification

Before the session starts, TrustExam.ai checks the front-facing camera, microphone, overhead camera, and software environment. Controls cover signs of virtualization, prohibited processes, and remote-access tools. Hardware and software changes are recorded before testing begins and whenever a session is reinitialized.

Automated monitoring

During the exam, the system records video and audio, monitors attempts to minimize the application, blocks copy and paste, and detects third-party processes. Algorithms flag the participant leaving the frame, multiple faces, head turns, gaze deviations, and external voices. Each suspicious event becomes an alert with an exact timestamp and a link to the relevant workstation.

A unified situation center

Situation-center staff can see active exam sessions, workstation status, and incoming alerts. The same interface provides video from front-facing and overhead cameras, as well as room and corridor cameras. An operator can open an event, review the relevant video, and record the decision for the incident.

Local duty staff can access only their own testing center, while central proctors can monitor authorized sessions across the network. Permissions are role-based, and stronger authentication is used for authorized personnel. Every incident card retains the time, workstation, and event type, while an integrated chat supports escalation and the exchange of evidence.

A specialized solution for thin clients

Supporting legacy thin clients was one of the most difficult parts of the project. The standard client configuration could not provide the required stability on this hardware.

The team reworked the ArtLinux system environment, updated critical components, optimized performance, and corrected incoming and outgoing audio. Specialized client logic was developed for thin clients and integrated with both automated proctoring and the existing examination platform.

Joint testing in examination rooms confirmed that both systems could launch and operate reliably on ArtLinux thin clients. This made it possible to retain the existing equipment fleet and bring it into the common proctoring environment.

Centralized operations and support

To operate thousands of workstations, the project introduced centralized update delivery with installation control. The administration panel added remote restart and shutdown functions together with managed administrator authorization.

During peak exam periods, support specialists analyzed system logs, connected to workstation diagnostics, and helped local teams restore service. They resolved startup freezes, server connection errors, device-detection issues, virtual-machine checks, camera and microphone initialization problems, and restart failures after reboot. When recovery required more time, the participant could be moved to a reserve workstation and continue the exam.

The solution was also integrated with a Nimble Streamer-based video service. A computer-vision module for the overhead camera was tested to flag hands below the desk, mobile devices, and unauthorized objects in the testing area.

Implementation results

The organization received a unified control layer for exam sessions across distributed infrastructure. The situation center combines video, workstation status, and automated events, while local teams and central proctors operate within one role-based model.

• Around 6,000 workstations are included in the common monitoring workflow.

• Identity is verified before the exam and continuously checked during the session.

• Remote-access tools, virtualization, and prohibited processes are covered by examination-environment controls.

• Alerts are linked to video, event time, and the specific workstation.

• Thin clients operate in a specialized ArtLinux environment.

• Updates and basic workstation operations are managed centrally.

The project demonstrated that automated proctoring can be embedded into an existing distributed infrastructure while preserving the examination system and a heterogeneous equipment fleet. The key outcome was manageability: the organization can see the progress of the exam, receive structured risk signals, and respond from a single situation center.

A solution for large-scale testing

TrustExam.ai is designed for organizations that need to run exams simultaneously across many sites and workstations. It combines identity verification, device and software controls, video monitoring, automated alerts, and centralized operations tools.

If you are planning a large-scale, sector-specific, or corporate testing program, the TrustExam.ai team can assess your infrastructure and design a proctoring environment around your operational requirements.

Frequently asked questions

How does proctoring help control large-scale testing?

The system verifies the participant, devices, and software environment, records video streams, and creates alerts for suspicious events. A proctor reviews each event in the situation center and records the decision.

How does the system detect remote access?

Before and during the exam, the client monitors the software environment, prohibited processes, signs of virtualization, and remote-access tools. Deviations are recorded as monitoring events.

Can proctoring run on thin clients?

Yes. A specialized ArtLinux-based environment was prepared for the project infrastructure and jointly tested with the automated proctoring system and the existing examination platform.

What does a situation-center operator see?

The operator sees active sessions, workstation status, video streams from several camera types, automated alerts, and incident cards linked to the relevant time and computer.