Bianca Andrei
Software Developer
Root Cause Analysis — Transforming chaos into clarity
Root Cause Analysis for Modern Telecom and IT Ops
Rad1X unifies observability across domains, correlates telemetry in real time, and isolates the single failing component responsible for service disruptions.
Unified Observability. Clear Root Cause. Faster Resolution.
Reduce MTTR by up to 80%AI/ML predictive assurance + Explainable AI (XAI)TMF Autonomous Networks Level 4 EnablerTM Forum aligned (TMF638/TMF642/TMF621) + closed-loop ready (TMF921/TMF753/TMF640/TMF664/TMF641)
Rad1X
The Challenge
The realities of operating complex, mission-critical systems
Complexity in Modern Networks
5G Standalone, CNFs, orchestrated clouds, edge infrastructure, and containerized workloads introduce intricate dependencies. A single degraded component can ripple across multiple services.
The Explosion of Alerts
Thousands of systems generate millions of alarms per day. NOCs drown in noise, making it very difficult to separate symptoms from the true root cause.
Fragmented Tools, Partial Visibility
Operational visibility is typically split across disconnected tools and teams. Rad1X unifies infrastructure, network, and applications through customer-defined service graphs, with native multi-tenancy for department-specific views and responsibilities.
Direct Business Impact
Downtime leads to revenue loss, SLA penalties, churn, and regulatory pressure. Repeated outages erode trust and increase operational cost.
Rad1X
Unified Observability. Clear Root Cause. Faster Resolution.
Built to isolate the one actionable root cause across modern telecom complexity.
AI/ML-Powered Insights
Learns behavior, detects anomalies early, and predicts degradation with GRU and STGNN features explained with an XAI layer.
Cross-Domain Correlation
Consolidates telemetry, correlates across domains, promotes the real root cause, and suppresses irrelevant alarms.
End-to-End Observability
Hardware to cloud-native layers to CNFs—full visibility and service dependency mapping in one interface.
Rad1X gathers telemetry from every layer, correlates it in real time, and automatically pinpoints the true root cause of an alarm. A unified dashboard delivers cross-domain visibility and rapid action.
Modular architecture with a universal data model—new connectors typically onboard in about three months.
Rad1X
How Rad1X Works
Four stages that turn fragmented telemetry into one actionable root cause.
1
Step 1
Telemetry Ingestion via Connector Factory
2
Step 2
Real-Time Correlation Engine
3
Step 3
Service Modeling & Graph-Based Topology
4
Step 4
AI/ML Predictive Engine + Closed-Loop Remediation (TM Forum–Aligned)
Step 1
Telemetry Ingestion via Connector Factory
ZabbixPrometheusGrafana exportersIPMIgNMISNMPKubernetes APIOpenStackAWS / Azure / GCP5G Core, CNF logs, OSS/EMSApache NiFi / FlinkOpenTelemetryLog filesFault systems
Step 2
Real-Time Correlation Engine
- Telemetry is evaluated and associated with services based on defined relations in the service tree/graph and on physical topology, virtual topology, K8s namespaces, CNF components, and business KPIs.
- Correlation is based on a set of catalogs that are fully configurable.
- Stages: alarm creation; alarm root cause analysis with deduplication; alarm federation with external systems; incident creation based on alarm criteria.
- Includes alarm federation using TMF642 Alarm API.
- Rad1X identifies the impact chain and places each alarm within a hierarchical context.
Step 3
Service Modeling & Graph-Based Topology
- Service Inventory built on TMF638 standard, flexible service models aligned to the operator's operational architecture.
- Graph-based topology engine (Neo4j) maps dependencies from hardware and virtual infrastructure to applications and services.
- Instantly reveals which services are impacted by a fault or degradation.
Step 4
AI/ML Predictive Engine + Closed-Loop Remediation (TM Forum–Aligned)
- Machine learning detects anomalies, recurring patterns, and forecasts degradation for proactive intervention.
- Proposed remediation actions are updated in active closed-loop instances that reconfigure services via TMF640, TMF664, and TMF641 APIs to restore service health.
- Remediation actions can be validated against real-time telemetry to confirm resolution and return services to the intended state.
Rad1X
Key Capabilities
What Rad1X delivers out of the box.
Noise Reduction & Alert Suppression
Root Cause Isolation
Multi-Domain Correlation
Predictive Assurance
Service Impact Analysis
Cross-Cloud & Cross-Vendor Visibility
Unified Dashboards & Reporting
Open APIs & TM Forum Compliance (TMF638, TMF642, TMF621, TMF640)
Rad1X
Business Benefits
Fewer alarms, faster fixes, and proactive assurance across every domain.
Noise Reduction
Suppresses redundant alarms, focusing on the true root cause.
Faster Resolution
Identifies what failed and why, cutting MTTR.
Proactive Assurance
AI/ML predicts failures before service impact.
Unified Visibility
Correlates across hybrid deployments with multi-cloud and on-premise servers and applications.
SLA Compliance
Ensures infrastructure and business services stay healthy.
Impact
Reduce MTTR by up to 80%
Based on streamlined RCA and alarm suppression.
Pain Points Solved
Alert fatigueSiloed monitoring toolsSlow troubleshootingReactive postureReduced OPEX
Rad1X
Dashboards & Visual Analytics
Visual clarity that ties infrastructure health to business impact.
service health scoring
domain and rack views
root cause tree visualization
predictive alerts timeline
business service impact mapping
Rad1X
Rad1X In Action
A quick visual tour. Cycle through recent Rad1X screens, dashboards, and visual insights.
Rad1X
Chat Agent Assistance + Knowledge Base
Guided diagnostics, explainable answers, and knowledge-grounded remediation context.
Integrated AI chat agent lets users ask questions about metrics, alarms, and service behavior and receive clear root-cause explanations and supporting details.
Knowledge base allows uploading technical documents so the agent incorporates organization-specific expertise into diagnostic workflows.
Chat Agent
Context-aware RCA assistant
What is causing degraded voice quality in the core?
Voice KPI degradation correlates with packet loss in the CNF UPF namespace. Root cause traces to a misconfigured network policy affecting gNMI traffic.
Recommended action: apply remediation playbook via TMF640 and validate recovery through TMF642 alarm clear events.
Knowledge base reference: Uploaded “5G Core RCA guide.pdf” + operator topology model (Neo4j).
Rad1X
Cloud-Native Architecture Built for Scale
Built for resilience, scale, and closed-loop intelligence.
Architecture Overview
Distributed foundation for resilient assurance
Rad1X is built on a cloud-native, distributed microservices architecture designed for scale, reliability, and real-time intelligence.
Running on Kubernetes, it scales elastically and delivers high availability.
Kafka, TimescaleDB, Redis MQ, and Neo4j graph DB enable fast ingestion, real-time correlation, and end-to-end visibility across multi-cloud and multi-vendor environments.
Modular design ensures continuous evolution toward AI-driven prediction and full closed-loop automation.
Rad1X represents all monitored artefacts as services organized in service trees and graphs.
Services are characterized by health indicators (metrics and events), dependency relationships, and telemetry connectors.
Rad1X is designed for sovereign environments, with all observability, analytics, and AI models operating within customer-owned infrastructure and governance boundaries.
Aligned with TMF Autonomous Networks and powered by open TMF APIs, any asset, function, or domain can be abstracted as a service with defined relationships and connectors, enabling autonomous assurance and closed-loop automation.
Designed for the TM Forum Autonomous Networks roadmap; supports operators moving toward AN Levels 4 and 5—combining real-time RCA, graph-based service impact analysis, intent-aligned service modeling, and autonomous remediation workflows.
Technology Stack
Core Services & Frameworks
Java (JDK 17+)
Spring Boot & Spring Framework (Web, Data, Security)
Hibernate (ORM)
Jackson (JSON processing)
Data & Storage
PostgreSQL (relational store)
TimescaleDB (time-series telemetry storage)
Neo4j (graph database for service topology)
Redis (in-memory cache, fast lookups)
Messaging & Event Processing
Apache Kafka (event bus, pub/sub)
Redis MQ
Apache NiFi (data flow automation)
Apache Flink (real-time log and stream processing)
APIs & Integration
REST (Spring Web)
gRPC (high-performance RPC)
Swagger / OpenAPI
Containerization & Orchestration
Docker (container packaging)
Kubernetes + Helm (deployment, scaling, lifecycle management)
Build & DevOps
Git (source control)
Gradle (build automation)
CI/CD pipelines (Jenkins / GitHub Actions / GitLab CI)
Load balancing, horizontal scaling
Rad1X
The Team Behind Rad1X
Rad1X is built by a multidisciplinary team of experts with decades of combined experience in telecom operations, observability, AI-driven analytics, and large-scale software engineering at Realworld Systems.
Bringing together product leadership, solution architecture, software development, and project management expertise, the team works collaboratively to design a robust Root Cause Analysis platform that turns operational complexity into actionable clarity while maintaining a strong focus on reliability, scalability, and real business impact.
With deep expertise in real-time telemetry correlation, cross-domain observability, and AI-powered service assurance, the team is dedicated to helping organizations move beyond reactive troubleshooting toward predictive and autonomous operations.
By combining telecom domain knowledge with advanced analytics and automation capabilities, Rad1X continuously evolves to deliver faster incident resolution, improved service visibility, and greater operational confidence for modern telecom networks.
Meet the experts
Serban Bogatoniu
Software Developer
Vladimir Coman
Software Developer
Ioana Dragus
Enterprise Architect
Adrian Lici
Solution Architect
Bogdan Nae
Software Developer
Alexandra Racareanu
Project Manager
Alexandra Rizea
Software Developer
Rober Voicu
Software Developer
Cristian Sandu
Software Developer
Rad1X
Use Cases
Purpose-built scenarios that span IT, cloud, and telecom operations.
IT Ops
Identify root cause of outages across servers, VMs, Kubernetes, and networks.
Cloud Ops
Correlate metrics from AWS, Azure, GCP alongside on-prem systems.
Telecom Ops
Link CNF logs to business KPIs (voice, SMS, data health).
DevOps
Reduce alert fatigue, ensure reliable CI/CD deployments.
Cross-domain troubleshooting
Automatically correlates faults across network, cloud, compute, and application domains.
Core network change assurance
Prevents configuration-induced faults
Rad1X
Customer Journey
Clear phases that de-risk adoption and accelerate outcomes.
1
Phase 1
Pre-Sales Workshop
Define goals, use cases, integration scope.
2
Phase 2
Pilot or PoC
Validate Rad1X in real operational scenarios.
3
Phase 3
Deployment & Rollout
Multi-domain onboarding, configuration, dashboards.
4
Phase 4
Continuous Enablement
Fine-tuning, AI model refinement, roadmap alignment.
Rad1X
Why Us
Experience, standards alignment, and relentless delivery.
- 30+ years of telecom expertise
- Proven solutions deployed across global Tier-1 operators
- Active contributor to 6G-IA innovation
- European engineering excellence
- Deep knowledge of OSS/BSS and telecom operations
- Strong roadmap and commitment to continuous evolution
Brand Promise
We do what we promise: fast and efficient, innovative and realistic solutions.
Rad1X
Licensing Calculator
Estimate the total licenses based on number of monitored services and features. Licensing is based on Managed Entities, which are used to weigh monitored services and features in scope.
SSmall
< 1,000 ME
MMedium
1,000 - 10,000 ME
LLarge
10,001 - 20,000 ME
XLExtra Large
> 20,000 ME
Services
0ME
Connectors
0ME
Premium Licenses
0ME
Data Retention
0ME
S
Total0 ME
Services
0
Connectors
0
Premium
0
Data
0
Services
Core product base license. Each monitored object contributes its ME weight to your total.
0 ME
ME Definition: A managed entity quantifies the weight of a monitored object in scope.
Physical server
1 ME / unit
Physical server
K8S cluster
*all dynamic K8S services in the cluster are included
3 ME / unit
K8S cluster
*all dynamic K8S services in the cluster are included
CNF
5 ME / unit
CNF
VNF
5 ME / unit
VNF
Switch / Router
3 ME / unit
Switch / Router
Public Cloud Artefacts / VM
3 ME / unit
Public Cloud Artefacts / VM
Application Server / API, Business Service
5 ME / unit
Application Server / API, Business Service
Connectors Package
Standard connectors are included in the base package. Specialized connectors add to your ME count.
0 ME
Standard Connectors (included)
IPMIGNMIKubernetesOpenstackVMWarePrometheusGrafanaElasticOpenTelemetry
Specialized Connectors
Server connector
20 ME / unit
Server connector
Log File Connector
20 ME / unit
Log File Connector
OSS connector
30 ME / unit
OSS connector
Proprietary connector
30 ME / unit
Proprietary connector
Premium Licenses
Per-domain licenses providing specialized capabilities. 1 Premium License = 100 ME.
0 ME
Premium License: Specialized capabilities per domain for prediction and SLA effectiveness.
AI/ML Prediction per Domain
100 ME / unit
AI/ML Prediction per Domain
Agent Chat
70 ME / unit
Agent Chat
Knowledge Base
70 ME / unit
Knowledge Base
Closed-loop Remediation per Domain
90 ME / unit
Closed-loop Remediation per Domain
Business Catalog
100 ME / unit
Business Catalog
Data Retention
Includes 90 days/service in history DB and 6 days/service in production operational DB.
0 ME
History database90 days / service
Production operational DB6 days / service
Reporting Dashboard & KPIsIncluded
Optional Add-on
Data Warehouse License
40 ME / unit
Data Warehouse License
Total Managed Entities
0ME
Services: 0 MEConnectors: 0 MEPremium: 0 MEData Retention: 0 ME
License Tier
S
Small
< 1,000 ME
Activation Fee
Activation Fee of the Product, per ME size, is added and paid one time.
All values shown in Managed Entities (ME). Contact us for list pricing per ME.
Rad1X
FAQ
Answers grounded in the Rad1X one-pager.
Rad1X correlates telemetry across domains, associates alarms to service context, suppresses duplicates, and promotes the single failing component responsible for the disruption.
Rad1X
Request a Demo
Cut through alert noise, isolate the real root cause, and move from reactive troubleshooting to proactive, automated assurance.
What to expect
- Discovery of your domains, telemetry sources, and TM Forum alignment.
- Guided demo on RCA, predictive assurance, and closed-loop readiness.
- Next steps for PoC or rollout aligned to your operational roadmap.
Rad1X is designed for TM Forum Autonomous Networks roadmap and supports AN Levels 4 and 5 progression.