UNLEASHING YOUR NETWORK’S FULL POTENTIAL

Topology • Capacity • Automation

UNLEASHING YOUR NETWORK’S FULL POTENTIAL

Topology • Capacity • Automation

UNLEASHING YOUR NETWORK’S FULL POTENTIAL

Topology • Capacity • Automation

Advanced Network Data-Mining

Nomad Automate is a leading network data-mining expert. Since 2005 it has developed as a reference partner for advanced end-to-end network discovery, capacity management and anomaly detection in multi-vendor transport networks (wireless and optical transmission, IP/MPLS, mobile backhaul, fixed access).

We make complex things easier for our clients. We design smart sampling and discovery algorithms offering unmatched clarity in the network. Leveraging on the latest technologies in machine learning we deliver state-of-the-art multi-layer network modelling and impactful network performance predictions to address our customers’ biggest pain-points: End-to-end capacity utilization monitoring & forecast ♦ Scenario planning ♦ Accurate service route finding (i.e. RAN site to controller) ♦ Cross-layer correlation (i.e. IP/MPLS layer down to physical layer) for service routes, alarms, performance metrics.

Our Customers

End-to-End Network Discovery

Multi-layer, multi-vendor topology discovery
A reliable network topology representation is the fundamental prerequisite for efficient root-cause analysis, capacity planning or automation. Our discovery algorithms deliver a multi-layer modelling of the network, regardless of vendors or technology. Simply navigate top-down through the model to visualize the real underlying route of a high-end service (i.e. RAN 4G site) across the transport network. Browse the model bottom-up to identify the upper-layer services (i.e. 802.1Q VLAN, Gigabit Ethernet, IP/MPLS Tunnels, etc) potentially impacted by a fiber-cut or a link outage.
ML-powered topology stitching
In order to be usable the discovered network topology must be continuous across the various domains and vendors installed-bases. We call the act of connecting adjacent topology subtrees: Topology Stitching.

Leveraging on machine-learning we have developed advanced topology stitching algorithms to perform topology stitching automatically and at large scale. As a result any service routes can tracked end-to-end.
Hardware & license inventory
A complete inventory of the hardware components (chassis, modules, SFPs, ODUs, etc) and licenses is performed. Network planners can visualize chassis, slots and ports provisioning states while asset managers can track the multi-vendor installed-base over time.

End-to-End Network Discovery

Multi-layer, multi-vendor topology discovery
A reliable network topology representation is the fundamental prerequisite for efficient root-cause analysis, capacity planning or automation. Our discovery algorithms deliver a multi-layer modelling of the network, regardless of vendors or technology. Simply navigate top-down through the model to visualize the real underlying route of a high-end service (i.e. RAN 4G site) across the transport network. Browse the model bottom-up to identify the upper-layer services (i.e. 802.1Q VLAN, Gigabit Ethernet, IP/MPLS Tunnels, etc) potentially impacted by a fiber-cut or a link outage.
ML-powered topology stitching
In order to be usable the discovered network topology must be continuous across the various domains and vendors installed-bases. We call the act of connecting adjacent topology subtrees: Topology Stitching.

Leveraging on machine-learning we have developed advanced topology stitching algorithms to perform topology stitching automatically and at large scale. As a result any service routes can tracked end-to-end.
Hardware & license inventory
A complete inventory of the hardware components (chassis, modules, SFPs, ODUs, etc) and licenses is performed. Network planners can visualize chassis, slots and ports provisioning states while asset managers can track the multi-vendor installed-base over time.

Capacity Management & Planning

End-to-end Capacity Utilization
Capacity usage monitoring and congestion detection is a critical aspect of network operations and planning. We deliver reliable near real-time capacity utilization monitoring for the entire network (OTN/WDM, microwave, IP/MPLS). Traffic-related metrics such as transmit rates (Mbps), capacity utilization (%) or discarded frame ratio (%) are monitored over time for all multi-vendor devices and services across the network.

Capacity-related metrics are automatically correlated with the multi-layer network model. Network engineers can therefore visualize immediately on the network map the congested connections (at any layer) or locate in the whole network the underlying contributor(s) of packet drops on a higher layer service.
Traffic predictions

Measurements are stored for a long period of time (at least one year) to offer precise trend analysis using aggregation or comparative reports. Leveraging on these accurate measurements we compute statistical regressions to provide network planners with robust traffic predictions on every links in the network, at any layer. These predictions correlated with the end-to-end topology help enhance the capacity planning process dramatically.
Capacity planning
Planning capacity upgrades in the transport network is a complex activity. While it can be a time-consuming and lengthy process, inaccurate planning may results in overspending or drops of the quality of service. Combining our state-of-the-art network discovery capability with the powerful capacity usage monitoring & predictions, we offer network planners the possibility to simulate capacity upgrades scenarios directly from the network map using real and predicted utilization figures.

An precise and agile capacity planning capability is critical for network operators, we deliver exactly this: responsive yet accurate capacity planning.

Capacity Management & Planning

End-to-end Capacity Utilization
Capacity usage monitoring and congestion detection is a critical aspect of network operations and planning. We deliver reliable near real-time capacity utilization monitoring for the entire network (OTN/WDM, microwave, IP/MPLS). Traffic-related metrics such as transmit rates (Mbps), capacity utilization (%) or discarded frame ratio (%) are monitored over time for all multi-vendor devices and services across the network.

Capacity-related metrics are automatically correlated with the multi-layer network model. Network engineers can therefore visualize immediately on the network map the congested connections (at any layer) or locate in the whole network the underlying contributor(s) of packet drops on a higher layer service.
Traffic predictions

Measurements are stored for a long period of time (at least one year) to offer precise trend analysis using aggregation or comparative reports. Leveraging on these accurate measurements we compute statistical regressions to provide network planners with robust traffic predictions on every links in the network, at any layer. These predictions correlated with the end-to-end topology help enhance the capacity planning process dramatically.
Capacity planning
Planning capacity upgrades in the transport network is a complex activity. While it can be a time-consuming and lengthy process, inaccurate planning may results in overspending or drops of the quality of service. Combining our state-of-the-art network discovery capability with the powerful capacity usage monitoring & predictions, we offer network planners the possibility to simulate capacity upgrades scenarios directly from the network map using real and predicted utilization figures.

An precise and agile capacity planning capability is critical for network operators, we deliver exactly this: responsive yet accurate capacity planning.

Anomaly Detection & Performance Monitoring

Fault & performance monitoring
In order to provide efficient root-cause analysis, a wide range of performance metrics and alarms are monitored. Alarms, error counters (ES, SES, UAS), signal levels (TSL/RSL), optical power levels, modulation statistics are constantly monitored. Comparative reports and statistical analysis help identify the slightest degradation over long period of times.
Correlation & root-cause analysis
In order to perform efficient root-cause analysis, alarms and performance metrics are correlated in a bi-dimensional way: vertically through the different network layers, horizontally across the network (end-to-end). As a result, an alarm on a given service is automatically associated to the corresponding alarms on links of lower network layers. Similarly, performance metrics of a given service are also correlated with traffic and performance metrics of all underlying services, end-to-end. The unique contributors of any degradation in the network can therefore be quickly and accurately identified across the network, at any layer.
Reconciliation planning vs. reality
Mismatches between network planning and real network configuration are common and may cause avoidable troubles in the network. An efficient way to avoid such issues or to speed up their resolution is to first ensure that the network is actually configured according to planning.

Network planning data, particularly microwave link engineering data (i.e. link budget) can be loaded in the app. ML-powered algorithms automatically reconcile planning data with the real network topology, configuration and performance. Discrepancies between planning and reality are automatically reported (frequencies, power levels. modulation, etc). Network planners can work on accurate datasets while anomalies resulting from planning misconfiguration are immediately detected and resolved.

Anomaly Detection & Performance Monitoring

Fault & performance monitoring
In order to provide efficient root-cause analysis, a wide range of performance metrics and alarms are monitored. Alarms, error counters (ES, SES, UAS), signal levels (TSL/RSL), optical power levels, modulation statistics are constantly monitored. Comparative reports and statistical analysis help identify the slightest degradation over long period of times.
Correlation & root-cause analysis
In order to perform efficient root-cause analysis, alarms and performance metrics are correlated in a bi-dimensional way: vertically through the different network layers, horizontally across the network (end-to-end). As a result, an alarm on a given service is automatically associated to the corresponding alarms on links of lower network layers. Similarly, performance metrics of a given service are also correlated with traffic and performance metrics of all underlying services, end-to-end. The unique contributors of any degradation in the network can therefore be quickly and accurately identified across the network, at any layer.
Reconciliation planning vs. reality
Mismatches between network planning and real network configuration are common and may cause avoidable troubles in the network. An efficient way to avoid such issues or to speed up their resolution is to first ensure that the network is actually configured according to planning.

Network planning data, particularly microwave link engineering data (i.e. link budget) can be loaded in the app. ML-powered algorithms automatically reconcile planning data with the real network topology, configuration and performance. Discrepancies between planning and reality are automatically reported (frequencies, power levels. modulation, etc). Network planners can work on accurate datasets while anomalies resulting from planning misconfiguration are immediately detected and resolved.

Get Started with Your Personalized Demo

Hahngasse 17/1
1090 Vienna
AUSTRIA
office@nomadautomate.com