ATLAS INTELLIGENT MONITORING
Intelligent Monitoring is being developed to help Layer 3 understand network and service quality — not just whether a device responds. Latency, packet loss, availability, recent state and customer impact all provide important context.
A USEFUL PICTURE NEEDS MORE THAN ONE SIGNAL
Is the service reachable?
What does available evidence show?
Has behaviour changed?
What else may explain the symptom?
VISIBILITY BEFORE GUESSWORK
Traditional monitoring may answer whether something is online. Useful monitoring should build a broader picture where sufficient data exists.
Is the supported monitored service reachable?
Has monitored behaviour changed?
Do supported quality indicators look unusual?
Does a wider site or one component appear affected?
Monitoring does not automatically prove the cause of every problem. It provides evidence that can make diagnosis more informed.
THE SIGNALS THAT HELP BUILD THE PICTURE
Information depends on the monitored service, supported equipment and configuration. Not every category is available on every installation.
Whether supported monitored services or devices are reachable.
Where supported, latency and packet loss add context.
Status from supported monitored infrastructure where available.
Whether a supported connection changes state.
Historical patterns where suitable data is recorded.
Monitored changes around when a problem began.
ONLINE DOES NOT ALWAYS MEAN HEALTHY
A service may remain reachable while users experience packet loss, latency, congestion, intermittent connectivity, poor Wi-Fi, a device-specific fault or an application problem.
Online/offline is useful information, but it is only one part of diagnosing customer experience.
WHEN THE PROBLEM DISAPPEARS BEFORE SUPPORT LOOKS
“It was broken ten minutes ago, but now it's working.”
When did behaviour change?
How long did it last?
Was an entire monitored site affected?
Has something similar happened before?
Historical evidence may help only where appropriate monitoring and history exist.
MORE THAN RED OR GREEN
The value is using available signals to understand what requires attention and what the evidence does — or does not — support.
01
02
03
04
05
No autonomous AI reasoning or automatic remediation is claimed.
START WITH THE SYMPTOM
Customer symptoms are starting points, not diagnoses.
Evidence may establish when supported connectivity changed.
Supported latency, packet-loss or utilisation data may narrow possibilities.
History may help with intermittent faults.
Evidence may show whether the wider monitored service is affected.
Online/offline alone is insufficient.
Recorded events may add context.
MONITORING FINDS CLUES
If a collector stops ingesting data, the last known value may still look normal. Intelligent Monitoring should distinguish healthy current data from stale, missing or failed data.
Observes supported signals.
Interprets evidence in context.
Chooses the right next step.
No new data is itself an operational signal. Freshness, heartbeat and last successful ingest must be considered before trusting a status.
TWO DIFFERENT TYPES OF VISIBILITY
Atlas Intelligent Monitoring focuses on supported connectivity, service conditions, network infrastructure and performance evidence.
Network and service visibility.
Supported computers, device health, maintenance, updates, protection and backup context.
Explore Device MonitoringVISIBILITY IS NOT MAGIC
Monitoring does not automatically repair hardware, increase capacity, improve weak Wi-Fi, fix Windows, remove malware, repair applications or resolve every upstream fault.
It helps create the evidence needed to determine the appropriate next step.
DON'T WAIT FOR A PHONE CALL TO CREATE THE FIRST CLUE
When appropriately configured, monitoring data can provide information before, during or after a reported problem.
Monitoring evidence and alert delivery are related, but they are not the same promise.
Explore Proactive AlertsPART OF THE ATLAS DIRECTION
A degraded device or link may affect one customer, one sector, one site, multiple downstream services or many customers. Atlas Intelligent Monitoring is being developed to connect technical evidence to operational impact.
Atlas is Layer 3's network-management and operational-intelligence platform. Customer-impact correlation remains a developing capability and no real customer data is shown here.
THE MORE IMPORTANT THE CONNECTION, THE MORE USEFUL THE EVIDENCE
Atlas must monitor Atlas. Dependency health should consider heartbeat, last successful ingest, stale-data detection, error state, API availability and worker or collector health. Zabbix, syslog, IDS feeds, UISP, RADIUS, DNS services, Redis, Postgres, collectors and APIs are architectural examples, not claims that every integration is complete.
Recurring supported connectivity problems.
Calls and productivity depend on connection quality.
Multiple users and critical connectivity.
Several supported locations or services.
Multiple links and remote infrastructure complicate troubleshooting.
Explore Farms & LodgesNOT EVERY CONNECTION NEEDS THE SAME LEVEL OF MONITORING
A home connection may not require the same approach as a business, lodge, farm, remote site, multi-link connection or multi-site organisation.
Monitoring should solve a visibility or operational problem — not be added simply because the technology exists.
QUESTIONS PEOPLE ACTUALLY ASK
Capabilities remain dependent on supported monitored services and configuration.
Only for supported monitored-service availability where configured.
History may help where suitable data exists.
Evidence may narrow the problem; wireless diagnosis still needs appropriate evidence.
No.
Not necessarily.
No. This is network-oriented; Device Monitoring is endpoint-oriented.
“Intelligent” means using information with context; autonomous AI is not claimed.
No such claim is made.
That depends on Proactive Alerts and configuration.
MONITOR THE SERVICE — NOT THE PERSON
The intended focus is technical health, availability, performance evidence, supported infrastructure and operational conditions.
No claim is made to monitor personal documents, message content, private conversations or employee productivity.
MONITORING FOR REAL SOUTH AFRICAN NETWORK CONDITIONS
Fibre, Fixed Wireless, LTE/5G, remote connectivity and multi-link sites expose different symptoms and constraints.
Nationwide availability is not claimed without verification for the specific service and site.
INTELLIGENT MONITORING
A network-oriented Atlas capability for supported connectivity evidence.
Availability, quality, infrastructure, state, trends and events may be available depending on configuration.
No.
No.
Recorded history may help where suitable data exists.
No automatic remediation is promised.
It is positioned around technical conditions, not personal-content surveillance.
Yes.
Potentially, where supported and useful.
Start with the visibility problem, frequency and impact.
START WITH WHAT YOU NEED TO KNOW
Intelligent Monitoring is being developed to help Layer 3 understand service quality, detect meaningful changes and give technical teams better evidence before deciding what to do next.