Watchdog Data Sheet
In this article
Watchdog Overview
Watchdog provides network monitoring for enterprises, carriers and ISPs, edge IoT fleets, and remote workers.
The Watchdog Network Monitoring application uses the EdgeLQ platform to orchestrate distribution and enablement of monitoring agents as native applications on PCs, containers on dedicated hardware and Cisco routers and switches, and VMs.
Watchdog agents ping internet endpoints to report internet quality by targets, target groups, geography, agent, agent groups, and ISPs. Real-time and time-series monitoring data is available by API or in the Watchdog Dashboard for data visualization.
Watchdog System Overview
Deploy Watchdog across the network
- Deploy the agent to existing hardware like Cisco switches and routers as a docker container
- Or install the watchdog agent on dedicated hardware like Dell EMC VEP 1425 or Raspberry Pi and connecting them to the branch ethernet
- No configuration is required. The device will connect automatically onboard to Watchdog in minutes
|
- Run the Watchdog Agent in a VM with a custom image for each installation
- VM configuration is as easy as assigning the activation token and naming the VM on Watchdog Dashboard
- No further configuration is required. The VM will connect automatically onboard to Watchdog in minutes
|
- Windows and macOS native application
- Desktop UI for users to monitor the hardware, WiFi, and ISP conditions that shape their internet quality
- Security and privacy ensured
Agent does NOT allow remote access to the agent software or host device
Agent is NOT a browser-based application that collects surfing behavior - End-user control and transparency
No root access required for installation/upgrades
Pause internet monitoring at any time or pause speed tests to conserve bandwidth when using mobile networks
View logs of all monitoring activity
Submit internet quality ratings to IT
Override location detection to provide better location specificity |
Watchdog Agent Requirements
| CPU: vCPU |
| Headless: 100 MB (no user interface) |
| Desktop Application: 250 MB |
| Protocol | Ports | Destination | Direction | Purpose |
| TCP | 80/443 | *.edgelq.com Configured TCP targets | Outbound | Access to Watchdog controlers (HTTP2 must be supported when an HTTP proxy is configured) TCP probes |
| ICMP | NA | * | Outbound | ICMP Probes to default and custom targets |
| UDP | 53 | DNS Server IP | outbound | DNS Lookups |
| UDP | 3478, 19302 | * | Outbound | Stun server - used to determine public IP Address |
| ICMP | NA | * | Inbound | ICMP Time exceeded messages should not be blocked by firewall for hop trace |
| UDP | * | Targets for which hop trace is enabled | Outbound | Hop tracing on Windows OS (ICMP is used on Linux) |
Watchdog Metrics
| Method | Size/bandwidth |
| ICMP | packet size: 64 bytes per ping |
| HTTP/HTTPS | Packet size: Varies by endpoint |
| Hops | Packet size: 64 bytes per packet
Number of packets varies by number of hops |
| Speed test for broadband and wireless | Bandwidth used: 4 MB upload/4 MB download |
| Speed test for 10 Gbps high-speed networks | Bandwidth used: 4 MB upload/4 MB download |
| Minimum Interval: 1 second |
Recommended defaults - Public targets: 30 seconds
- Local targets in LAN: 1 second
|
| Data Resolution: 1 minute |
Threshold-based quality profiles - Compare real-time and time-series data by latency, jitter, and Loss
- Create multiple customized quality profiles per project
|
Selectable measures - Evaluate real-time and time-series data by the aggregation method
- Median: Midpoint data collected
- Percentile95: 95/100th value average
- Percentile99: 99/100th value average
|
Latency, jitter loss by: - Geography: Country, state/province, city/locality
- Targets and target groups
|
Metrics configuration by API and dashboard interface for: - Targets and target groups
- Target distribution to agents by tag and group
- Agent groups to create aggregated views for teams, hardware types, etc.
|
| CPU |
| Memory |
| Wifi signal strength |
| Network interface metrics |
| 1-minute resolution: One week |
| 1-hour resolution: After one week |
| Data deletion based on contract requirements |
| Data transfer before deletion not currently supported |
Watchdog SaaS Features
| User management | - Role-based access control
- Dashboard user management email invitation, role setting, role promotion/demotion, revoke user access
|
| Agent Management | - Multiple activation modes for Agents
- Shared Token and individual token for Desktop Agent
- Embedded credentials for OS image
- Desktop agent download and token-based onboarding emails to end-users using dashboard forms or Excel upload batch operations.
- Auditing of user operations
|
| Data Visualization | - Aggregate views by map location
- Real-time and time-series data for targets, ISPs, hardware
- Aggregated by geography and groups
- Viewable by an individual agent
|
| Ratings and alerts | - Ratings aggregated by geography and groups
- Downloadable PDF views and link sharing
|
| Additional features | - Data separated by the project in a logical database with access control
- Remote device management via SSH and SCP access for devices running EdgeLQ OS Raspberry Pi, Dell VEP 1425
|
| Language Support | Dashboard and Knowlege Base |
| Documented REST and GRPC API |
The EdgeLQ platform provides the highest-level data security architecture and practices that enterprise SASE and Zero Trust applications and services require - All data is encrypted in transit
- Multi-layer access control to all systems
- Multi-tenant data storage protects all client data
- Data stored and secured on Google Cloud Platform
|
- Level 3 & 4 support to reseller/integrator
- Level 1 & 2 support provided by reseller/integrator
|
