Modern networks are continuously evolving for higher performance and limitless scalability to support constantly changing modern applications workload demands. To ensure that these complex networks are up and running at optimal performance, it has become essential for enterprises and service providers to have end-to-end visibility with real-time network telemetry information to know what is happening on the network at any time.
While using traditional network troubleshooting and monitoring solutions using active and passive network performance measurement approaches have become insufficient to monitor and optimize virtualized on-demand cloud networks, a new hybrid approach that enables packet-level telemetry, by having each node along the path to incorporate timestamps and other potential information in the headers of data plane packets, allowing fine-grained network performance measurement and congestion detection. The hybrid approach uses Push, Poll or In-band(INT) data collector methods to monitor data center network infrastructure more closely and accurately than ever before.
VOLANSYS Network Telemetry Solution
VOLANSYS cloud or on-premise network telemetry solution monitors, detects and optimize network devices to realize utmost network infrastructure performance and availability. The solution traces health of each network device, by using embedded telemetry agent and provides granular-level information including CPU utilization, temperature, geographical location, port statistics, queue statistics, buffer statistics and other table metrics.
The solution integrates with Network Operating Systems (NOS) and analytics engines, to deliver a comprehensive set of telemetry features that monitor network status in real-time with nanoseconds granularity and at scale. A comprehensive view of network infrastructure offered by easily programmable VOLANSYS Network Telemetry solution helps network engineers, operations teams, support desk staff, security analysts, developers and business users—with an accurate, up-to-the-moment understanding of what’s happening in the network.
Using programmable network telemetry solution, network administrators can easily identify and update network device configurations, data collection methods, and customize events responses.
- Automatic device type detection
- Filter-based advanced device search
- Topology formation
- View device port configurations
- Set data collection interval
- Set threshold for each device attribute alerts
- Device simulator
- Data collector methods – Push, Poll or INT
See what’s really going inside the network using granular performance metrics, logs, tasks, events, and topology from each network device. It empowers administrators with an accurate real-time view of network health by proactively identifying performance and capacity bottlenecks to comply with defined SLAs.
- Device status monitoring
- Device history management
- Device Egress queue and buffer statistics
- Queue utilization based on watermark levels
- Packet monitoring and filtration to identify network exceptions
- Packet-based telemetry (INT)
- Latency and distributions metrics (per path, per hop)
- Overall & per path average latency distribution
- In-band switch states report and traffic path analysis
- User-defined path transportation & metadata
Predict downtime with immediate visibility of data center infrastructure health and workload distribution. Identify network devices that are over- or under-used, and network paths that are running out of capacity.
- Egress queue congestion monitoring
- Identify inflanted intra and inter hop latencies
- User-defined path transportation
- Real-time microburst detection with nanoseconds granularity
- Autonomous congestion correction
- Drop counters analysis
- Flows identification assistance (‘Agressor vs Victim’ or ‘Elephant vs Mice’)
Use the secured user interface and powerful analytics engine to spot performance issues and root cause.
- Role-based user management
- Customizable rules and features management
- Programmable data visualization & reports
- Email, SMS and in-app alerts and notifications
- Secure communication with SSL/TLS & REST APIs
- JWT token management
VOLANSYS programmable network telemetry agent can be easily ported on a network device to perform real-time event-driven telemetry data capture and organization of information for consumption by network telemetry application and analytics engines for data visualization.
- Easily portable architecture
- Integration with third-party Cloud Solutions
- Integration with third-party Network Device SDK
- SAL for portability with any other operating system
- Modular design to supports feature insertion or removal at runtime
- Priority queueing for a packet and interrupt-based data
- Optimized CPU utilization
- Configurable data fetching interval per feature
- Support to push feature data only at the change
- Event monitoring and logging support
- Multiple log levels for ease in debugging
- Capability to backup and maintain multiple log files
- Multiple log levels to ease debugging
- Supports backup and maintains multiple log files
- Identify network resource contention and isolate workloads, which can cause resource contention
- Identify microburst and irregularities in network flows that undermine performance
- Identify load changes, latency fluctuations, unwarranted threats and path distribution that negatively impact application performance
- Proactive troubleshooting on single dashboard built-with programmable visualizations and support for integrations with analytics engines
- Real-time state streaming per flow, in ns resolution, to pinpoint device performance issues
- Detect silent packets loss, packet processing exceptions and inconsistent configurations that inflate latencies
- Set real-time alerts and notifications for network administrators to notify device performance degradation or network failures
- Predict capacity saturation and accordingly plan resource efficiency benchmarks
- Test configurations before deployment and continuously monitor system health
- On-premise or cloud ready modular and future-proof open-source architecture
- Eliminate hardware vendor assistance – supports telemetry data extraction for analytics application with no host CPU Software Intervention
- Support for 3rd party analytics engine integrations
- Easy to extend and evolve with programmable P4 Telemetry data