Video summary
[Hindi] Day-1: Why we need SDWAN
Main summary
Key takeaways
Summary (Technological Concepts + SD-WAN / Legacy WAN Analysis)
Legacy WAN vs SD-WAN (HD WAN)
- The tutorial explains that legacy WAN devices are being retired because they are:
- Mostly IP-based
- Manually driven
- In legacy environments, routing/switching decisions depend largely on a routing table.
- Legacy devices typically cannot understand application intent—they “see” IP rather than the traffic’s meaning.
Old Traffic-Processing Model (Limited Intelligence)
- When a user opens google.com:
- The PC resolves URL → IP using DNS
- WAN/edge devices forward traffic mainly using IP destination matching from their routing tables
- If the destination IP is not found, the traffic is dropped.
- As a result, forwarding decisions are less traffic-aware, reducing control and visibility.
What SD-WAN Adds (Application/Traffic Awareness + Automation)
- SD-WAN shifts to a model where the edge device infers more than just IP, including:
- Source/destination
- Routing context
- Which application is being accessed
- This is important because in real networks, link quality can drop (e.g., packet loss and reliability issues), causing failures at branches (examples include):
- Banking/payment systems
- Billing
- Ticketing
- With SD-WAN, actions such as dropping, load balancing, and reporting can be driven by:
- Traffic/application behavior
- Link conditions
- This reduces reliance on manual engineer interventions.
Control-Plane Structure: “Box Inside a Box” and Input-Output Module
- The explanation uses an input-output module concept:
- After login/configuration, the device uses internal modules (described as “age models” / modules)
- These modules represent the input-output handling of traffic
- Packet handling during forwarding is governed by internal state:
- Routing
- Policies
- Routing protocols may still run inside the device, but decisions remain tied to the device’s internal routing/policy tables.
Problem with Manual Failover in Legacy Environments
- When a primary WAN circuit degrades, legacy recovery typically involves:
- Inspecting and identifying the impacted branch/service
- Logging in and checking reliability metrics (CRC-like/reliability indicators are mentioned)
- Manually failing over
- The tutorial emphasizes this process is:
- Slow
- Dependent on expertise
- Sometimes not fully successful (even after failover, service impact may remain)
- SD-WAN is positioned as improving this by enabling:
- Automatic detection
- Automated reroute using thresholds/conditions (packet loss thresholds like 10% / 5% are referenced)
Operational Efficiency: Less Third-Party Tooling
- Legacy monitoring/automation often requires third-party tools, for example:
- SolarWinds (mentioned for config storage)
- Or manual backups to shared storage
- SD-WAN is presented as offering integrated management/visibility via:
- vManage (described as providing “everything”)
- This reduces dependence on additional third-party solutions.
Simplified Replacement and Zero-Touch Style Recovery
- In legacy networks, replacing a failed router may require:
- Replacement
- Reconfiguration/migration effort
- In the SD-WAN scenario, replacement is described as easier:
- Remove the faulty unit
- Power on the replacement
- The replacement retrieves policies and routing tables automatically
- This is framed as making the operational lifecycle simpler.
Migration / Compatibility: Converting Legacy Routers to SD-WAN
- The tutorial suggests a customer can migrate a legacy router by:
- Changing the image/software
- Updating licenses
- After this, the device becomes SD-WAN-capable (the video references “iOS”/image and license conversion).
- Deployment model details mentioned:
- vManage is virtual (not a physical box)
- SD-WAN site devices are described as physical boxes with WAN/land/console ports (and auxiliary ports)
- These devices have increased features due to SD-WAN capability
Key Themes Across the Tutorial
- Why SD-WAN is needed:
- Automation
- Application-aware decisions
- Better reliability management
- How SD-WAN differs:
- Goes beyond IP-only routing
- Adds more intelligence for traffic and policy-based actions
- Operational benefits:
- Less manual work
- Less third-party tooling
- Simpler failover and replacement
- Centralized management with vManage
Main Speakers / Sources
- Gaurav Sir (repeatedly referenced; provides demos and explanations)
- Pankaj (asks questions)
- Varsha and Steven (mentioned as examples of vendors/participants)
- Other vendor names referenced in subtitles (e.g., “Systo” / “Taylor”)
- No external source links explicitly shown, except a mention of a YouTube link for a prior class/demo.