Saturday, May 16, 2026

CCDE Enterprise Case Study Part 4: MPLS vs VPLS vs SD-WAN Decision Analysis for Independent Store WAN Architecture

CCDE Enterprise Case Study Part 4 – MPLS Contract Renewal and Independent Store WAN Strategy

CCDE Enterprise Case Study Part 4 – MPLS Contract Renewal and Independent Store WAN Strategy

In this section of the enterprise architecture case study, we analyze one of the most realistic enterprise networking challenges:

🎯 Cost Optimization vs Operational Stability

This is one of the most important topics in enterprise architecture.

The challenge is not simply:

  • Choosing a WAN technology
  • Selecting MPLS or SD-WAN
  • Reducing costs

Instead, architects must balance:

  • Business risk
  • Operational simplicity
  • Migration complexity
  • Service continuity
  • Technical debt
  • Scalability
  • Cost optimization

The Scenario

The MPLS provider for independent stores, Annet, has increased pricing by:

$$ 28\% $$

Annet proposes:

  • Migration to VPLS
  • Reuse of existing CE routers
  • Minimal downtime
  • No need for CE refresh

The customer states:

  • Independent stores are non-critical
  • Outages up to six hours are acceptable
  • Minimal change is preferred
  • Alternative WAN options may be explored later

Table of Contents

1. Correct Answer Analysis

✅ Correct Answer:

Plan to pay the uplift to maintain continuity while researching alternative solutions in parallel.

At first glance, this answer may seem disappointing.

Many engineers instinctively want:

  • New technology
  • Migration projects
  • SD-WAN transformation
  • Cost reduction initiatives

However, CCDE-level thinking focuses on:

$$ Risk\\ vs\\ Reward $$

The key architectural insight is:

💡 Sometimes maintaining the status quo is the optimal enterprise decision.

2. Understanding the Business Requirement

The independent stores are fundamentally different from Jacobs stores.

Independent Stores Characteristics

Feature Impact
Own local payment systems Reduced dependency on Jacobs DC
Own telephony systems Reduced WAN criticality
Can tolerate 6-hour outage Lower resilience requirements
Single WAN connection acceptable No HA requirement
Minimal changes desired Migration complexity must stay low

This means:

$$ Business\\ Criticality = Low $$

Therefore:

$$ Aggressive\\ Migration\\ Projects = Difficult\\ to\\ Justify $$

Business Risk Formula

$$ Risk = Impact \times Probability $$

Since:

  • Business impact is low
  • Stores can operate independently

The enterprise should avoid unnecessary migration risk.

3. Why VPLS Is Risky

The provider suggests:

Migrating from MPLS Layer 3 VPN to VPLS

This initially sounds attractive because:

  • No CE refresh needed
  • Lower cost
  • Fast migration
  • Provider automation available

However, the architectural implications are major.

Current MPLS L3VPN Model

Currently:

  • The provider handles routing
  • The provider handles MPLS forwarding
  • The provider isolates sites
  • The provider maintains scalability

In MPLS Layer 3 VPN:

$$ Provider\\ Handles\\ Routing $$

What Changes with VPLS?

VPLS fundamentally changes the WAN model.

Instead of Layer 3 separation:

$$ Layer\\ 2\\ Extension $$

is introduced across the WAN.

💡 Critical Enterprise Insight

VPLS does NOT provide routing.

It only provides Layer 2 transport.

Major Architectural Problems with VPLS

  • Store routing must now be managed internally
  • Static routes become necessary
  • Broadcast domains expand
  • Store isolation becomes harder
  • Policy-based routing becomes more complex

VPLS Complexity Formula

$$ Operational\\ Complexity \propto Sites + Broadcast\\ Domains + Static\\ Routing $$

Why This Becomes Dangerous

Currently:

  • Store-to-store communication is blocked
  • Traffic is centrally controlled
  • PBR enforces policy

VPLS complicates this model significantly.

Without MPLS L3VPN separation:

$$ Traffic\\ Isolation \downarrow $$

and:

$$ Operational\\ Complexity \uparrow $$

Code Example – Static Routing Explosion

ip route 10.0.0.0 255.0.0.0 172.16.1.1

ip route 10.1.0.0 255.255.0.0 172.16.1.1
ip route 10.2.0.0 255.255.0.0 172.16.1.1
ip route 10.3.0.0 255.255.0.0 172.16.1.1
ip route 10.4.0.0 255.255.0.0 172.16.1.1

Now imagine scaling this across:

$$ 102\\ Independent\\ Stores $$

This becomes operationally inefficient very quickly.

Why VPLS Is Commonly Misunderstood

Many engineers think VPLS is simply:

  • "Cheaper MPLS"

But architecturally:

  • MPLS L3VPN = routed WAN
  • VPLS = Layer 2 transport fabric

This changes operational ownership significantly.

4. Why SD-WAN Is Not Optimal Yet

SD-WAN is a modern WAN technology and may eventually become a good solution.

However:

🚫 Timing Matters in Enterprise Architecture

The current enterprise constraints are:

  • Limited time before contract renewal
  • Minimal change requirement
  • Internet bandwidth limitations
  • No existing SD-WAN infrastructure

Critical WAN Capacity Problem

The documentation shows:

  • Jacobs ISP utilization already approaches saturation
  • Peak inbound traffic approaches 0.9Gbps

This means:

$$ Existing\\ Internet\\ Capacity \approx Saturated $$

If SD-WAN is deployed:

$$ Independent\\ Store\\ Traffic + Internet\\ Traffic $$

must traverse the same Internet edge.

Result

$$ Congestion\\ Risk \uparrow $$

Enterprise Migration Challenge

SD-WAN deployment would require:

  • Internet provisioning
  • Security redesign
  • Tunnel architecture
  • New edge devices
  • Controller deployment
  • Training

Migration Complexity Formula

$$ Migration\\ Complexity \propto Sites \times Technologies \times Policies $$

With:

$$ 102\\ Stores $$

the migration becomes a major project.

5. Why Migrating to Bluesky or Taco Is Risky

This option initially appears attractive because:

  • Existing provider relationship exists
  • Operational consolidation may improve
  • Potential long-term simplification exists

However, there are serious issues.

Capacity Problem

The utilization charts show:

  • Bluesky MPLS already reaches near 8Gbps peak
  • Annet MPLS peaks near 4Gbps

Combined:

$$ 8 + 4 = 12Gbps $$

But:

$$ Current\\ DC\\ MPLS\\ Capacity = 10Gbps $$

💡 Key Capacity Insight

The WAN edge does not currently support combined traffic loads.

Additional Challenges

  • New MPLS tail circuits required
  • Store migrations required
  • New VRFs/VPNs needed
  • PBR redesign required
  • Potential delivery delays

This becomes:

$$ A\\ Major\\ WAN\\ Transformation\\ Project $$

which directly violates:

  • Minimal change requirement
  • Limited timeline

6. Why Maintaining the Current MPLS Service Is Best

The best architectural decision is:

Maintain continuity while evaluating alternatives carefully.

This approach provides:

  • Operational stability
  • Business continuity
  • Time for analysis
  • Reduced migration risk
  • Controlled transformation planning

Why This Is Good Enterprise Architecture

Enterprise architects must avoid:

  • Panic-driven migrations
  • Reactive technology changes
  • Poorly understood projects

The correct strategy is:

$$ Stabilize\\ First $$

Then:

$$ Research\\ and\\ Transform\\ Carefully $$

Enterprise Stability Formula

$$ Stability \propto Predictability + Operational\\ Simplicity $$

7. Enterprise WAN Mathematics

WAN Utilization Formula

$$ Utilization = \frac{Traffic}{Capacity} $$

Bandwidth Saturation

$$ If\\ Traffic > Capacity \Rightarrow Congestion $$

Migration Risk Formula

$$ Risk \propto Time\\ Pressure \times Infrastructure\\ Changes $$

Operational Complexity Formula

$$ Complexity \propto Protocols + Policies + Providers $$

Scalability Relationship

$$ Scalability \propto Automation + Standardization $$

8. CLI and Design Examples

Example – MPLS CE Router Static Redistribution

router bgp 64700
 redistribute static

router bgp 64700

 neighbor 10.10.10.1 remote-as 65000

 redistribute static

Example – Policy-Based Routing

route-map BLOCK-STORE permit 10

route-map BLOCK-STORE permit 10

 match ip address 101

 set ip next-hop 192.168.10.1

Example – Basic VRF Segmentation

vrf definition INDEPENDENT-STORES

vrf definition INDEPENDENT-STORES

 rd 65000:200

 address-family ipv4

 route-target export 65000:200
 route-target import 65000:200
Why MPLS L3VPN Scales Better Than VPLS Here

MPLS L3VPN:

  • Separates routing domains
  • Reduces operational burden
  • Prevents Layer 2 scaling issues
  • Simplifies isolation

VPLS:

  • Extends Layer 2 domains
  • Increases operational complexity
  • Requires customer-managed routing

9. Machine Learning and WAN Capacity Planning

Modern enterprises increasingly use:

  • Traffic forecasting
  • Capacity prediction
  • Anomaly detection
  • Predictive analytics

Understanding network utilization trends is a data science problem.

Useful supporting concepts include:

These concepts become critical for:

  • WAN optimization
  • Capacity planning
  • Traffic engineering
  • AIOps platforms

10. Final Architectural Takeaways

🎯 Correct Answer

✅ Plan to pay the uplift to maintain continuity while researching alternative solutions in parallel.

Why This Is Correct

  • Minimizes operational disruption
  • Avoids rushed migration
  • Maintains stability
  • Provides planning time
  • Avoids major infrastructure changes
  • Supports business continuity

Why the Other Answers Are Incorrect

Option Problem
Migrate to Bluesky/Taco Capacity limitations and major migration project
Deploy SD-WAN Internet edge already near saturation
Use LTE/5G Transport only, not full WAN solution
Migrate to VPLS Operational complexity and routing challenges

CCDE Enterprise Case Study Series

Related CCIE Networking Articles

Final Conclusion

This question teaches one of the most important lessons in enterprise architecture:

Not every cost increase justifies a major transformation project.

Architects must evaluate:

  • Risk
  • Business impact
  • Migration complexity
  • Operational overhead
  • Capacity constraints

The technically exciting answer is not always the correct answer.

The optimal enterprise decision here is:

$$ Maintain\\ Stability + Buy\\ Time + Evaluate\\ Carefully $$

That is exactly the mindset expected from a CCDE-level architect.

By at
Labels:

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)

Featured Post

How HMT Watches Lost the Time: A Deep Dive into Disruptive Innovation Blindness in Indian Manufacturing

The Rise and Fall of HMT Watches: A Story of Brand Dominance and Disruptive Innovation Blindness The Rise and Fal...

Popular Posts