1.2 WAN Traffic Analysis
| Item |
Details |
| Document Version |
1.0 |
| Last Updated |
December 2025 |
| Author |
Network Architecture Team |
| Organization |
Abhavtech.com |
| Classification |
Internal Use Only |
1.2.1 Overview
This section provides a comprehensive analysis of Abhavtech.com's WAN traffic patterns, application flows, bandwidth utilization, and performance requirements. Understanding current traffic characteristics is essential for designing appropriate SD-WAN policies, QoS configurations, and capacity planning.
Analysis Objectives
- Characterize application traffic by type and volume
- Identify peak usage patterns and bandwidth requirements
- Document latency and jitter sensitivity for critical applications
- Map traffic flows between sites for topology optimization
- Establish baselines for post-migration comparison
Data Collection Period
| Parameter |
Value |
| Collection Start |
October 1, 2025 |
| Collection End |
November 30, 2025 |
| Duration |
61 days |
| Collection Method |
NetFlow/IPFIX, SNMP, Application Visibility |
| Tools Used |
SolarWinds NPM, Cisco Prime, Splunk |
1.2.2 Aggregate Bandwidth Analysis
Global WAN Bandwidth Utilization
DAILY BANDWIDTH UTILIZATION PATTERN
===================================
Mbps
^
800 │ ████████████
│ ███ ███
700 │ ███ ███
│ ██ ██
600 │ ███ ███
│ ██ ██
500 │ ██ ██
│ ██ ██
400 │ ██ ██
│ ██ ██
300 │ ██ ██
│ ██ ██
200 │██ ████
├────┬────┬────┬────┬────┬────┬────┬────┬────┬────┬────┬────┬────►
0 2 4 6 8 10 12 14 16 18 20 22 24 Hour
Legend: ████ Peak Hours (09:00-18:00 IST) ░░░░ Off-Peak Hours
Site-by-Site Bandwidth Utilization
| Site |
Circuit Capacity |
Avg Utilization |
Peak Utilization |
95th Percentile |
| Mumbai |
800 Mbps |
412 Mbps (52%) |
687 Mbps (86%) |
623 Mbps (78%) |
| Chennai |
500 Mbps |
234 Mbps (47%) |
398 Mbps (80%) |
356 Mbps (71%) |
| Bangalore |
200 Mbps |
156 Mbps (78%) |
189 Mbps (95%) |
178 Mbps (89%) |
| Delhi |
150 Mbps |
98 Mbps (65%) |
134 Mbps (89%) |
121 Mbps (81%) |
| Noida |
150 Mbps |
112 Mbps (75%) |
142 Mbps (95%) |
131 Mbps (87%) |
| London |
300 Mbps |
167 Mbps (56%) |
256 Mbps (85%) |
234 Mbps (78%) |
| Frankfurt |
300 Mbps |
189 Mbps (63%) |
267 Mbps (89%) |
245 Mbps (82%) |
| New Jersey |
500 Mbps |
312 Mbps (62%) |
445 Mbps (89%) |
398 Mbps (80%) |
| Dallas |
300 Mbps |
178 Mbps (59%) |
267 Mbps (89%) |
234 Mbps (78%) |
Bandwidth Utilization Observations
| Finding |
Impact |
Recommendation |
| Bangalore at 95% peak utilization |
Risk of congestion |
Increase capacity to 300 Mbps |
| Noida at 95% peak utilization |
Risk of congestion |
Increase capacity to 300 Mbps |
| Mumbai DC under 60% average |
Headroom available |
Optimize with SD-WAN AAR |
| Inter-region traffic high at 40% |
Global MPLS dependency |
Enable DIA for SaaS |
1.2.3 Application Traffic Classification
Application Distribution by Volume
APPLICATION TRAFFIC DISTRIBUTION
=================================
┌────────────────────────────────────────────────────┐
│ Office 365 & Cloud SaaS ████████████ 28%│
│ SAP ERP/S4HANA █████████ 22% │
│ Video Conferencing (Webex/Teams) ███████ 18% │
│ Voice (SIP/RTP) ████ 8% │
│ Web Browsing ████ 8% │
│ File Transfer (SMB/CIFS) ███ 6% │
│ Database Replication ███ 5% │
│ Backup Traffic ██ 3% │
│ Other █ 2% │
└────────────────────────────────────────────────────┘
Detailed Application Inventory
| Application |
Protocol |
Ports |
Daily Volume |
Peak Bandwidth |
QoS Class |
| Office 365 |
HTTPS |
443 |
892 GB |
245 Mbps |
Business Critical |
| SAP ERP |
RFC/HTTPS |
3300/443 |
734 GB |
198 Mbps |
Business Critical |
| Webex |
UDP/HTTPS |
9000/443 |
612 GB |
312 Mbps |
Real-Time |
| MS Teams |
UDP/HTTPS |
3478-3481/443 |
523 GB |
287 Mbps |
Real-Time |
| Voice (SIP) |
UDP |
5060-5061 |
89 GB |
45 Mbps |
Real-Time |
| Voice (RTP) |
UDP |
16384-32767 |
167 GB |
78 Mbps |
Real-Time |
| Web Browsing |
HTTPS/HTTP |
443/80 |
267 GB |
89 Mbps |
Default |
| File Transfer |
SMB |
445 |
201 GB |
67 Mbps |
Bulk Data |
| Database Rep |
Oracle |
1521 |
156 GB |
123 Mbps |
Business Critical |
| Backup |
Custom |
Various |
98 GB |
234 Mbps (night) |
Bulk Data |
| GitHub/GitLab |
HTTPS |
443 |
78 GB |
45 Mbps |
Business Critical |
| Salesforce |
HTTPS |
443 |
67 GB |
34 Mbps |
Business Critical |
| ServiceNow |
HTTPS |
443 |
45 GB |
23 Mbps |
Business Critical |
Application Criticality Matrix
| Criticality |
Applications |
Bandwidth Share |
Downtime Impact |
| Mission Critical |
SAP, Database Rep, Voice |
35% |
Immediate revenue impact |
| Business Critical |
O365, Webex, Teams, CRM |
50% |
Productivity loss |
| Best Effort |
Web, File Transfer |
12% |
Limited impact |
| Scavenger |
Backup, Updates |
3% |
Can be scheduled |
1.2.4 Traffic Flow Analysis
Inter-Site Traffic Matrix (Average Daily Mbps)
INTER-SITE TRAFFIC MATRIX
═══════════════════════════════════════════════════════════════════
FROM ▼ │ Mumbai │Chennai│ BLR │ Delhi│ Noida│London│ FRA │ NJ │Dallas
─────────┼────────┼───────┼──────┼──────┼──────┼──────┼──────┼──────┼──────
Mumbai │ - │ 156 │ 89 │ 67 │ 78 │ 45 │ 34 │ 56 │ 23
Chennai │ 145 │ - │ 45 │ 34 │ 45 │ 23 │ 12 │ 34 │ 12
BLR │ 78 │ 34 │ - │ 23 │ 34 │ 12 │ 8 │ 15 │ 8
Delhi │ 56 │ 28 │ 18 │ - │ 23 │ 8 │ 6 │ 12 │ 6
Noida │ 67 │ 34 │ 28 │ 18 │ - │ 8 │ 6 │ 12 │ 6
London │ 42 │ 18 │ 10 │ 6 │ 6 │ - │ 34 │ 23 │ 15
Frankfurt│ 32 │ 10 │ 6 │ 5 │ 5 │ 28 │ - │ 18 │ 12
NJ │ 48 │ 28 │ 12 │ 10 │ 10 │ 18 │ 15 │ - │ 56
Dallas │ 20 │ 10 │ 6 │ 5 │ 5 │ 12 │ 10 │ 48 │ -
═══════════════════════════════════════════════════════════════════
Traffic Flow Visualization
MAJOR TRAFFIC FLOWS
==================
┌─────────┐
│ Mumbai │
│ DC │
└────┬────┘
│
┌────────────────────┼────────────────────┐
│ │ │
156 Mbps 89 Mbps 78 Mbps
│ │ │
▼ ▼ ▼
┌─────────┐ ┌─────────┐ ┌─────────┐
│ Chennai │ │ BLR │ │ Noida │
│ DR │ │ │ │ │
└────┬────┘ └─────────┘ └─────────┘
│
45 Mbps ←── DC Replication (Off-peak: 234 Mbps)
│
▼
┌──────────────────────────────────────────────────────────────┐
│ REGIONAL TRAFFIC FLOWS │
│ │
│ Mumbai ◄──── 45 Mbps ────► London ◄──── 34 Mbps ────► │
│ ▲ Frankfurt │
│ Mumbai ◄──── 56 Mbps ────► NJ │ │
│ │ │
│ 18 Mbps │
│ │ │
│ ▼ │
│ Dallas │
└──────────────────────────────────────────────────────────────┘
Traffic Flow Optimization Opportunities
| Flow Pattern |
Current Path |
Optimized Path |
Savings |
| Branch → SaaS |
Branch → DC → Internet |
Branch → Local DIA |
60% latency |
| DC Replication |
MPLS only |
Scheduled MPLS + Internet |
Cost reduction |
| Voice Regional |
MPLS hairpin |
Direct via SD-WAN |
40% latency |
| Video Conference |
MPLS + Internet |
Local DIA with QoS |
Quality improvement |
1.2.5 Latency Analysis
Current Latency Measurements
| Path |
Measured RTT |
Target RTT |
Status |
Primary Use |
| Mumbai ↔ Chennai |
18 ms |
<25 ms |
✅ Good |
DC Replication |
| Mumbai ↔ Bangalore |
22 ms |
<30 ms |
✅ Good |
Application Access |
| Mumbai ↔ Delhi |
28 ms |
<35 ms |
✅ Good |
Application Access |
| Mumbai ↔ Noida |
25 ms |
<35 ms |
✅ Good |
Application Access |
| Mumbai ↔ London |
142 ms |
<150 ms |
✅ Good |
Inter-region |
| Mumbai ↔ Frankfurt |
138 ms |
<150 ms |
✅ Good |
Inter-region |
| Mumbai ↔ New Jersey |
198 ms |
<200 ms |
⚠️ Marginal |
Inter-region |
| Mumbai ↔ Dallas |
212 ms |
<200 ms |
❌ Exceeds |
Inter-region |
| London ↔ Frankfurt |
12 ms |
<20 ms |
✅ Good |
EMEA internal |
| New Jersey ↔ Dallas |
38 ms |
<50 ms |
✅ Good |
Americas internal |
Latency Distribution Analysis
LATENCY DISTRIBUTION BY APPLICATION
===================================
Application 0ms 50ms 100ms 150ms 200ms 250ms
│ │ │ │ │ │
Voice/RTP ├──────┤ Target: <50ms ✅
│░░░░░░│
│ │
Video Conference ├──────────┤ Target: <100ms ✅
│░░░░░░░░░░│
│ │
SAP Interactive ├───────────────┤ Target: <150ms ✅
│░░░░░░░░░░░░░░░│
│ │
Web Browsing ├─────────────────────┤ Target: <200ms ⚠️
│░░░░░░░░░░░░░░░░░░░░░│
│ │
File Transfer ├───────────────────────────┤ Tolerant ✅
│░░░░░░░░░░░░░░░░░░░░░░░░░░░│
│ │
One-Way Delay (OWD) Requirements
| Application Type |
Max OWD |
Acceptable Jitter |
Packet Loss Tolerance |
| Voice (VoIP) |
75 ms |
<30 ms |
<1% |
| Video Conference |
100 ms |
<30 ms |
<1% |
| Real-Time Collaboration |
100 ms |
<50 ms |
<1% |
| Interactive Applications |
150 ms |
<100 ms |
<2% |
| Bulk Data Transfer |
N/A |
N/A |
<0.1% |
1.2.6 Jitter and Packet Loss Analysis
| Path |
Avg Jitter |
Max Jitter |
Packet Loss |
Impact |
| Mumbai ↔ Chennai |
3 ms |
8 ms |
0.02% |
None |
| Mumbai ↔ Bangalore |
5 ms |
12 ms |
0.05% |
None |
| Mumbai ↔ Delhi |
6 ms |
15 ms |
0.08% |
Minimal |
| Mumbai ↔ London |
12 ms |
28 ms |
0.12% |
Video quality |
| Mumbai ↔ New Jersey |
18 ms |
35 ms |
0.18% |
Voice quality |
| Mumbai ↔ Dallas |
22 ms |
45 ms |
0.25% |
Voice/Video |
Quality Degradation Events (Last 60 Days)
| Event Type |
Count |
Avg Duration |
Affected Sites |
Root Cause |
| High Jitter (>30ms) |
23 |
18 min |
Mumbai-NJ |
MPLS congestion |
| Packet Loss (>1%) |
12 |
8 min |
Delhi, Noida |
Last mile |
| Brownout (<50% BW) |
8 |
45 min |
Bangalore |
Circuit issue |
| Blackout |
3 |
2 hr |
Various |
Provider outage |
1.2.7 Cloud and SaaS Traffic Analysis
SaaS Application Traffic
| Application |
Daily Traffic |
Peak Users |
Destination Region |
Current Path |
| Office 365 |
892 GB |
2,850 |
Multi-region |
Via Mumbai DC |
| Salesforce |
67 GB |
450 |
ap3 (Singapore) |
Via Mumbai DC |
| ServiceNow |
45 GB |
380 |
EU (Germany) |
Via Mumbai DC |
| Webex |
523 GB |
1,200 |
Global |
Via Mumbai DC |
| GitHub |
78 GB |
340 |
US (Virginia) |
Via Mumbai DC |
| AWS (IaaS) |
234 GB |
N/A |
ap-south-1 |
Via Mumbai DC |
| Azure (IaaS) |
189 GB |
N/A |
centralindia |
Via Mumbai DC |
Cloud Traffic Optimization Potential
CURRENT VS. OPTIMIZED SaaS PATH
================================
CURRENT (via DC):
┌────────┐ MPLS ┌────────┐ Internet ┌─────────────┐
│ Branch │ ────────────► │Mumbai │ ────────────► │ SaaS Cloud │
│ Delhi │ ← 28ms → │ DC │ ← 45ms → │ (Singapore) │
└────────┘ └────────┘ └─────────────┘
Total: ~73ms RTT
OPTIMIZED (Direct Internet Access):
┌────────┐ Internet ┌─────────────┐
│ Branch │ ────────────► │ SaaS Cloud │
│ Delhi │ ← 35ms → │ (Singapore) │
└────────┘ └─────────────┘
Total: ~35ms RTT (52% improvement)
Cloud OnRamp Candidates
| Application |
Current Latency |
DIA Latency |
Improvement |
Priority |
| Office 365 |
85-120 ms |
35-55 ms |
45-55% |
High |
| Salesforce |
75-95 ms |
40-50 ms |
40-50% |
High |
| Webex |
70-110 ms |
30-45 ms |
50-60% |
High |
| AWS ap-south-1 |
15-25 ms |
10-15 ms |
30-40% |
Medium |
| Azure centralindia |
12-20 ms |
8-12 ms |
30-40% |
Medium |
1.2.8 Time-Based Traffic Patterns
Hourly Traffic Distribution
| Hour (IST) |
Mumbai |
Chennai |
Branches |
EMEA |
Americas |
Total |
| 00:00-02:00 |
125 |
78 |
45 |
89 |
234 |
571 |
| 02:00-04:00 |
98 |
67 |
34 |
67 |
256 |
522 |
| 04:00-06:00 |
89 |
56 |
28 |
56 |
198 |
427 |
| 06:00-08:00 |
156 |
89 |
67 |
78 |
145 |
535 |
| 08:00-10:00 |
345 |
178 |
134 |
156 |
112 |
925 |
| 10:00-12:00 |
456 |
234 |
167 |
189 |
98 |
1144 |
| 12:00-14:00 |
412 |
212 |
156 |
201 |
89 |
1070 |
| 14:00-16:00 |
478 |
245 |
178 |
212 |
145 |
1258 |
| 16:00-18:00 |
498 |
256 |
189 |
178 |
198 |
1319 |
| 18:00-20:00 |
234 |
145 |
98 |
134 |
245 |
856 |
| 20:00-22:00 |
178 |
112 |
67 |
112 |
278 |
747 |
| 22:00-00:00 |
145 |
89 |
56 |
98 |
256 |
644 |
Peak Hour Analysis
| Metric |
Global Peak |
India Peak |
EMEA Peak |
Americas Peak |
| Time |
14:00-16:00 IST |
14:00-16:00 IST |
14:00-16:00 GMT |
10:00-12:00 EST |
| Bandwidth |
1,319 Mbps |
923 Mbps |
212 Mbps |
278 Mbps |
| Overlap |
18:30-21:00 IST |
India + Americas overlap period |
|
|
Weekly Traffic Pattern
| Day |
Avg Traffic |
Peak Traffic |
Notes |
| Monday |
1,245 Mbps |
1,456 Mbps |
Week start, high sync activity |
| Tuesday |
1,189 Mbps |
1,389 Mbps |
Normal operations |
| Wednesday |
1,201 Mbps |
1,412 Mbps |
Normal operations |
| Thursday |
1,234 Mbps |
1,445 Mbps |
Pre-weekend activity |
| Friday |
1,156 Mbps |
1,312 Mbps |
Reduced afternoon |
| Saturday |
345 Mbps |
567 Mbps |
Backup operations |
| Sunday |
289 Mbps |
478 Mbps |
Maintenance window |
1.2.9 Traffic Analysis Summary
Key Findings
| Category |
Finding |
Implication |
| Bandwidth |
78-95% peak at branches |
Capacity upgrade needed |
| Applications |
50% SaaS-bound traffic |
DIA with Cloud OnRamp |
| Latency |
Inter-region exceeds targets |
Regional optimization |
| Traffic Patterns |
40% DC-bound from branches |
Direct breakout opportunity |
| Peak Hours |
14:00-16:00 IST global peak |
QoS critical during peak |
SD-WAN Design Recommendations
| Recommendation |
Priority |
Expected Benefit |
| Implement Cloud OnRamp for SaaS |
High |
45-55% latency reduction |
| Enable DIA at all branches |
High |
Reduced DC load, better SaaS |
| Increase branch bandwidth 50% |
High |
Eliminate congestion |
| Implement AAR for voice/video |
High |
Quality assurance |
| Configure FEC for long-haul paths |
Medium |
Packet loss mitigation |
| Schedule bulk transfers off-peak |
Medium |
Peak hour relief |
Baseline Metrics for Post-Migration
| Metric |
Current Baseline |
Target Post-Migration |
| Avg WAN Latency (Regional) |
25 ms |
<20 ms |
| Avg WAN Latency (Inter-region) |
175 ms |
<160 ms |
| Avg Jitter |
15 ms |
<10 ms |
| Packet Loss |
0.15% |
<0.05% |
| SaaS Application Latency |
85 ms |
<45 ms |
| Branch Utilization (Peak) |
90% |
<70% |
| Voice MOS Score |
3.8 |
>4.2 |
References
| Document |
Description |
| NetFlow Analysis Report |
Raw traffic data |
| SolarWinds NPM Reports |
Bandwidth utilization |
| Cisco Prime Application Visibility |
Deep packet inspection data |
| Splunk Traffic Logs |
Historical traffic patterns |
| SD-WAN Design Guide |
Cisco CVD reference |
Document Version: 1.0
Last Updated: December 2025
Classification: Internal Use Only
Abhavtech.com - SD-WAN Documentation