How Does DNS Load Balancing Work and What Are the Common Solutions - 面试题

DNS Load Balancing is a technique that distributes user requests to multiple servers through DNS resolution. It is one of the simplest and most commonly used load balancing solutions, utilizing DNS round-robin mechanisms or other strategies to spread traffic across different server nodes.

How DNS Load Balancing Works

Basic Implementation

dns
; Configure multiple A records for the same domain
www.example.com.    300    IN    A    192.0.2.1
www.example.com.    300    IN    A    192.0.2.2
www.example.com.    300    IN    A    192.0.2.3

When users query www.example.com, the DNS server returns these three IP addresses in sequence, achieving traffic distribution.

Round Robin Mechanism

shell
User 1 queries → Returns 192.0.2.1
User 2 queries → Returns 192.0.2.2
User 3 queries → Returns 192.0.2.3
User 4 queries → Returns 192.0.2.1 (cycles)

Pros and Cons of DNS Load Balancing

Advantages

Advantage	Description
Simple to Use	No additional hardware or software needed, just configure DNS records
Low Cost	Most DNS services support it without extra fees
Global Distribution	Naturally supports cross-regional server distribution
No Single Point of Failure	When one server fails, others continue serving

Disadvantages

Disadvantage	Description
Cannot Detect Server Status	DNS doesn't know if servers are healthy
Caching Causes Imbalance	DNS caching makes same users always access same server
Switching Delay	Failover requires waiting for TTL expiration
Session Persistence Difficult	Same user may access different servers

Common DNS Load Balancing Solutions

1. Simple Round Robin

Configuration Example:

dns
www.example.com.    300    IN    A    192.0.2.1
www.example.com.    300    IN    A    192.0.2.2
www.example.com.    300    IN    A    192.0.2.3

Characteristics:

Simplest load balancing method
Returns IP addresses in cyclic order
Doesn't consider server performance differences

2. Weighted Round Robin

Configuration Example (BIND view configuration):

bind
; Different weights for servers with different performance
view "client1" {
    match-clients { 192.0.2.0/24; };
    zone "example.com" {
        type master;
        file "example.com.client1";
    };
};

Or using DNS services that support weights:

shell
Server A (high performance): Weight 3
Server B (medium performance): Weight 2
Server C (low performance): Weight 1

Characteristics:

Assign different weights based on server performance
High-performance servers handle more requests
Requires intelligent DNS service support

3. Geographic-based Load Balancing (GeoDNS)

How It Works:

shell
Beijing user queries → Returns Beijing server IP
Shanghai user queries → Returns Shanghai server IP
Guangzhou user queries → Returns Guangzhou server IP
US user queries → Returns US server IP

Implementation:

Intelligent DNS services (like Cloudflare, AWS Route 53)
Determine geographic location based on user IP
Return nearest server address

Advantages:

Reduce network latency
Improve user experience
Meet data compliance requirements

4. Health Check-based Load Balancing

Workflow:

shell
1. DNS service regularly checks backend server health status
2. Only return IP addresses of healthy servers
3. Failed servers are automatically removed from resolution results
4. Servers automatically rejoin after recovery

Health Check Methods:

HTTP/HTTPS health checks
TCP port checks
ICMP Ping checks

5. ISP-based Load Balancing

Scenario:

shell
China Telecom users → Return China Telecom line server IP
China Unicom users → Return China Unicom line server IP
China Mobile users → Return China Mobile line server IP

Implementation:

Determine ISP based on DNS query source
Return optimized lines for corresponding ISPs
Reduce cross-network access latency

Comparison of Major DNS Load Balancing Services

Provider	Round Robin	Weighted	Geo Routing	Health Check	Price
Cloudflare	✅	✅	✅	✅	Free/Paid
AWS Route 53	✅	✅	✅	✅	Pay per query
Alibaba Cloud DNS	✅	✅	✅	✅	Free/Paid
Tencent Cloud DNSPod	✅	✅	✅	✅	Free/Paid
NS1	✅	✅	✅	✅	Enterprise paid
BIND + Custom	✅	⚠️	⚠️	❌	Free

Practical Applications of DNS Load Balancing

Scenario 1: High Availability Website Architecture

shell
User Request
    ↓
DNS Round Robin (3 IPs)
    ↓
┌─────────┬─────────┬─────────┐
│  Web 1  │  Web 2  │  Web 3  │
│  Nginx  │  Nginx  │  Nginx  │
└────┬────┴────┬────┴────┬────┘
     └─────────┼─────────┘
               ↓
        Shared Database

Scenario 2: CDN Acceleration

shell
User Request
    ↓
Intelligent DNS (GeoDNS)
    ↓
┌─────────┬─────────┬─────────┐
│ Beijing │ Shanghai│ Guangzhou│
│CDN Edge │CDN Edge │CDN Edge │
└─────────┴─────────┴─────────┘

Scenario 3: Multi-Active Data Centers

shell
User Request
    ↓
DNS Load Balancing + Health Check
    ↓
┌─────────────┐    ┌─────────────┐
│ Beijing DC  │    │ Shanghai DC │
│ Primary     │◄──►│ Secondary   │
└─────────────┘    └─────────────┘

Optimization Strategies for DNS Load Balancing

1. Set TTL Reasonably

dns
; High availability scenario - Short TTL for quick failover
www.example.com.    60    IN    A    192.0.2.1
www.example.com.    60    IN    A    192.0.2.2

; Stable scenario - Long TTL to reduce DNS queries
static.example.com.    3600    IN    A    192.0.2.3

2. Combine with Application Layer Load Balancing

shell
User → DNS Load Balancing (distributes to different data centers)
           ↓
    ┌──────┴──────┐
    ↓             ↓
 DC A           DC B
    ↓             ↓
 LVS/Nginx     LVS/Nginx
    ↓             ↓
 App Cluster   App Cluster

3. Session Persistence Solutions

Solution	Description
Sticky Session	IP-based session persistence, but affected by DNS caching
Session Replication	Synchronize session data between servers
Centralized Session	Use Redis/Memcached to store sessions
JWT Token	Stateless authentication, no session dependency

Common Interview Questions

Q: What's the difference between DNS load balancing and application layer load balancing (like Nginx)?

DNS Load Balancing: Distributes at DNS resolution stage, simple but cannot detect server status
Application Layer Load Balancing: Distributes after request arrives, can health check, more flexible, but requires additional deployment

Q: How to solve the session persistence problem in DNS load balancing?

Use centralized session storage (Redis)
Adopt stateless architecture (JWT)
Combine with application layer load balancing session persistence

Q: What scenarios is DNS load balancing suitable for?

Cross-regional traffic distribution
Simple traffic sharing
Complement to application layer load balancing

Summary

Solution	Complexity	Cost	Suitable Scenarios
Simple Round Robin	Low	Low	Small-scale applications
Weighted Round Robin	Medium	Medium	Heterogeneous servers
Geographic Routing	Medium	Medium	Global applications
Health Check	Medium	Medium-High	High availability requirements
ISP Routing	Medium	Medium	Multi-ISP in China