Global server load balancing (GSLB) in HLD - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

How does GSLB improve user experience globally?

Imagine you have users all over the world accessing your website. How does Global Server Load Balancing (GSLB) help improve their experience?

A By sending all user requests to a single central server to maintain consistency.

B By directing users to the closest or fastest server to reduce latency and improve response time.

C By randomly distributing user requests to any server regardless of location.

D By caching all content on the user's device to avoid server requests.

💡 Hint

Think about how distance affects loading speed.

GSLB directs users to the nearest or best-performing server, reducing delay and improving speed.

❓ Architecture

intermediate

Which component is essential in a GSLB architecture for health checks?

In a GSLB system, which component regularly checks if servers are healthy and available?

A Global DNS server that performs health checks and updates routing accordingly.

B User's browser that pings servers before sending requests.

C Load balancer inside each data center without global coordination.

D Content Delivery Network (CDN) that caches static content only.

💡 Hint

Think about who decides where to send user requests globally.

The global DNS server monitors server health to avoid sending users to down servers.

❓ scaling

advanced

How to scale GSLB to handle millions of global users?

You want your GSLB system to support millions of users worldwide. Which approach best supports this scale?

A Rely on client-side load balancing to distribute traffic evenly.

B Use a single global DNS server with a large database to handle all requests.

C Deploy multiple global DNS servers with geo-redundancy and use Anycast IP for fast routing.

D Use only local load balancers in each data center without global coordination.

💡 Hint

Consider how to reduce latency and avoid single points of failure.

Multiple global DNS servers with Anycast provide fast, reliable routing and fault tolerance.

❓ tradeoff

advanced

Tradeoff between DNS TTL and failover speed in GSLB

In GSLB, what is the tradeoff when setting a very low DNS TTL (time-to-live) value?

A Low TTL eliminates the need for health checks on servers.

B Low TTL reduces DNS query load but causes slow failover during outages.

C Low TTL improves caching efficiency but risks stale routing information.

D Low TTL allows fast failover but increases DNS query load and latency for users.

💡 Hint

Think about how often DNS records are refreshed and its impact.

Lower TTL means DNS records update quickly for failover but cause more frequent DNS lookups, adding load and latency.

❓ estimation

expert

Estimate DNS query capacity for a global GSLB system

Your global website has 100 million daily active users. Each user makes 10 DNS queries per day on average. You deploy 5 global DNS servers. Estimate the average DNS queries per second each server must handle.

A Approximately 2300 queries per second per server.

B Approximately 2000 queries per second per server.

C Approximately 23000 queries per second per server.

D Approximately 20000 queries per second per server.

💡 Hint

Calculate total queries per day, then per second, then divide by servers.

Total queries = 100M users * 10 = 1 billion queries/day. Seconds per day = 86400. Queries per second total = ~11574. Divide by 5 servers = ~2315 queries/sec per server.