Reliability Concepts
Scaling Techniques
Fault Tolerance
Maintainability
Performance Metrics
100

What is the difference between fault and failure?

Faults are deviations from specifications, while failures affect system functionality

100

What is vertical scaling?

Adding more resources to a single machine

100

What is fault tolerance?

A system's ability to continue functioning despite faults

100

What is simplicity in maintainability?

Managing system complexity

100

What is throughput?

Number of processed data points per unit time

200

How can hardware faults be tolerated?

Through rolling upgrades and fault-tolerant architectures

200

What is horizontal scaling?

Adding more machines to the system

200

Name a type of fault that replication can handle.

Hardware faults

200

Define evolvability.

Making a system easy to change over time

200

What is latency?

Time taken to respond to a request

300

What is eventual consistency?

Ensures all replicas will eventually converge to the same state

300

Why is partitioning important for scalability?

Distributes data to prevent bottlenecks and improves performance

300

What is the purpose of leader election in replication?

To appoint a new leader after the current one fails

300

What does operability focus on?

Simplifying operational tasks for engineers

300

Why is the 99th percentile important in performance?

It represents the worst-case scenarios most users experience

400

What is a rolling upgrade?

Updating a system incrementally without downtime

400

How does dynamic rebalancing help scalability?

Adjusts data distribution as load increases

400

How does replication improve reliability?

By keeping multiple copies of data across nodes

400

Why does maintenance dominate software costs?

Long-term upkeep often exceeds development costs

400

What is hybrid scaling?

Combining vertical and horizontal scaling

500

How do distributed systems handle replication lag?

Through strategies like direct reads from leaders or using indexes

500

What is a hot spot, and how is it mitigated?

Overloaded nodes, mitigated by hashing and replication

500

What is the difference between synchronous and asynchronous replication?

Synchronous waits for acknowledgment; asynchronous does not

500

How does good maintainability improve developer productivity?

Simplifies adding features and resolving issues

500

How does load balancing improve performance?

Distributes requests evenly across servers