Chapter 1: Basics of System Design

Question 1

How to give a SD interview

Answer 1

1. Comprehend scope of the question.
2. Never assume things that are not explicitly stated
3. Gather all the requirements
4. You are required to scale breadth and depth of a blurred problem.
5. The solution is not the goal of the interview, it is the conversation, which should cover breadth as well as depth.
6. Break problem into modules and tackle modules individually.
7. There are no solutions to many system design problems, its about making progress.
8. Summary
   a. Scoping the problem.
   b. Sketching up and abstract design
   c. Identify and address bottlenecks.

Question 2

System Design basics

Answer 2

1. What are different architectural pieces that can be used
2. How do these pieces work with each other?
3. How can we best utilize these pieces and what are the tradeoffs?

Question 3

Load balancing

Answer 3

Can be added at three layers

1. Between user and webserver
2. Between webserver and appserver
3. Between appserver and database.
4. Smart client
5. Hardware load balancer.
6. Software load balancer

Question 4

Caching

Answer 4

1. Appserver cache
2. Distributed cache (consistent hashing)
3. Global cache
4. Content distribution network
5. Cache invalidation
   a. Write through
   b. Write around
   c. Write back
6. Eviction policies
   a. FIFO
   b. LIFO
   c. LRU
   d. LFU
   e. Random

Question 5

Sharding or Data Partitioning

Answer 5

Methods:
1. Horizontal - different rows on different machines. Also called range based partitioning. Unbalanced - side effect.
2. Vertical partitioning
3. Directory based partitioning - There will be a directory server which will handle redirecting queries to the server.
Criteria:
1. Key or hash based partitioning.
2. List partitioning - based on group like region
3. Round robin.
4. Composite of above

Common problems of sharding

1. Joins and Denormalization
2. Referential integrity.
3. Rebalancing

Question 6

Indexes

Answer 6

Nothing much

Question 7

Proxy

Answer 7

Proxy sits between user and app server and can cache data.

Question 8

Queues

Answer 8

For large scale processing

Question 9

Redundancy and Replication

Answer 9

Shared nothing architecture

Question 10

SQL vs NoSQL

Answer 10

SQL:
Relational databases, mysql, oracle etc.
NoSQL:
Key - Value stores, document databases, Wide-Column databases, graph databases

Question 11

CAP Theorem

Answer 11

Consistency, Availability and partition all three cannot be achieved.

Question 12

Consistent Hashing

Answer 12

Having n hash servers and dividing data with hash % n will not scale. Some servers become hot and others cold depending on data.

ConsistentHashing:
/Users/manish/Downloads/687474703a2f2f626c6f672e636f64696e676c6162732e6f72672f75706c6f6164732f70696374757265732f636f6e73697374656e742d68617368696e672f362e706e67.png

Question 13

Long-Polling Vs WebSockets vs Server sent events

Answer 13

1. Ajax
2. Long polling:
   Server may not immediately respond.
   Client may immediately re request, so that server will be able to push as soon as it has data. This is sometimes called hanging get.

Question 14

Websockets

Answer 14

Provides full duplex communication over TCP.

Question 15

Server side events

Answer 15

Only server to client communication.

Question 16

A step by step guide to system design interview

Answer 16

1. Requirements clarification.
2. Interface definition
3. Back of envolope estimation
4. Defining data model
5. High level design
6. Detailed design