Interview Notes

Components to know about

• Domain Name System (DNS)
• Proxies and Reverse Proxies
• Load Balancers
• Relational and NoSQL Databases
• Object Storage
• Content Delivery Network (CDN)
• Servers
• Cache

Concepts to know

• CAP
• PACELC
• Distributed Systems Requirements
• Testing
• Consistent Hashing

Data Structures to Know

• Trie - Search Autocomplete
• Quadtree - Location Based Indexing
• R-Tree - Location Based Indexing, sercing multi-dimension shapes, asuch as nearest neighbor lookups
• Geohash - Location Based Indexing
• Skiplist - in memory index type
• Hash Index - in-memory index type - common implementation of the "Map" datastructure
• SSTable - disk based "Map" datastructure
• LSM Tree - skiplist (memory) + sstable (disk) combination to provide high write throughput
• B-Tree - disk based index with consistent read/write performance - most popular index used in databases
• Inverted Index - used for document indexing in search
• Suffix Tree - used for string pattern search, such as string suffix match

Algorithms to Know

• Consistent Hashing - Balancing load within a cluster of services
• Bloomfilter - Eliminate costly lookups
• Leaky Bucket - Rate limiter
• Token Bucket - Rate limited
• RSync - File transfers
• Raft/Paxos - Consensus algorithm
• Merkle Tree - Identify inconsistencies between nodes
• HyperLogLog - Count unique values fast
• Count-min Sketch - Estimate frequencies of items

Key Topics

System Requirements Analysis

• Understand the functional and non-functional requirements of the system.
• Discuss use cases, user expectations, and system constraints.

System Architecture

• Design a high-level architecture that addresses the system's requirements.
• Discuss components, their responsibilities, and interactions.
• Consider microservices, service-oriented architecture (SOA), or serverless architecture based on the requirements.

Scalability

• Discuss horizontal and vertical scaling strategies.
• Consider sharding, partitioning, and replication techniques.
• Discuss load balancing and auto-scaling mechanisms.

Availability and Fault Tolerance

• Discuss redundancy, failover, and disaster recovery strategies.
• Consider techniques such as replication, data mirroring, and distributed consensus algorithms.
• Discuss how to handle network partitions and node failures.

Consistency and Concurrency

• Discuss consistency models (e.g., eventual consistency, strong consistency) based on application requirements.
• Consider distributed locking, consensus algorithms (e.g., Paxos, Raft), and coordination services (e.g., ZooKeeper, etcd).
• Discuss isolation levels and transaction management.

Data Storage and Retrieval

• Discuss database selection based on requirements (e.g., relational databases, NoSQL databases, distributed key-value stores).
• Consider data partitioning, indexing, and caching strategies.
• Discuss data replication, consistency, and durability guarantees.

Messaging and Communication

• Discuss message queuing, publish-subscribe patterns, and event-driven architectures.
• Consider message brokers (e.g., Kafka, RabbitMQ) and communication protocols (e.g., HTTP, gRPC).

Security

• Discuss authentication, authorization, encryption, and data privacy requirements.
• Consider network security, access control mechanisms, and secure communication protocols.

Monitoring and Management

• Discuss logging, monitoring, and alerting requirements.
• Consider metrics collection, distributed tracing, and centralized logging solutions.
• Discuss deployment strategies and continuous integration/continuous deployment (CI/CD) pipelines.

Cost Optimization

• Discuss resource utilization, cost-effective scaling strategies, and resource provisioning.
• Consider serverless computing, containerization, and cloud services pricing models.

Writing Service Interfaces

Write the outline of the service interface to easily describe the service behavior to the interviewer, by including:

1. message data structures - describes the request and response messages that a service uses to communicate over a netwrk
2. method signature - contains a methods name, return type, and parameter list

GetCourseRequest
    int64 user_id
    int64 timestamp
    string course_name

GetCourseResponse
    int64 userid
    repeated string lessons
    int64 timestamp

# CourseService has a method getCourse which accepts a request message GetCourseRequest and returns the response message GetCourseResponse

# It accepts a single argument, the request, and returns a single object, the response

CourseService
    GetCourseResponse getCourse (GetCourseRequest request)

Reminder: the message data structure is different from the data model - e.g. song_names could be stored as rows that are aggregated into a sequential datastructure like an array in the course response.

A system would have multiple services, and each service would have multiple methods. Making them descriptive makes the interview self-explanatory and self-documenting.