Distributed Systems for Practitioners

0/7 Objectives Completed (0%)

Introduction to Distributed Systems

Before Getting Started

Introduction to Distributed Systems

Getting Started Fallacies of Distributed Computing Difficulties Designing Distributed Systems Measures of Correctness in Distributed Systems System Models Types of Failures The Tale of Exactly-Once Semantics Failure in the World of Distributed Systems Stateless and Stateful Systems Quiz

Core Concepts and Theoretical Foundations

Basic Concepts and Theorems

Partitioning Algorithms for Horizontal Partitioning Replication Primary-Backup Replication Algorithm Multi-Primary Replication Algorithm Quorums in Distributed Systems Let AI Evaluate Your Understanding of Replication Algorithms Safety Guarantees in Distributed Systems ACID Transactions The CAP Theorem Consistency Models CAP Theorem's Consistency Model Isolation Levels and Anomalies Prevention of Anomalies in Isolation Levels Consistency and Isolation Hierarchy of Models Why All the Formalities?Quiz

Understanding Distributed Transactions

Distributed Transactions

Introduction to Distributed Transactions Quiz

Achieving Isolation

Achieving Serializability Pessimistic Concurrency Control (PCC)Optimistic Concurrency Control (OCC)Achieving Snapshot Isolation Achieving Full Serializable Snapshot Isolation Quiz

Achieving Atomicity

Hard to Guarantee Atomicity 2-Phase Commit (2PC)3-Phase Commit (3PC)Quorum-Based Commit Protocol Quiz

Concluding Distributed Transactions

How It All Fits Together Long-Lived Transactions and Sagas

Consensus and Time in Distributed Systems

Consensus

Defining the Consensus Problem FLP Impossibility The Paxos Algorithm Intricacies of Paxos Paxos in Real Life Replicated State Machine via Consensus Distributed Transactions via Consensus An introduction to Raft Communication among Raft Nodes Raft's Implementation Standing on the Shoulders of Giants Quiz

Time

What is Different in a Distributed System?A Practical Perspective A Theoretical Perspective Logical Clocks Quiz

Order

Total and Partial Ordering The Concept of Causality Lamport Clocks Vector Clocks Version Vectors Dotted Version Vectors Distributed Snapshot Problem Solving the Distributed Snapshot Problem Physical and Logical Time: Closing Thoughts Quiz

Networking and Security

Networking

Introduction The Physical Layer The Link Layer - Services The Link-Layer Protocols The Network Layer The Transport Layer The Application Layer Taking a Step Back Quiz

Security

Introduction Authentication Confidentiality Integrity A Cryptography Primer Symmetric/Asymmetric Encryption and Digital Signatures Quiz

Security Protocols

Transport Layer Security (TLS)Public-Key Infrastructure (PKI)Web of Trust (PGP)OAuth Protocol Quiz

Practical Applications and Case Studies

From Theory to Practice

Case Study 1: Distributed File Systems

Hadoop Distributed File System and Google File System Creating and Reading Files Writing and Deleting Files GFS Consistency Model Quiz

Case Study 2: Distributed Coordination Service

Coordination Service Zookeeper Guarantees Provided by Zookeeper Zookeeper's ZAB Protocol Examples of Powerful Primitives by Zookeeper API Quiz

Case Study 3: Distributed Data Stores

Introduction BigTable/HBase Architecture Appends and Read Operations in HBase Guarantees Provided by HBase Cassandra's Data Model Cassandra's Cluster Internode Communication Cassandra's Consistency Levels Linearizability Violations in Cassandra Linearizability Guarantees by Cassandra Cassandra Performing Queries Efficiently Spanner's Data Model Spanner's Architecture Spanner using TrueTime Spanner Operations FaunaDB Quiz

Case Study 4: Distributed Messaging System

Introduction to Kafka Kafka Levers Kafka's Messaging Guarantees Transactions, Storage Layout, and other Guarantees Quiz

Case Study 5: Distributed Cluster Management

Kubernetes Components of Manager and Worker Nodes Quiz

Case Study 6: Distributed Ledger

Introduction to Corda Corda's Data Model Corda's Architecture Backwards Compatibility provided by Corda Quiz

Case Study 7: Distributed Data Processing Systems

Introduction Introduction to MapReduce MapReduce's Manager-Worker Architecture Introduction to Apache Spark DAG of Stages in Apache Spark Perks of Apache Spark Apache Flink Time and Watermarks in Flink Failure Recovery in Flink Quiz

Patterns and Practices in Distributed Systems

Practices & Patterns

Communication Patterns

Creating and Parsing Data Transfer of Data Datastores for Asynchronous Communication Communication Models

Coordination Patterns

Coordination Patterns

Data Synchronization

Data Synchronisation Event Sourcing Change Data Capture (CDC)

Shared-nothing Architectures

Sharing Problems and their Solution Benefits and Drawbacks

Distributed Locking

Leases in Distributed Systems Preventing Safety Risks in Leases

Compatibility Patterns

Backwards Compatibility Maintaining Backwards Compatibility

Dealing with Failure

Failure Handling Techniques Applying Failure Handling Techniques Retries Containing Impact of Failure Backpressure Reacting to Backpressure

Distributed Tracing

Recording Program's Execution

Concluding this Course

Recap of the Course Some More Things to Discover

Introduction

Let's look at some prerequisites and what you can expect from this course.

We'll cover the following

Target audience
What to expect
What not to expect

To take this course, you don’t need any prior knowledge of the concepts and algorithms for distributed systems.

Target audience

This course is for:

Software engineers with little to no experience with distributed systems
People that don’t develop software, but want to learn about the field of distributed systems

What to expect

This course will help you get the underlying logic behind a concept or an algorithm.

It also provides the necessary references to the original papers, so that you can study other parts of interest in greater depth.

It offers a gradual introduction to all necessary terms, as well as explanations of the basic algorithms, with the help of diagrams and examples.

What not to expect

This course does not aim to provide a full analysis or proof of every single algorithm.