Programming in Java

Unit 11: Exceptions, Assertions & Multithreading

From graceful error handling to concurrent execution — master Java's exception hierarchy, assertion mechanisms, and multithreading to build production-grade applications like Zerodha and Paytm.

⏱️ 8 hrs theory + 6 hrs lab | 💰 ₹15K–₹50K/month | 📝 30 MCQs (Bloom's Mapped)

💼 Jobs this unlocks: Java Backend Developer (₹6–12 LPA) | Multithreaded Systems Engineer (₹8–18 LPA) | Fintech Developer (₹10–25 LPA)

Section A

Opening Hook — When 10 Lakh Orders Hit Zerodha at 9:15 AM

🏢 How Zerodha Handles ₹15 Lakh Crore Without Crashing

Every trading day at 9:15 AM, the Indian stock market opens. In that single second, Zerodha processes over 10 lakh concurrent buy/sell orders. Each order is a separate thread. When two traders try to modify the same shared portfolio balance simultaneously, Java's synchronized keyword ensures only one thread touches the balance at a time — preventing ₹15 lakh crore worth of data corruption.

But what happens when a UPI payment to Zerodha fails mid-transaction? A network timeout? An invalid stock symbol? The system doesn't crash — it catches the exception gracefully, logs the error, sends you a friendly "Order failed — please retry" message, and keeps running for the other 9,99,999 orders.

This unit teaches you both pillars: Exceptions (how to handle failures without crashing) and Multithreading (how to run thousands of tasks simultaneously). Together, they make Java the #1 language for fintech, banking, and high-frequency trading systems across India.

🇮🇳 Zerodha🇮🇳 Paytm🇮🇳 PhonePe🇮🇳 Razorpay🇮🇳 HDFC Bank🇮🇳 NSE/BSE

Zerodha's backend is built primarily in Java and Go. Their Kite platform handles 15+ million orders daily with sub-millisecond latency. Every order goes through exception handling (for invalid quantities, insufficient balance, market closed) and multithreading (for concurrent order matching). Nithin Kamath, Zerodha's founder, has stated that robust error handling saved them from ₹100+ crore losses during market volatility events.

Section B

Learning Outcomes — Bloom's Taxonomy Mapped

Bloom's Level	Learning Outcome
🔴 Remember	List the exception hierarchy: Throwable → Error/Exception → RuntimeException, and differentiate checked vs unchecked exceptions
🔴 Remember	Identify the five states in the Java thread lifecycle: NEW → RUNNABLE → BLOCKED → WAITING → TERMINATED
🟠 Understand	Explain the execution flow of try-catch-finally and how try-with-resources auto-closes resources
🟠 Understand	Describe how `synchronized` prevents race conditions in concurrent access to shared data
🟢 Apply	Write custom checked and unchecked exceptions (InsufficientBalanceException) with exception chaining
🟢 Apply	Create multithreaded programs using Thread class, Runnable interface, and lambda expressions
🟡 Analyze	Analyze deadlock scenarios and determine prevention strategies using lock ordering
🟡 Analyze	Compare Thread class extension vs Runnable implementation vs lambda-based thread creation
🔵 Evaluate	Evaluate when to use `ExecutorService` thread pools vs manual thread management in production systems
🔵 Evaluate	Assess the trade-offs of assert vs exception throwing for input validation in different deployment scenarios
🟣 Create	Design a producer-consumer system (Zerodha Order Book) using wait/notify with full exception handling
🟣 Create	Build a multithreaded bank transfer simulation with synchronized blocks and custom exception chains