Problems · Performance Testing 101

Common Performance Problems

The 7 most frequent performance issues encountered during testing and in production — what they mean, why they happen, and how they manifest.

7 Problems Every Tester Must Know

Reduced Throughput

📉

The system is processing fewer requests/transactions per second than expected. Users experience delays as the system cannot handle incoming traffic fast enough.

Why It Happens

Slow Backend Calls Database Slowness Insufficient CPU Poorly Written Code Network Bottlenecks

Example

EXPECTED 100 req/sec → ACTUAL 40 req/sec

Poor Response Time

🐌

Each request takes too long to return a response. The delay is felt directly by end users as sluggish interactions.

Why It Happens

Heavy Server Processing Slow DB Queries High Network Latency Thread Pool Saturation

Example

SLA Login API: 200 ms → ACTUAL 2,000 ms

10× slower than expected — users abandon the page.

Lack of Scalability

📊

Performance does not increase when more resources (CPU / RAM / instances) are added. Throwing hardware at the problem doesn't fix it.

Why It Happens

Architecture Limitations Single-Threaded Processes DB Bottlenecks Global Locks

Example

ACTION Servers: 2 → 4 instances

RESULT Throughput barely increases

Server Can't Handle X Users

💥

The application crashes, hangs, or becomes slow when user count crosses a certain limit.

Why It Happens

CPU Spikes to 100% Memory Exhaustion Thread Pool Full DB Pool Maxed Out

Example

STABLE Up to 500 users ✓

FAILS At 600+ users — unstable

Heap Problem

🧠

Java heap memory is not managed well → leads to OutOfMemoryError, long GC pauses, or memory leaks. Memory keeps increasing and never comes down.

Symptoms

Memory Never Reclaimed Frequent Full GC GC Pause Times ↑ OutOfMemoryError

Example

UNDER LOAD Heap usage → 100%

Very slow responses → eventual application crash.

Thread Pool Issue

🧵

All available threads are occupied → new requests wait in queue or get rejected. The server is alive but unable to process more work.

Why It Happens

Slow Backend Calls Blocking I/O Small Thread Pool Size

Example

POOL 50 threads ← INCOMING 200 requests

150 requests wait → high response time → 503 Service Unavailable errors.

Long Running DB Queries

🗄️

Queries take too much time due to poor SQL design or missing indexes. This bottleneck cascades and slows down the entire application.

Why It Happens

Missing Indexes Large Data Scans Complex Joins Query Locks High DB CPU

Example

NORMAL "Get User Orders": 50 ms

UNDER LOAD 3,000 ms (table scan + lock contention)

60× slower → entire application bottlenecked.

∑

Quick Reference Summary

Problem	Meaning	Example
Reduced Throughput	System handles fewer req/sec	Expected 100 TPS → actual 40 TPS
Poor Response Time	Requests take longer to complete	Login: 2 sec instead of 200 ms
Lack of Scalability	More resources ≠ better performance	+= servers but same TPS
Server User Limit	Crashes/slow at certain user count	Stable till 500, fails at 600
Heap Problem	Memory leak / GC issues	Heap 100%, app slows or crashes
Thread Pool Issue	No free threads → requests wait	Pool: 50; Incoming: 200 requests
Long DB Queries	Slow SQL → delays entire app	Query: 3 sec instead of 50 ms