Quick Start Guide

Quick Start Guide Beginner

Got 5 minutes? This guide gets you from zero to productive fast. Skip the theory and jump straight to practical solutions.

::: info This is a reference guide, not a traditional tutorial. For step-by-step learning, start with Chapter 1: Introduction to Algorithms. :::

What You’ll Learn

Estimated time: 15 minutes

By the end of this quick start guide, you will:

• Get productive with algorithms immediately using copy-paste ready solutions
• Learn which algorithm to use for common scenarios (sorting, searching, caching, path-finding)
• Master the decision tree for algorithm selection based on your data and use case
• Discover quick performance wins that can 10x your application speed
• Access framework-specific optimizations for Laravel and Symfony

Prerequisites

No prerequisites required - dive right in! This guide is designed for PHP developers of all levels who want practical solutions now.

🎯 “I Need To…”

Sort Data

quick-sort-example.php

<?php
// Small array (<100 items) or nearly sorted
$data = [64, 34, 25, 12, 22, 11, 90];
sort($data);  // PHP's built-in - BEST choice 99% of the time

// Custom sorting
$users = [
    ['name' => 'John', 'age' => 30],
    ['name' => 'Jane', 'age' => 25],
];
usort($users, fn($a, $b) => $a['age'] <=> $b['age']);

// When to use custom algorithms:
// - Learning/interview prep: See Chapter 5-10
// - Specific constraints: See Chapter 28 (Selection Guide)

→ Full Sorting Guide

Search for Something

search-examples.php

<?php
declare(strict_types=1);

// In unsorted array
$needle = 'value';
$found = in_array($needle, $haystack, true);  // Linear search O(n)

// In sorted array - use binary search
function binarySearch(array $arr, mixed $target): int {
    $left = 0;
    $right = count($arr) - 1;

    while ($left <= $right) {
        $mid = (int)(($left + $right) / 2);
        if ($arr[$mid] === $target) return $mid;
        if ($arr[$mid] < $target) $left = $mid + 1;
        else $right = $mid - 1;
    }

    return -1;  // Not found
}

// Frequent lookups - use hash table
$lookup = array_flip($haystack);  // O(1) search after O(n) setup
$found = isset($lookup[$needle]);

::: tip Performance Tip Binary search is 100x faster than linear search on large sorted arrays. Hash tables give instant O(1) lookups but require O(n) setup time. :::

→ Binary Search | Hash Tables

Cache Results

simple-cache.php

<?php
declare(strict_types=1);

// Simple in-memory cache
class SimpleCache {
    private array $cache = [];

    public function remember(string $key, callable $callback): mixed {
        if (!isset($this->cache[$key])) {
            $this->cache[$key] = $callback();
        }
        return $this->cache[$key];
    }

    public function forget(string $key): void {
        unset($this->cache[$key]);
    }

    public function clear(): void {
        $this->cache = [];
    }
}

// Usage
$cache = new SimpleCache();
$result = $cache->remember('expensive_op', function(): array {
    // Expensive database query or calculation
    return expensiveOperation();
});

// Production: Use Redis
$redis = new Redis();
$redis->connect('127.0.0.1');
$result = $redis->get('key') ?: $redis->setex('key', 3600, expensiveOperation());

::: tip Cache Everything Caching is the #1 performance win. A simple cache can reduce response times from 500ms to 5ms (100x faster)! :::

→ Caching Strategies | Performance Optimization

Find Shortest Path

shortest-path-bfs.php

<?php
declare(strict_types=1);

// Unweighted graph (all edges equal) - Use BFS
function shortestPath(array $graph, int $start, int $end): ?array {
    if ($start === $end) return [$start];
    if (!isset($graph[$start])) return null;

    $queue = [[$start]];
    $visited = [$start => true];

    while (!empty($queue)) {
        $path = array_shift($queue);
        $node = end($path);

        if ($node === $end) return $path;

        foreach ($graph[$node] ?? [] as $neighbor) {
            if (!isset($visited[$neighbor])) {
                $visited[$neighbor] = true;
                $newPath = [...$path, $neighbor];  // PHP 8.4 spread operator
                $queue[] = $newPath;
            }
        }
    }

    return null;  // No path found
}

// Weighted graph - See Dijkstra's algorithm

→ BFS | Dijkstra

Process Large Files

stream-large-file.php

<?php
declare(strict_types=1);

// ❌ BAD: Loads entire file into memory
$lines = file('huge-file.csv');  // OOM for large files

// ✅ GOOD: Stream with generator
function readLargeFile(string $filename): Generator {
    if (!file_exists($filename)) {
        throw new RuntimeException("File not found: $filename");
    }

    $handle = fopen($filename, 'r');
    if ($handle === false) {
        throw new RuntimeException("Cannot open file: $filename");
    }

    try {
        while (($line = fgets($handle)) !== false) {
            yield $line;
        }
    } finally {
        fclose($handle);
    }
}

// Usage - constant memory regardless of file size
foreach (readLargeFile('huge-file.csv') as $lineNumber => $line) {
    processLine($line);
}

::: warning Memory Management Generators use constant memory regardless of file size. A 10GB file uses the same ~100KB memory whether you’re processing 1 line or 1 billion lines! :::

→ Performance Optimization | Stream Processing

Optimize Slow Code

optimization-examples.php

<?php
// 1. Profile first
$start = microtime(true);
slowFunction();
echo (microtime(true) - $start) * 1000 . "ms\n";

// 2. Common fixes:

// ❌ N+1 database queries
foreach ($users as $user) {
    $posts = $db->query("SELECT * FROM posts WHERE user_id = ?", [$user['id']]);
}

// ✅ Single query with JOIN
$results = $db->query("
    SELECT u.*, p.*
    FROM users u
    LEFT JOIN posts p ON u.id = p.user_id
");

// ❌ Repeated calculations in loop
for ($i = 0; $i < count($array); $i++) {  // count() called every iteration
    process($array[$i]);
}

// ✅ Calculate once
$n = count($array);
for ($i = 0; $i < $n; $i++) {
    process($array[$i]);
}

// ❌ String concatenation in loop
$result = '';
foreach ($items as $item) {
    $result .= $item;  // Creates new string each time
}

// ✅ Use array and implode
$parts = [];
foreach ($items as $item) {
    $parts[] = $item;
}
$result = implode('', $parts);

→ Performance Guide

---

🚦 Decision Tree

”Which Algorithm Should I Use?”

Need to process data?
├─ Search for specific item?
│  ├─ Data sorted? → Binary Search (O(log n))
│  ├─ Many searches? → Hash Table (O(1))
│  └─ Unsorted, one search? → Linear Search (O(n))
│
├─ Sort data?
│  ├─ Use PHP sort() → DONE (best choice 99% of time)
│  ├─ Need custom order? → usort() with comparator
│  └─ Learning? → See Chapters 5-10
│
├─ Find path between nodes?
│  ├─ Unweighted graph? → BFS (shortest path)
│  ├─ Weighted graph? → Dijkstra's Algorithm
│  └─ Just any path? → DFS
│
├─ Optimize repeated calculations?
│  ├─ Overlapping subproblems? → Dynamic Programming
│  ├─ Expensive function calls? → Memoization/Caching
│  └─ Database queries? → Query result caching
│
└─ Process large dataset?
   ├─ Fits in memory? → Regular arrays
   ├─ Too large for memory? → Generators/Streaming
   └─ Need fast lookups? → Hash table/Redis

---

📊 Complexity Cheat Sheet

Complexity	Max Input Size	Examples
O(1)	Unlimited	Array access, hash table lookup
O(log n)	Billions	Binary search, balanced tree ops
O(n)	100 million	Linear search, array traversal
O(n log n)	10 million	Merge sort, quick sort, heap sort
O(n²)	10,000	Bubble sort, selection sort, nested loops
O(2^n)	~20	Subset generation, backtracking
O(n!)	~10	Permutations, TSP brute force

Rule of thumb: If n > 10,000 and you have O(n²), you need a better algorithm.

📏 Real-World Input Sizes

Understanding typical data scales helps you choose the right algorithm for your use case:

Use Case	Typical Size	Algorithm Needed	Why
User dropdown options	10-100	Any (even O(n²))	Too small to matter
Search autocomplete	1,000-10,000	Hash table/Trie	Need instant O(1) lookups
E-commerce catalog	10,000-100,000	Indexed DB search	Database handles it
Social media feed	100-1,000/page	Pagination + O(n)	Per-page processing
Analytics dashboard	10,000-1M rows	O(n log n) max	Must complete in seconds
Log file processing	1M-1B entries	Streaming/generators	Can’t fit in memory
ML training data	1M-100M records	Specialized + distributed	Requires infrastructure

Quick Size Check:

size-check-helper.php

<?php
declare(strict_types=1);

function chooseApproach(int $n): string {
    return match(true) {
        $n < 100 => "✅ Use any algorithm, even O(n²)",
        $n < 10_000 => "⚠️  O(n²) acceptable, O(n log n) better",
        $n < 1_000_000 => "⚡ Need O(n log n), prefer O(n)",
        default => "🔥 Must use O(n) or streaming"
    };
}

// Examples
echo chooseApproach(50) . "\n";        // Any algorithm
echo chooseApproach(5_000) . "\n";     // O(n²) acceptable
echo chooseApproach(500_000) . "\n";   // Need O(n log n)
echo chooseApproach(10_000_000) . "\n"; // Must stream

::: tip Pro Tip If your “large” dataset fits comfortably in a PHP array, it’s probably not that large. True big data requires databases, message queues, or streaming processors. :::

::: info Quick Reference Most Common: O(1) hash lookup, O(n) array scan, O(n log n) sort, O(log n) binary search.
Avoid in Production: O(n²) nested loops on large data, O(2^n) recursive solutions without memoization. :::

→ Full Complexity Guide | Appendix A

---

🛠️ Common Patterns

Two Pointers

two-pointers-pattern.php

<?php
declare(strict_types=1);

// Find pair that sums to target (sorted array)
function twoSum(array $arr, int $target): ?array {
    $left = 0;
    $right = count($arr) - 1;

    while ($left < $right) {
        $sum = $arr[$left] + $arr[$right];
        if ($sum === $target) return [$left, $right];
        if ($sum < $target) $left++;
        else $right--;
    }

    return null;
}

// Usage example
$numbers = [1, 2, 3, 4, 6, 8, 9];
$result = twoSum($numbers, 10);
// Returns: [1, 5] because $numbers[1] + $numbers[5] = 2 + 8 = 10

Sliding Window

sliding-window-pattern.php

<?php
declare(strict_types=1);

// Maximum sum of k consecutive elements
function maxSumSubarray(array $arr, int $k): int {
    if (count($arr) < $k) {
        throw new InvalidArgumentException("Array too small for window size");
    }

    $maxSum = $currentSum = array_sum(array_slice($arr, 0, $k));

    for ($i = $k; $i < count($arr); $i++) {
        $currentSum = $currentSum - $arr[$i - $k] + $arr[$i];
        $maxSum = max($maxSum, $currentSum);
    }

    return $maxSum;
}

// Usage example
$sales = [100, 200, 300, 400, 100, 200];
$maxThreeDaySales = maxSumSubarray($sales, 3);
// Returns: 900 (days 2-4: 200 + 300 + 400)

Fast & Slow Pointers

fast-slow-pointers-pattern.php

<?php
declare(strict_types=1);

// Detect cycle in linked list (Floyd's Cycle Detection)
function hasCycle(?ListNode $head): bool {
    if ($head === null) return false;

    $slow = $fast = $head;

    while ($fast !== null && $fast->next !== null) {
        $slow = $slow->next;
        $fast = $fast->next->next;

        if ($slow === $fast) return true;  // Cycle detected
    }

    return false;  // No cycle
}

// Also known as "Floyd's Tortoise and Hare" algorithm
// Time: O(n), Space: O(1)

::: tip Pattern Recognition These three patterns solve 80% of array/linked list interview questions. Master them and you’ll recognize when to apply them instantly. :::

---

🔍 Recognize Algorithms in Code

Learn to identify algorithms by their code patterns. This helps you understand existing code and spot optimization opportunities.

Common Algorithm Signatures

Two Nested Loops = O(n²) Brute Force:

<?php
# Pattern: Comparing all pairs

foreach ($arr as $i => $val1) {
    foreach ($arr as $j => $val2) {
        if ($val1 + $val2 === $target) {
            // O(n²) - check all combinations
        }
    }
}

Two Pointers = O(n) Optimized:

<?php
# Pattern: Left and right pointers moving toward each other

$left = 0;
$right = count($arr) - 1;
while ($left < $right) {
    // O(n) - single pass through sorted array
}

Divide in Half = O(log n) Binary Search:

<?php
# Pattern: Repeatedly dividing search space

$mid = (int)(($left + $right) / 2);
if ($arr[$mid] === $target) {
    return $mid;
}
// O(log n) - eliminates half each iteration

Hash Table Lookup = O(n) Space-Time Tradeoff:

<?php
# Pattern: Build lookup table, then query

$seen = [];
foreach ($arr as $val) {
    if (isset($seen[$val])) {
        return true;  // Found duplicate
    }
    $seen[$val] = true;
}
// O(n) time, O(n) space - trade memory for speed

Memoization Cache = Dynamic Programming:

<?php
# Pattern: Cache to avoid recomputation

static $cache = [];
if (isset($cache[$key])) {
    return $cache[$key];  // Return cached result
}
$result = expensiveCalculation($key);
$cache[$key] = $result;
// Overlapping subproblems → DP

Sort Then Process = O(n log n) Foundation:

<?php
# Pattern: Sort enables faster algorithms

sort($arr);  // O(n log n)
// Now can use binary search O(log n)
// Or two pointers O(n)
// Total: O(n log n)

Algorithm Spotting Cheat Sheet

Code Pattern	Algorithm Type	Complexity	Use Case
Two nested loops	Brute force	O(n²)	Small data or first solution
while ($left < $right)	Two pointers	O(n)	Sorted array problems
Divide by 2 repeatedly	Binary search	O(log n)	Find in sorted data
Build hash table first	Hash table	O(n)	Fast lookups needed
Recursive + cache	Dynamic programming	Varies	Overlapping subproblems
Queue for nodes	BFS	O(V+E)	Shortest path (unweighted)
Stack/recursion for nodes	DFS	O(V+E)	Path finding, traversal
sort() then iterate	Sort-based	O(n log n)	Need ordering first

::: info Quick Diagnosis See nested loops? → Probably O(n²), look for optimization.
See hash table? → Trading space for speed.
See recursion? → Check for duplicate work (DP candidate).
See sort first? → Likely O(n log n) is the best you can do. :::

---

🎯 By Use Case

E-Commerce

ecommerce-examples.php

<?php
// Product recommendations - collaborative filtering
$similarUsers = findSimilarUsers($userId);  // Graph/clustering
$productScores = calculateScores($similarUsers);  // Aggregation
arsort($productScores);  // Sort by score
$recommendations = array_slice(array_keys($productScores), 0, 10);

// Inventory sorting - priority queue
$orders = new SplPriorityQueue();
$orders->insert($order1, $order1->priority);
$topOrder = $orders->extract();

APIs

api-examples.php

<?php
// Rate limiting - token bucket
class RateLimiter {
    public function checkLimit(string $userId): bool {
        $key = "rate:$userId";
        $redis = new Redis();
        $redis->connect('127.0.0.1');

        $current = $redis->incr($key);
        if ($current === 1) {
            $redis->expire($key, 60);  // 1 minute window
        }

        return $current <= 100;  // 100 requests per minute
    }
}

// Response caching - LRU cache
$cache = new LRUCache(1000);
$response = $cache->get($cacheKey) ?? $cache->set($cacheKey, generateResponse());

Search

search-use-case.php

<?php
// Autocomplete - Trie data structure
class AutocompleteNode {
    public array $children = [];
    public bool $isEnd = false;
    public array $suggestions = [];
}

// Full-text search - inverted index + TF-IDF
$index = buildInvertedIndex($documents);  // term → document IDs
$scores = calculateTFIDF($queryTerms, $index);  // Relevance scores
arsort($scores);

---

💡 Quick Wins

1. Use PHP’s Built-in Functions (Don’t Reinvent the Wheel)

use-built-ins.php

<?php
declare(strict_types=1);

// ❌ DON'T: Implement basic algorithms yourself
function mySort(array $arr): array {
    // 50 lines of sorting code...
    return $arr;
}

// ✅ DO: Use PHP's optimized built-ins
sort($array);  // Implemented in C, handles edge cases
$found = in_array($item, $array, true);  // Strict comparison
$sum = array_sum($numbers);
$filtered = array_filter($array, $callback);
$unique = array_unique($array);
$reversed = array_reverse($array);
$keys = array_keys($array);
$values = array_values($array);

When to use custom algorithms:

• ✅ Specific business logic not in built-ins
• ✅ Performance-critical (after profiling proves it!)
• ✅ Unique constraints PHP doesn’t handle
• ✅ Learning/interview preparation

When to use built-ins:

• ✅ 99% of the time (seriously!)
• ✅ Standard operations (sort, search, filter, map)
• ✅ Production code (battle-tested, optimized)
• ✅ When you want maintainable code

2. Cache Expensive Operations

cache-expensive-ops.php

<?php
// ❌ Recalculates every time
function getStats($userId) {
    return calculateExpensiveStats($userId);  // 500ms
}

// ✅ Cache for 5 minutes
function getStats($userId) {
    $key = "stats:$userId";
    return $cache->remember($key, 300, fn() => calculateExpensiveStats($userId));
}

3. Use Generators for Large Data

use-generators.php

<?php
// ❌ 1GB memory for 1M records
$users = User::all();
foreach ($users as $user) {
    process($user);
}

// ✅ Constant memory
$users = User::cursor();  // Returns generator
foreach ($users as $user) {
    process($user);
}

4. Batch Database Operations

batch-operations.php

<?php
// ❌ 1000 queries
foreach ($users as $user) {
    DB::insert('INSERT INTO logs (user_id) VALUES (?)', [$user->id]);
}

// ✅ 1 query
$values = array_map(fn($u) => "({$u->id})", $users);
DB::insert('INSERT INTO logs (user_id) VALUES ' . implode(',', $values));

5. Enable OPcache in Production

; php.ini
opcache.enable=1
opcache.memory_consumption=256
opcache.interned_strings_buffer=16
opcache.max_accelerated_files=10000
opcache.validate_timestamps=0  ; Production only
opcache.jit=tracing  ; PHP 8.0+ JIT compiler
opcache.jit_buffer_size=128M

::: warning Production Only Never set opcache.validate_timestamps=0 in development! This disables file change detection. Your code changes won’t take effect without restarting PHP-FPM. :::

---

🎓 Learning Paths

”I Have 1 Hour”

1. Read this page ✓
2. Introduction & Big O (15 min)
3. Hash Tables (20 min)
4. Caching (25 min)

“I’m Preparing for Interviews”

1. All Sorting (Chapters 5-10)
2. Binary Search (Chapter 12)
3. Trees & Traversals (Chapters 18-19)
4. DFS & BFS (Chapters 22-23)
5. Dynamic Programming (Chapters 25-26)

→ Full Interview Prep Path

”I Want to Optimize My App”

1. Benchmarking
2. Algorithm Selection
3. Caching
4. Performance
5. Case Studies

→ Full Optimization Path

---

📖 Essential References

Quick Lookups:

• Complexity Cheat Sheet - Big O reference
• PHP Performance Tips - Optimization guide
• Glossary - Term definitions

Common Tasks:

• Sorting: Chapters 5-10
• Searching: Chapters 11-14
• Graphs: Chapters 21-24
• Optimization: Chapters 27-30

---

⚠️ Common Pitfalls to Avoid

1. Premature Optimization

avoid-premature-optimization.php

<?php
// ❌ DON'T: Optimize before measuring
// "I'll use a complex algorithm because it's O(n log n)"

// ✅ DO: Profile first, then optimize
$start = microtime(true);
simpleAlgorithm();  // Start simple
$time = microtime(true) - $start;

if ($time > 0.1) {  // Only if it's actually slow
    optimizedAlgorithm();
}

Rule: Make it work, make it right, make it fast - in that order.

::: danger Premature Optimization Kills Projects “Premature optimization is the root of all evil” — Donald Knuth. Profile first, optimize second. 97% of optimizations are wasted effort if applied to the wrong place. :::

2. Wrong Data Structure

choose-right-data-structure.php

<?php
// ❌ BAD: Using array with in_array() for lookups
$validUsers = [1, 2, 3, 4, 5, /* ...1000 more */];
if (in_array($userId, $validUsers)) {  // O(n) - SLOW!
    // ...
}

// ✅ GOOD: Use associative array for O(1) lookups
$validUsers = [1 => true, 2 => true, 3 => true, /* ... */];
if (isset($validUsers[$userId])) {  // O(1) - FAST!
    // ...
}

3. Hidden N+1 Queries (Laravel)

avoid-n-plus-one.php

<?php
// ❌ BAD: 1 + N queries
$users = User::all();  // 1 query
foreach ($users as $user) {
    echo $user->posts->count();  // N queries - ONE PER USER!
}
// Result: 1 + 1000 users = 1001 queries! 🔥

// ✅ GOOD: 2 queries total
$users = User::withCount('posts')->get();  // 2 queries with JOIN
foreach ($users as $user) {
    echo $user->posts_count;  // No query!
}
// Result: 2 queries total (500x faster) ⚡

::: warning N+1 Query Detector Enable Laravel Debugbar or Telescope in development to catch N+1 queries. This is the #1 performance killer in Laravel apps. :::

4. Memory Leaks in Loops

avoid-memory-leaks.php

<?php
// ❌ BAD: Memory accumulates
$results = [];
foreach ($hugeDataset as $item) {
    $results[] = process($item);  // Grows without bounds
}

// ✅ GOOD: Process and discard
foreach ($hugeDataset as $item) {
    $result = process($item);
    sendToOutput($result);  // Process one at a time
    unset($result);  // Free memory
}

5. Inefficient String Building

efficient-string-building.php

<?php
// ❌ BAD: O(n²) - Creates new string each time
$html = '';
foreach ($items as $item) {
    $html .= "<li>$item</li>";  // SLOW for large arrays
}

// ✅ GOOD: O(n) - Build array then join
$parts = [];
foreach ($items as $item) {
    $parts[] = "<li>$item</li>";
}
$html = implode('', $parts);

---

🔒 Quick Security Tips

1. Always Use Parameterized Queries

secure-queries.php

<?php
// ❌ DANGER: SQL injection
$result = $db->query("SELECT * FROM users WHERE id = {$_GET['id']}");

// ✅ SAFE: Prepared statements
$stmt = $db->prepare("SELECT * FROM users WHERE id = ?");
$stmt->execute([$_GET['id']]);

2. Avoid Timing Attacks

avoid-timing-attacks.php

<?php
// ❌ BAD: Vulnerable to timing attacks
if ($userHash === $expectedHash) {  // Character-by-character comparison
    return true;
}

// ✅ GOOD: Constant-time comparison
if (hash_equals($expectedHash, $userHash)) {
    return true;
}

3. Use Secure Random for Tokens

secure-random.php

<?php
// ❌ BAD: Predictable
$token = md5(time() . rand());

// ✅ GOOD: Cryptographically secure
$token = bin2hex(random_bytes(32));

---

🧰 Essential Tools

Profiling & Debugging

profiling-tools.php

<?php
// 1. Quick memory check
echo "Memory: " . memory_get_usage() / 1024 / 1024 . " MB\n";
echo "Peak: " . memory_get_peak_usage() / 1024 / 1024 . " MB\n";

// 2. Query logging (Laravel)
DB::enableQueryLog();
// ... run queries ...
dd(DB::getQueryLog());

// 3. Execution time breakdown
$times = [];
$start = microtime(true);
step1();
$times['step1'] = microtime(true) - $start;

$start = microtime(true);
step2();
$times['step2'] = microtime(true) - $start;

print_r($times);

Recommended Tools

• Xdebug: Development profiling (slow, detailed)
• Blackfire: Production profiling (fast, safe)
• Telescope (Laravel): Request/query monitoring
• Clockwork: Browser-based profiling
• New Relic/DataDog: APM for production

---

🎯 Framework-Specific Quick Wins

Laravel

laravel-optimizations.php

<?php
// 1. Eager loading (prevents N+1)
$users = User::with('posts')->get();  // 2 queries instead of N+1

// 2. Chunk large datasets
User::chunk(1000, function ($users) {
    foreach ($users as $user) {
        // Process in batches
    }
});

// 3. Cache query results
$users = Cache::remember('users', 3600, function () {
    return User::all();
});

// 4. Use whereIn instead of loop
$userIds = [1, 2, 3, 4, 5];
$users = User::whereIn('id', $userIds)->get();  // 1 query, not 5

// 5. Select only needed columns
$users = User::select(['id', 'name'])->get();  // Not SELECT *

Symfony

symfony-optimizations.php

<?php
// 1. Query builder for complex queries
$qb = $entityManager->createQueryBuilder();
$users = $qb->select('u')
    ->from('App\Entity\User', 'u')
    ->where('u.active = :active')
    ->setParameter('active', true)
    ->getQuery()
    ->getResult();

// 2. Batch processing
$batchSize = 1000;
$i = 0;
foreach ($users as $user) {
    $entityManager->persist($user);
    if (($i % $batchSize) === 0) {
        $entityManager->flush();
        $entityManager->clear();
    }
    $i++;
}

// 3. HTTP cache
$response->setCache([
    'public' => true,
    'max_age' => 3600,
]);

---

❓ FAQs

Q: Should I always use the most efficient algorithm? A: No! Use the simplest algorithm that meets your requirements. sort() is fine for 99% of cases. Optimize only when you have measured performance issues.

Q: When should I NOT implement a custom algorithm? A: When PHP has a built-in that does it (sort(), array_filter(), array_map()), when you haven’t profiled yet, when the built-in is “good enough,” or when maintainability matters more than micro-optimizations.

Q: When do I really need to know this stuff? A: Job interviews, optimizing slow code, working with large datasets, understanding framework internals, debugging production issues.

Q: Can’t PHP frameworks handle this for me? A: Frameworks help, but you need to understand when to use whereIn() vs where(), when to cache, when to eager load, etc. Frameworks give you tools, but you need to know when to use them.

Q: What’s the #1 optimization I should know? A: Caching. It’s the biggest bang for your buck. Cache database queries, API responses, expensive calculations. A simple cache can turn a 500ms request into a 5ms request.

Q: How do I know if my code is slow? A: Measure! Use microtime(), Xdebug, or Blackfire. Never optimize without profiling first. “Premature optimization is the root of all evil.”

Q: Should I learn all the algorithms in this series? A: Start with the essentials (sorting, searching, hash tables, caching). Learn advanced topics (dynamic programming, graph algorithms) when you need them or for interview prep.

Q: Where do I start if I’m completely new? A: Chapter 1: Introduction for comprehensive learning, or continue with this guide for copy-paste solutions.

Q: What about async/await in PHP? A: PHP 8.1+ supports fibers, but for practical async use ReactPHP, Swoole, or Amp. See Chapter 31: Concurrent Algorithms.

Q: How do I know if my data is “large” enough to worry about? A: If it’s < 10,000 items and fits in a PHP array comfortably, don’t overthink it. Use simple algorithms. If it’s > 100,000 or causes memory issues, then optimize. Profile first!

Q: How do I recognize what algorithm is being used in code I’m reading? A: Look for patterns: nested loops = O(n²), binary divide = O(log n), hash table = space/time tradeoff, recursion + cache = dynamic programming. See the “Recognize Algorithms in Code” section above.

---

🚀 Next Steps

Ready to dive deeper?

• Start Chapter 1 - Full course
• See all paths - Choose your journey
• Use case guide - Find what you need

Need help?

• Glossary - Look up terms
• Cheat Sheet - Quick reference
• Review this guide - Come back anytime

---

💾 Copy-Paste Snippets

Benchmark Function

benchmark-helper.php

<?php
declare(strict_types=1);

function benchmark(callable $fn, array $args = [], int $iterations = 1): float {
    $times = [];

    for ($i = 0; $i < $iterations; $i++) {
        $start = microtime(true);
        $fn(...$args);
        $times[] = (microtime(true) - $start) * 1000;
    }

    return $iterations === 1 ? $times[0] : array_sum($times) / count($times);
}

// Single run
echo "Execution time: " . benchmark(fn() => expensiveFunction()) . "ms\n";

// Average of 100 runs for accuracy
echo "Average time: " . benchmark(fn() => quickFunction(), [], 100) . "ms\n";

Simple Cache Class

cache-class.php

<?php
declare(strict_types=1);

class Cache {
    private static array $store = [];

    public static function remember(string $key, int $ttl, callable $callback): mixed {
        // Check if cached and not expired
        if (isset(self::$store[$key]) && self::$store[$key]['expires'] > time()) {
            return self::$store[$key]['value'];
        }

        // Generate and cache
        $value = $callback();
        self::$store[$key] = [
            'value' => $value,
            'expires' => time() + $ttl,
            'hits' => 0
        ];
        return $value;
    }

    public static function forget(string $key): void {
        unset(self::$store[$key]);
    }

    public static function flush(): void {
        self::$store = [];
    }

    public static function stats(): array {
        return [
            'keys' => count(self::$store),
            'memory' => memory_get_usage()
        ];
    }
}

Binary Search

binary-search-helper.php

<?php
declare(strict_types=1);

function binarySearch(array $arr, mixed $target): int {
    $left = 0;
    $right = count($arr) - 1;

    while ($left <= $right) {
        $mid = (int)(($left + $right) / 2);

        if ($arr[$mid] === $target) return $mid;
        if ($arr[$mid] < $target) $left = $mid + 1;
        else $right = $mid - 1;
    }

    return -1;  // Not found
}

// Usage
$numbers = [1, 3, 5, 7, 9, 11, 13, 15];
$index = binarySearch($numbers, 7);  // Returns: 3
$missing = binarySearch($numbers, 6);  // Returns: -1

File Stream Generator

file-stream-generator.php

<?php
declare(strict_types=1);

function streamFile(string $filename): Generator {
    if (!file_exists($filename)) {
        throw new RuntimeException("File not found: $filename");
    }

    $handle = fopen($filename, 'r');
    if ($handle === false) {
        throw new RuntimeException("Cannot open file: $filename");
    }

    try {
        while (!feof($handle)) {
            $line = fgets($handle);
            if ($line !== false) {
                yield $line;
            }
        }
    } finally {
        fclose($handle);
    }
}

// Usage: Process 10GB file with constant memory
foreach (streamFile('huge.log') as $lineNumber => $line) {
    if (str_contains($line, 'ERROR')) {
        echo "Line $lineNumber: $line";
    }
}

---

---

Ready for More?

This guide covers the essentials. For deep understanding, continue with the full series.

Start Full Course →