메모리 풀 : part 2 - 구현 및 성능 분석

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
#include <iostream>
#include <vector>
#include <chrono>
#include <thread>
#include <mutex>
#include <cstdlib>
#include <cstdint>
#include <stdlib.h>  // rand, srand
#include <ctime>     // time

using namespace std;
using namespace std::chrono;

//---------------------------------------------------------------------------
// MemoryHeader: 할당된 메모리 앞에 붙여 할당 크기를 기록합니다.
//---------------------------------------------------------------------------
struct MemoryHeader {
    int allocSize;
};

inline void* AttachHeader(MemoryHeader* header, int allocSize) {
    header->allocSize = allocSize;
    // 헤더 바로 뒤의 영역을 사용자에게 반환
    return static_cast<void*>(header + 1);
}

inline MemoryHeader* DetachHeader(void* ptr) {
    return (static_cast<MemoryHeader*>(ptr)) - 1;
}

//---------------------------------------------------------------------------
// MemoryPool: 단순한 스레드 안전 메모리풀 구현 (_aligned_malloc/_aligned_free 사용)
//---------------------------------------------------------------------------
class MemoryPool {
public:
    // poolSize: 관리할 블록의 크기(헤더 포함X)
    MemoryPool(int poolSize) : _poolSize(poolSize) {}

    ~MemoryPool() {
        for (MemoryHeader* header : _pool)
            _aligned_free(header);
    }

    MemoryHeader* Pop() {
        lock_guard<mutex> lock(_mutex);
        if (!_pool.empty()) {
            MemoryHeader* header = _pool.back();
            _pool.pop_back();
            return header;
        }
        // 풀에 없으면 새로 할당
        MemoryHeader* header = reinterpret_cast<MemoryHeader*>(_aligned_malloc(_poolSize, 16));
        return header;
    }

    void Push(MemoryHeader* header) {
        lock_guard<mutex> lock(_mutex);
        _pool.push_back(header);
    }

private:
    int _poolSize;
    mutex _mutex;
    vector<MemoryHeader*> _pool;
};

//---------------------------------------------------------------------------
// Memory: 여러 MemoryPool을 관리하여 요청 크기에 맞게 할당합니다.
//---------------------------------------------------------------------------

class Memory {
public:
    enum { MAX_ALLOC_SIZE = 4096 }; // 최대 관리 크기

    Memory() {
        int currentKey = 0;
        // [1단계] 32 ~ 1024 바이트: 32 단위 증가
        for (int s = 32; s <= 1024; s += 32) {
            MemoryPool* pool = new MemoryPool(s);
            _pools.push_back(pool);
            while (currentKey <= s && currentKey <= MAX_ALLOC_SIZE) {
                _poolTable[currentKey] = pool;
                currentKey++;
            }
        }
        // [2단계] 1024 초과 ~ 2048: 128 단위 증가
        for (int s = 1024 + 128; s <= 2048; s += 128) {
            MemoryPool* pool = new MemoryPool(s);
            _pools.push_back(pool);
            while (currentKey <= s && currentKey <= MAX_ALLOC_SIZE) {
                _poolTable[currentKey] = pool;
                currentKey++;
            }
        }
        // [3단계] 2048 초과 ~ 4096: 256 단위 증가
        for (int s = 2048 + 256; s <= 4096; s += 256) {
            MemoryPool* pool = new MemoryPool(s);
            _pools.push_back(pool);
            while (currentKey <= s && currentKey <= MAX_ALLOC_SIZE) {
                _poolTable[currentKey] = pool;
                currentKey++;
            }
        }
    }

    ~Memory() {
        for (auto pool : _pools)
            delete pool;
        _pools.clear();
    }

    // request: 헤더 제외한 요청 크기
    void* Allocate(int request) {
        int allocSize = request + sizeof(MemoryHeader);
        MemoryHeader* header = nullptr;
        if (allocSize > MAX_ALLOC_SIZE) {
            header = reinterpret_cast<MemoryHeader*>(_aligned_malloc(allocSize, 16));
        }
        else {
            header = _poolTable[allocSize]->Pop();
        }
        return AttachHeader(header, allocSize);
    }

    void Release(void* ptr) {
        MemoryHeader* header = DetachHeader(ptr);
        int allocSize = header->allocSize;
        if (allocSize > MAX_ALLOC_SIZE) {
            _aligned_free(header);
        }
        else {
            _poolTable[allocSize]->Push(header);
        }
    }

private:
    vector<MemoryPool*> _pools;
    MemoryPool* _poolTable[MAX_ALLOC_SIZE + 1];
};

Memory* gMemory = nullptr;

//---------------------------------------------------------------------------
// 시간 측정 유틸리티 (마이크로초 단위)
//---------------------------------------------------------------------------
template <typename Func>
long long measureTime(Func f) {
    auto start = high_resolution_clock::now();
    f();
    auto end = high_resolution_clock::now();
    return duration_cast<microseconds>(end - start).count();
}

const int ITERATIONS = 100000;

//---------------------------------------------------------------------------
// [기본 테스트] 단일 스레드 테스트: custom allocator
//---------------------------------------------------------------------------
void testCustomAllocatorSingleThread(int size) {
    vector<void*> allocated;
    allocated.reserve(ITERATIONS);

    long long allocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            void* ptr = gMemory->Allocate(size);
            allocated.push_back(ptr);
        }
        });

    long long deallocTime = measureTime([&]() {
        for (auto ptr : allocated) {
            gMemory->Release(ptr);
        }
        });

    long long combinedTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            void* ptr = gMemory->Allocate(size);
            gMemory->Release(ptr);
        }
        });

    cout << "== Custom Allocator (Size " << size << ") Single-thread ==" << endl;
    cout << " Allocation: " << allocTime << " µs" << endl;
    cout << " Deallocation: " << deallocTime << " µs" << endl;
    cout << " Combined (alloc+dealloc): " << combinedTime << " µs" << endl;
    cout << " Total (alloc + dealloc): " << (allocTime + deallocTime) << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// [기본 테스트] 단일 스레드 테스트: std::vector 이용
//---------------------------------------------------------------------------
void testStdVectorSingleThread(int size) {
    vector<vector<char>*> allocated;
    allocated.reserve(ITERATIONS);

    long long allocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            vector<char>* vec = new vector<char>(size);
            allocated.push_back(vec);
        }
        });

    long long deallocTime = measureTime([&]() {
        for (auto vec : allocated) {
            delete vec;
        }
        });

    long long combinedTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            vector<char>* vec = new vector<char>(size);
            delete vec;
        }
        });

    cout << "== std::vector Allocator (Size " << size << ") Single-thread ==" << endl;
    cout << " Allocation: " << allocTime << " µs" << endl;
    cout << " Deallocation: " << deallocTime << " µs" << endl;
    cout << " Combined (alloc+dealloc): " << combinedTime << " µs" << endl;
    cout << " Total (alloc + dealloc): " << (allocTime + deallocTime) << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// [기본 테스트] 멀티 스레드 테스트용 결과 구조체 및 함수
//---------------------------------------------------------------------------
struct ThreadResult {
    long long allocTime;
    long long deallocTime;
    long long combinedTime;
};

void customAllocatorThread(int size, int iterations, ThreadResult& result) {
    vector<void*> allocated;
    allocated.reserve(iterations);

    result.allocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            void* ptr = gMemory->Allocate(size);
            allocated.push_back(ptr);
        }
        });

    result.deallocTime = measureTime([&]() {
        for (auto ptr : allocated)
            gMemory->Release(ptr);
        });

    result.combinedTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            void* ptr = gMemory->Allocate(size);
            gMemory->Release(ptr);
        }
        });
}

void stdVectorThread(int size, int iterations, ThreadResult& result) {
    vector<vector<char>*> allocated;
    allocated.reserve(iterations);

    result.allocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            vector<char>* vec = new vector<char>(size);
            allocated.push_back(vec);
        }
        });

    result.deallocTime = measureTime([&]() {
        for (auto vec : allocated)
            delete vec;
        });

    result.combinedTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            vector<char>* vec = new vector<char>(size);
            delete vec;
        }
        });
}

void testCustomAllocatorMultiThread(int size, int numThreads) {
    vector<thread> threads;
    vector<ThreadResult> results(numThreads);
    int iterationsPerThread = ITERATIONS / numThreads;

    auto start = high_resolution_clock::now();
    for (int i = 0; i < numThreads; i++) {
        threads.emplace_back(customAllocatorThread, size, iterationsPerThread, ref(results[i]));
    }
    for (auto& t : threads)
        t.join();
    auto end = high_resolution_clock::now();
    long long overallElapsed = duration_cast<microseconds>(end - start).count();

    long long totalAlloc = 0, totalDealloc = 0, totalCombined = 0;
    for (const auto& r : results) {
        totalAlloc += r.allocTime;
        totalDealloc += r.deallocTime;
        totalCombined += r.combinedTime;
    }

    cout << "== Custom Allocator (Size " << size << ") Multi-thread (" << numThreads << " threads) ==" << endl;
    cout << " Total Allocation (sum): " << totalAlloc << " µs" << endl;
    cout << " Total Deallocation (sum): " << totalDealloc << " µs" << endl;
    cout << " Total Combined (sum): " << totalCombined << " µs" << endl;
    cout << " Overall elapsed time: " << overallElapsed << " µs" << endl;
    cout << endl;
}

void testStdVectorMultiThread(int size, int numThreads) {
    vector<thread> threads;
    vector<ThreadResult> results(numThreads);
    int iterationsPerThread = ITERATIONS / numThreads;

    auto start = high_resolution_clock::now();
    for (int i = 0; i < numThreads; i++) {
        threads.emplace_back(stdVectorThread, size, iterationsPerThread, ref(results[i]));
    }
    for (auto& t : threads)
        t.join();
    auto end = high_resolution_clock::now();
    long long overallElapsed = duration_cast<microseconds>(end - start).count();

    long long totalAlloc = 0, totalDealloc = 0, totalCombined = 0;
    for (const auto& r : results) {
        totalAlloc += r.allocTime;
        totalDealloc += r.deallocTime;
        totalCombined += r.combinedTime;
    }

    cout << "== std::vector Allocator (Size " << size << ") Multi-thread (" << numThreads << " threads) ==" << endl;
    cout << " Total Allocation (sum): " << totalAlloc << " µs" << endl;
    cout << " Total Deallocation (sum): " << totalDealloc << " µs" << endl;
    cout << " Total Combined (sum): " << totalCombined << " µs" << endl;
    cout << " Overall elapsed time: " << overallElapsed << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// [추가 스트레스 테스트] 
// 1. 일부 해제 후 재할당: 단일 스레드
//---------------------------------------------------------------------------
void testStressCustomAllocatorSingleThread(int size) {
    vector<void*> allocated;
    allocated.reserve(ITERATIONS);

    long long initialAllocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            allocated.push_back(gMemory->Allocate(size));
        }
        });

    long long partialFreeTime = measureTime([&]() {
        // 전체의 약 1/3을 해제 (매 3번째)
        for (int i = 0; i < ITERATIONS; i += 3) {
            gMemory->Release(allocated[i]);
            allocated[i] = nullptr;
        }
        });

    long long reAllocTime = measureTime([&]() {
        // 해제된 슬롯에 대해 재할당
        for (int i = 0; i < ITERATIONS; i += 3) {
            allocated[i] = gMemory->Allocate(size);
        }
        });

    long long finalFreeTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            gMemory->Release(allocated[i]);
        }
        });

    cout << "== Custom Allocator Stress Test (Size " << size << ") Single-thread ==" << endl;
    cout << " Initial Allocation: " << initialAllocTime << " µs" << endl;
    cout << " Partial Free (every 3rd): " << partialFreeTime << " µs" << endl;
    cout << " Re-allocation: " << reAllocTime << " µs" << endl;
    cout << " Final Free: " << finalFreeTime << " µs" << endl;
    cout << " Total Stress Test Time: " << (initialAllocTime + partialFreeTime + reAllocTime + finalFreeTime) << " µs" << endl;
    cout << endl;
}

void testStressStdVectorSingleThread(int size) {
    vector<vector<char>*> allocated;
    allocated.reserve(ITERATIONS);

    long long initialAllocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            allocated.push_back(new vector<char>(size));
        }
        });

    long long partialFreeTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i += 3) {
            delete allocated[i];
            allocated[i] = nullptr;
        }
        });

    long long reAllocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i += 3) {
            allocated[i] = new vector<char>(size);
        }
        });

    long long finalFreeTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            delete allocated[i];
        }
        });

    cout << "== std::vector Stress Test (Size " << size << ") Single-thread ==" << endl;
    cout << " Initial Allocation: " << initialAllocTime << " µs" << endl;
    cout << " Partial Free (every 3rd): " << partialFreeTime << " µs" << endl;
    cout << " Re-allocation: " << reAllocTime << " µs" << endl;
    cout << " Final Free: " << finalFreeTime << " µs" << endl;
    cout << " Total Stress Test Time: " << (initialAllocTime + partialFreeTime + reAllocTime + finalFreeTime) << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// [추가 테스트] 혼합 크기 스트레스 테스트 (단일 스레드, custom allocator)
// 여러 크기를 랜덤하게 선택하여 할당/해제하여 단편화 현상을 관찰할 수 있습니다.
void testMixedSizeStressSingleThread() {
    vector<int> sizes = { 32, 64, 128, 256, 512, 1024, 2048, 4096 };
    vector<void*> allocated;
    allocated.reserve(ITERATIONS);

    long long initialAllocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            int idx = rand() % sizes.size();
            void* ptr = gMemory->Allocate(sizes[idx]);
            allocated.push_back(ptr);
        }
        });

    long long partialFreeTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            if (rand() % 100 < 30) { // 약 30% 해제
                gMemory->Release(allocated[i]);
                allocated[i] = nullptr;
            }
        }
        });

    long long reAllocTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            if (allocated[i] == nullptr) {
                int idx = rand() % sizes.size();
                allocated[i] = gMemory->Allocate(sizes[idx]);
            }
        }
        });

    long long finalFreeTime = measureTime([&]() {
        for (int i = 0; i < ITERATIONS; i++) {
            gMemory->Release(allocated[i]);
        }
        });

    long long totalTime = initialAllocTime + partialFreeTime + reAllocTime + finalFreeTime;
    cout << "== Custom Allocator Mixed Size Stress Test Single-thread ==" << endl;
    cout << " Initial Allocation: " << initialAllocTime << " µs" << endl;
    cout << " Partial Free: " << partialFreeTime << " µs" << endl;
    cout << " Re-allocation: " << reAllocTime << " µs" << endl;
    cout << " Final Free: " << finalFreeTime << " µs" << endl;
    cout << " Total Time: " << totalTime << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// [멀티 스레드 스트레스 테스트] 결과 구조체 및 함수
//---------------------------------------------------------------------------
struct ThreadStressResult {
    long long initialAllocTime;
    long long partialFreeTime;
    long long reAllocTime;
    long long finalFreeTime;
};

void customAllocatorStressThread(int size, int iterations, ThreadStressResult& result) {
    vector<void*> allocated;
    allocated.reserve(iterations);

    result.initialAllocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            allocated.push_back(gMemory->Allocate(size));
        }
        });

    result.partialFreeTime = measureTime([&]() {
        for (int i = 0; i < iterations; i += 3) {
            gMemory->Release(allocated[i]);
            allocated[i] = nullptr;
        }
        });

    result.reAllocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i += 3) {
            allocated[i] = gMemory->Allocate(size);
        }
        });

    result.finalFreeTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            gMemory->Release(allocated[i]);
        }
        });
}

void stdVectorStressThread(int size, int iterations, ThreadStressResult& result) {
    vector<vector<char>*> allocated;
    allocated.reserve(iterations);

    result.initialAllocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            allocated.push_back(new vector<char>(size));
        }
        });

    result.partialFreeTime = measureTime([&]() {
        for (int i = 0; i < iterations; i += 3) {
            delete allocated[i];
            allocated[i] = nullptr;
        }
        });

    result.reAllocTime = measureTime([&]() {
        for (int i = 0; i < iterations; i += 3) {
            allocated[i] = new vector<char>(size);
        }
        });

    result.finalFreeTime = measureTime([&]() {
        for (int i = 0; i < iterations; i++) {
            delete allocated[i];
        }
        });
}

void testCustomAllocatorStressMultiThread(int size, int numThreads) {
    vector<thread> threads;
    vector<ThreadStressResult> results(numThreads);
    int iterationsPerThread = ITERATIONS / numThreads;

    auto start = high_resolution_clock::now();
    for (int i = 0; i < numThreads; i++) {
        threads.emplace_back(customAllocatorStressThread, size, iterationsPerThread, ref(results[i]));
    }
    for (auto& t : threads)
        t.join();
    auto end = high_resolution_clock::now();
    long long overallElapsed = duration_cast<microseconds>(end - start).count();

    long long totalInitial = 0, totalPartial = 0, totalReAlloc = 0, totalFinal = 0;
    for (const auto& r : results) {
        totalInitial += r.initialAllocTime;
        totalPartial += r.partialFreeTime;
        totalReAlloc += r.reAllocTime;
        totalFinal += r.finalFreeTime;
    }

    cout << "== Custom Allocator Stress Test (Size " << size << ") Multi-thread (" << numThreads << " threads) ==" << endl;
    cout << " Total Initial Allocation: " << totalInitial << " µs" << endl;
    cout << " Total Partial Free: " << totalPartial << " µs" << endl;
    cout << " Total Re-allocation: " << totalReAlloc << " µs" << endl;
    cout << " Total Final Free: " << totalFinal << " µs" << endl;
    cout << " Overall elapsed time: " << overallElapsed << " µs" << endl;
    cout << endl;
}

void testStdVectorStressMultiThread(int size, int numThreads) {
    vector<thread> threads;
    vector<ThreadStressResult> results(numThreads);
    int iterationsPerThread = ITERATIONS / numThreads;

    auto start = high_resolution_clock::now();
    for (int i = 0; i < numThreads; i++) {
        threads.emplace_back(stdVectorStressThread, size, iterationsPerThread, ref(results[i]));
    }
    for (auto& t : threads)
        t.join();
    auto end = high_resolution_clock::now();
    long long overallElapsed = duration_cast<microseconds>(end - start).count();

    long long totalInitial = 0, totalPartial = 0, totalReAlloc = 0, totalFinal = 0;
    for (const auto& r : results) {
        totalInitial += r.initialAllocTime;
        totalPartial += r.partialFreeTime;
        totalReAlloc += r.reAllocTime;
        totalFinal += r.finalFreeTime;
    }

    cout << "== std::vector Stress Test (Size " << size << ") Multi-thread (" << numThreads << " threads) ==" << endl;
    cout << " Total Initial Allocation: " << totalInitial << " µs" << endl;
    cout << " Total Partial Free: " << totalPartial << " µs" << endl;
    cout << " Total Re-allocation: " << totalReAlloc << " µs" << endl;
    cout << " Total Final Free: " << totalFinal << " µs" << endl;
    cout << " Overall elapsed time: " << overallElapsed << " µs" << endl;
    cout << endl;
}

//---------------------------------------------------------------------------
// main: 전역 메모리 객체 초기화 및 각 테스트 실행
//---------------------------------------------------------------------------
int main() {
    srand((unsigned)time(NULL));

    // 전역 메모리 풀 생성
    gMemory = new Memory();

    vector<int> testSizes = { 32, 64, 128, 256, 512, 1024, 2048, 4096 };
    int numThreads = thread::hardware_concurrency();
    if (numThreads == 0)
        numThreads = 4;

    cout << "===== Single-thread Basic Tests =====" << endl;
    for (int size : testSizes) {
        testCustomAllocatorSingleThread(size);
        testStdVectorSingleThread(size);
        cout << "---------------------------------" << endl;
    }

    cout << "\n===== Multi-thread Basic Tests =====" << endl;
    for (int size : testSizes) {
        testCustomAllocatorMultiThread(size, numThreads);
        testStdVectorMultiThread(size, numThreads);
        cout << "---------------------------------" << endl;
    }

    cout << "\n===== Single-thread Stress Tests =====" << endl;
    for (int size : testSizes) {
        testStressCustomAllocatorSingleThread(size);
        testStressStdVectorSingleThread(size);
        cout << "---------------------------------" << endl;
    }

    cout << "\n===== Multi-thread Stress Tests =====" << endl;
    for (int size : testSizes) {
        testCustomAllocatorStressMultiThread(size, numThreads);
        testStdVectorStressMultiThread(size, numThreads);
        cout << "---------------------------------" << endl;
    }

    cout << "\n===== Mixed Size Stress Test (Custom Allocator) Single-thread =====" << endl;
    testMixedSizeStressSingleThread();

    delete gMemory;

    return 0;
}