This document discusses code optimization techniques in Java applications. It begins with an overview of code effectiveness and optimization, noting that optimization should not be done prematurely. It then covers various optimization techniques including JVM options, code samples, measurements using JMH of different techniques like method vs field access, strings, arrays, collections, and loops vs streams. It finds that techniques like using ArrayList/HashMap, compiler and JIT optimization, and measurement tools can improve performance. The document emphasizes measuring optimizations to determine real effectiveness.

1. Effectiveness and code optimization
in Java applications
Sergey Morenets
May, 21 2016

2. SPEAKER
FOUNDER
DEVELOPER 12 YEARS
TRAINER 4 YEARS
WRITER 3 BOOKS

4. Spring Data Vs JDBC

5. Preface

6. JDBC

9. Agenda

10. Agenda
What is effectiveness?
Code optimization
JVM optimization
Code samples
Measurements

11. Ideal code
Concise Readable
Self-describing Reusable
Testable Modern
Flexible Scalable
Effective

12. Effectiveness
Hard to determine on code/design review stages or in
unit-tests
Cannot be defined in development environment
Optimization is interactive
Premature optimization is evil
Hardware-specific
The only aspect of the ideal code that affects users

13. Optimization
Developer
Java compiler
JIT compiler
JVM

14. Premature optimization

15. Premature optimization

16. Premature optimization

17. Premature optimization

18. Premature optimization

19. Effectiveness
CPU usage
Memory
footprint
Network I/O
utilization
Disk I/O
utilization

20. Tuning
JVM options
Metaspace/heap/stack size
Garbage collector options
http://blog.sokolenko.me/2014/11/javavm-options-
production.html
http://www.javaspecialists.eu/

21. Why is it important?
Time-
consuming
processes
Big data
Real-time
systems

22. Code optimization

23. Code optimization
public int execute();
Code:
0: iconst_2
1: istore_1
2: iinc 1, 1
5: iconst_1
6: ireturn

24. Code optimization

25. Code optimization
public void execute();
Code:
0: return

26. Code optimization

27. Code optimization
public static boolean get();
Code:
0: iconst_1
1: ireturn

28. Code optimization

29. Code optimization
public void execute();
Code:
0: return

30. Code optimization

31. Code optimization
public int execute();
Code:
0: iconst_2
1: istore_1
2: iconst_4
3: istore_2
4: iload_1
5: iload_2
6: iadd
7: ireturn

32. Code optimization

33. Code optimization
public int execute();
Code:
0: bipush 6
2: ireturn

34. Measurements
JMH is micro benchmarking framework
Developed by Oracle engineers
First release in 2013
Requires build tool(Maven, Gradle)
Can measure throughput or average time
Includes warm-up period
Part of Java 9

35. Measurements

36. Warm-up

37. Environment
JMH 1.12
Maven 3.3.9
JDK 1.8.0.91
Intel Core i7, 4 cores, 16 GB

38. Method vs Field

39. Method vs Field
1,95 ns
1,93 ns

40. Strings

41. Strings
7,62 ns
13,4 ns
7,32 ns
7,24 ns

42. Conditions

43. Conditions
2,05 ns
2,03 ns

44. Strings

45. Strings

46. Measurements
Type(characters) Time(ns)
Multiple(3) 627
Single(3) 284
Multiple(30) 773
Single(30) 387

47. Strings

48. Strings

49. Measurements
Type(characters) Time(ns)
Multiple(3) 634
Single(3) 283
Pattern(3) 151

50. Arrays

51. Measurements
Type Time(ns)
For (10 elements) 4,9
For-each (10 elements) 5,1
For (1000 elements) 260
For-each (1000 elements) 259,9
For (50000 elements) 12957
For-each (50000 elements) 12958

52. Arrays

53. Measurements
Type Time(ns)
Sequential (10 elements) 5
Parallel (10 elements) 6230
Sequential (1000 elements) 263
Parallel (1000 elements) 8688
Sequential (50000 elements) 13115
Parallel (50000 elements) 34695

54. Measurements
Type Time(ns)
Sequential (10 elements) 5
Parallel (10 elements) 6230
Sequential (1000 elements) 263
Parallel (1000 elements) 8688
Sequential (50000 elements) 13115
Parallel (50000 elements) 34695
Sequential (5 000 000 elements) 1 765 206
Parallel (5 000 000 elements) 2 668 564
Sequential (500 000 000) 183 ms
Parallel (500 000 000) 174 ms

55. Autoboxing

56. Measurements
Type Time(ns)
Array(100 elements) 58
List(100 elements) 390
Array(10 000 elements) 4776
List(10 000 elements) 48449

57. Collections

58. Collections
Type Time(ns)
Fill HashMap(1000 elements) 16000
Fill TreeMap(1000 elements) 40115
Fill HashMap(100 000 elements) 2 027 116
Fill TreeMap(100 000 elements) 11 195 422
Iteration HashMap(1000 elements) 3086
Iteration TreeMap(1000 elements) 5038

59. Sample

61. Lists

62. Measurements
Type Time(ns)
ArrayList (1000 elements) 4766
ArrayList (100 000 elements) 381707
LinkedList (1000 elements) 5504
LinkedList (100 000 elements) 504231

63. Lists

64. Measurements
Type Time(ns)
ArrayList (1000 elements) 26767
ArrayList (100 000 elements) 276(ms)
LinkedList (1000 elements) 300971
LinkedList (100 000 elements) 3424(ms)

65. Lists

66. Measurements
Type Time(ns)
ArrayList (1000 elements) 774
ArrayList (100 000 elements) 144814
LinkedList (1000 elements) 2161
LinkedList (100 000 elements) 292364

67. Comparison
Operations ArrayList LinkedList
Add
Delete
Get
Iterate

68. Comparison

69. Comparison

70. Measurements
Type Time(ns)
ArrayList (1000 elements) 49710
ArrayList (100 000 elements) 632 754 873
LinkedList (1000 elements) 8507
LinkedList (100 000 elements) 874 483

71. Speed
Memory
footprint
Big data
structures
I/O
support

72. Measurements
Type (elements) Time(ns)
ArrayList (1000) 4732
ArrayList (100 000) 387692
LinkedList (1000) 5775
LinkedList (100 000) 511646
ObjectArrayList(1000) 3168
ObjectArrayList(100 000) 322811

73. Lightweight
Hashing
strategies
Primitive
types
Hash table
improvements

74. Measurements
Type (elements) Time(ns)
ArrayList (1000) 4595
ArrayList (100 000) 394645
ObjectArrayList(1000) 328363
ObjectArrayList(100 000) 3187
TIntArrayList(1000) 189608
TIntArrayList(100 000) 1925

75. LIFO

76. LIFO
LinkedList
Stack
ArrayDeque

77. Stack

78. Measurements
Type Time(ns)
Stack(1000 elements) 5729
LinkedList(1000 elements) 7928
ArrayDeque(1000 elements) 4366
Stack(100 000 elements) 588770
LinkedList(100 000 elements) 773684
ArrayDeque(100 000 elements) 484280

79. Mapping

80. Mapping

81. Mapping

82. Measurements
Type Time(ns)
HashMap 4,60
EnumMap 3,40

83. Mapping

84. Measurements
Type Time(ns)
HashMap 4,53
EnumMap 3,40
Direct usage 2,36

85. Loops vs Streams

86. Loops vs Streams

87. Measurements
Type 10 elements(ns) 1000 elements 100000 elements
For-each 85 8002 766765
Stream 147 11784 1103383

88. Conclusion
Compiler and JIT optimization
Speed and memory optimization
Prefer ArrayList/HashMap
Use measurement tools

89. Theory

93. Enterprise
Web server performance
Database programming
ORM & DI frameworks
REST services
Build management systems

94. Q&A
• Sergey Morenets, sergey.morenets@gmail.com