Minborg's Java Pot

constant folding of instance fields · Chart 1, Shows the frequency of various name byte sizes.

Hamburg;12.0
Bulawayo;8.9
Palembang;38.8
St. John's;15.2
Cracow;12.6
Bridgetown;26.9
Istanbul;6.2
Roseau;34.4
Conakry;31.2
Istanbul;23.0
public interface Point {
    int x();
    int y();
}
public final class RegularPoint implements Point {

    private final int x;
    private final int y;

    public RegularPoint(int x, int y) {
        this.x = x;
        this.y = y;
    }

    @Override
    public int x() {
        return x;
    }

    @Override
    public int y() {
        return y;
    }

    // Implementations of toString(), hashCode() and equals()
    // omitted for brevity

}
public record RecordPoint(int x, int y) implements Point {}
public static final Point ORIGIN = new RecordPoint(0, 0);
public static boolean isOrigin(Point point) {
        return point.x() == ORIGIN.x() &&
               point.y() == ORIGIN.y();
}
public class Demo {

    public static final Point ORIGIN = new RecordPoint(0, 0);

    public static void main(String[] args) {
        analyze(new RegularPoint(0, 0));
        analyze(new RegularPoint(1, 1));
    }

    public static void analyze(Point point) {
        System.out.format("The point %s is %s at the origin.%n",
                point, isOrigin(point) ? "" : "not");
    }

    public static boolean isOrigin(Point point) {
        return point.x() == ORIGIN.x() &&
               point.y() == ORIGIN.y();
    }

}
The point RegularPoint{x=0, y=0} is at the origin.
The point RegularPoint{x=1, y=1} is not at the origin.
The point RecordPoint[x=0, y=0] is at the origin.
The point RecordPoint[x=1, y=1] is not at the origin.
@BenchmarkMode(Mode.AverageTime)
@OutputTimeUnit(TimeUnit.NANOSECONDS)
@State(Scope.Benchmark)
@Warmup(iterations = 5, time = 1)
@Measurement(iterations = 5, time = 1)
@Fork(value=3)
public class Bench {

    private static final RegularPoint REGULAR_ORIGIN = new RegularPoint(0, 0);
    private static final RecordPoint RECORD_ORIGIN = new RecordPoint(0, 0);

    private List<RegularPoint> regularPoints;
    private List<RecordPoint> recordPoints;

    @Setup
    public void setup() {
        regularPoints = IntStream.range(0, 16)
                .mapToObj(i -> new RegularPoint(i, i))
                .toList();

        recordPoints = IntStream.range(0, 16)
                .mapToObj(i -> new RecordPoint(i, i))
                .toList();
    }

    @Benchmark
    public void regular(Blackhole bh) {
        for (RegularPoint point: regularPoints) {
            if (point.x() == REGULAR_ORIGIN.x() && point.y() == REGULAR_ORIGIN.y()) {
                bh.consume(1);
            } else {
                bh.consume(0);
            }
        }
    }

    @Benchmark
    public void record(Blackhole bh) {
        for (RecordPoint point: recordPoints) {
            if (point.x() == RECORD_ORIGIN.x() && point.y() == RECORD_ORIGIN.y()) {
                bh.consume(1);
            } else {
                bh.consume(0);
            }
        }
    }

    public static void main(String[] args) throws Exception {
        org.openjdk.jmh.Main.main(args);
    }

}
Benchmark      Mode  Cnt   Score   Error  Units
Bench.regular  avgt   15  10.424 ± 0.257  ns/op
Bench.record   avgt   15   9.412 ± 0.181  ns/op
Benchmark                           (size)  Mode  Cnt    Score   Error  Units
ToJavaStringTest.jni_readString          5  avgt   30   86.520 ? 1.842  ns/op
ToJavaStringTest.jni_readString         20  avgt   30   97.151 ? 1.459  ns/op
ToJavaStringTest.jni_readString        100  avgt   30  143.853 ? 1.287  ns/op
ToJavaStringTest.jni_readString        200  avgt   30  189.867 ? 2.337  ns/op
ToJavaStringTest.panama_readString       5  avgt   30   21.380 ? 0.351  ns/op
ToJavaStringTest.panama_readString      20  avgt   30   36.250 ? 0.520  ns/op
ToJavaStringTest.panama_readString     100  avgt   30   43.368 ? 0.544  ns/op
ToJavaStringTest.panama_readString     200  avgt   30   53.442 ? 2.048  ns/op

Benchmark                         (size)  Mode  Cnt    Score   Error  Units
ToCStringTest.jni_writeString          5  avgt   30   47.450 ? 0.832  ns/op
ToCStringTest.jni_writeString         20  avgt   30   56.208 ? 0.422  ns/op
ToCStringTest.jni_writeString        100  avgt   30  108.341 ? 0.459  ns/op
ToCStringTest.jni_writeString        200  avgt   30  157.119 ? 1.669  ns/op
ToCStringTest.panama_writeString       5  avgt   30   45.361 ? 0.717  ns/op
ToCStringTest.panama_writeString      20  avgt   30   47.742 ? 0.554  ns/op
ToCStringTest.panama_writeString     100  avgt   30   47.580 ? 0.673  ns/op
ToCStringTest.panama_writeString     200  avgt   30   49.060 ? 0.694  ns/op
// ordinary static initialization
private static final Logger LOGGER = Logger.getLogger("com.foo.Bar");
...
LOGGER.log(...);
// Initialization-on-demand holder idiom
Logger logger() {
    class Holder {
         static final Logger LOGGER = Logger.getLogger("com.foo.Bar");
    }
    return Holder.LOGGER;
}
...
logger().log(...);
// Double-checked locking idiom
class Foo {

    private volatile Logger logger;

    public Logger logger() {
        Logger v = logger;
        if (v == null) {
            synchronized (this) {
                v = logger;
                if (v == null) {
                    logger = v = Logger.getLogger("com.foo.Bar");
                }
            }
        }
        return v;
    }
}
...
foo.logger().log(...);
class Bar {
    // 1. Declare a computed constant value
    private static final ComputedConstant<Logger> LOGGER =
            ComputedConstant.of( () -> Logger.getLogger("com.foo.Bar") );

    static Logger logger() {
        // 2. Access the computed value
        //    (evaluation made before the first access)
        return LOGGER.get();
    }
}
Benchmark      Mode  Cnt  Score   Error  Units
staticHolder   avgt   15  0.561 ? 0.002  ns/op
doubleChecked  avgt   15  1.122 ? 0.003  ns/op
constant       avgt   15  0.563 ? 0.002  ns/op // static ComputedConstant
class Fibonacci {
    static final List<ComputedConstant<Integer>> FIBONACCI =
            ComputedConstant.of(1_000, Fibonacci::fib);

    static int fib(int n) {
        return (n < 2)
                ? n
                : FIBONACCI.get(n - 1) + FIBONACCI.get(n - 2);
    }

    int[] fibs = IntStream.range(0, 10)
            .map(Fibonacci::fib)
            .toArray(); // { 0, 1, 1, 2, 3, 5, 8, 13, 21, 34 }

}
               ___________fib(5)___________
              /                            \
        ____fib(4)____                ____fib(3)____
       /              \              /              \
     fib(3)         fib(2)         fib(2)          fib(1)
    /      \       /      \       /      \
  fib(2)  fib(1) fib(1)  fib(0) fib(1)  fib(0)
Benchmark      Mode  Cnt  Score   Error  Units
staticHolder   avgt   15  0.570 ? 0.005  ns/op // int[] in a holder class
doubleChecked  avgt   15  1.124 ? 0.044  ns/op
constant       avgt   15  0.562 ? 0.005  ns/op // List<ComputedConstant>
Benchmark      Mode  Cnt  Score   Error  Units
doubleChecked  avgt   15  1.259 ? 0.023  ns/op
constant       avgt   15  0.728 ? 0.022  ns/op // ComputedConstant
public int readInt() throws IOException {
    if (read(byteBuf, 0, 4) !=  4) {
        throw new EOFException();
    }

    if (byteOrder == ByteOrder.BIG_ENDIAN) {
        return
            (((byteBuf[0] & 0xff) << 24) | ((byteBuf[1] & 0xff) << 16) |      // (1)
             ((byteBuf[2] & 0xff) <<  8) | ((byteBuf[3] & 0xff) <<  0));
    } else {
        return
            (((byteBuf[3] & 0xff) << 24) | ((byteBuf[2] & 0xff) << 16) |      // (2)
             ((byteBuf[1] & 0xff) <<  8) | ((byteBuf[0] & 0xff) <<  0));
    }
}
private static final VarHandle INT =
        MethodHandles.byteArrayViewVarHandle(int[], ByteOrder.BIG_ENDIAN);

static int getInt(byte[] b, int off) {
    return (int) INT.get(b, off);
}
public int readInt() throws IOException {
    if (read(byteBuf, 0, 4) !=  4) {
        throw new EOFException();
    }

    return (byteOrder == ByteOrder.BIG_ENDIAN)
            ? ByteArray.getInt(byteBuf, 0)
            : ByteArrayLittleEndian.getInt(byteBuf, 0);
}
jpaStreamer.stream(Film.class)
    .sorted(Film$.length.reversed())
    .limit(15)
    .map(Film$.title)
    .forEach(System.out::println);

    <dependency>
        <groupId>com.speedment.jpastreamer</groupId>
        <artifactId>jpastreamer-core</artifactId>
        <version>0.1.8</version>
    </dependency>

$ docker pull restsql/mysql-sakila
$ docker run -d --publish 3306:3306 --name mysqld restsql/mysql-sakila

@Entity
@Table(name = "film", schema = "sakila")
public class Film implements Serializable {

    @Id

    @GeneratedValue(strategy = GenerationType.IDENTITY)
    @Column(name = "film_id", nullable = false, updatable = false, columnDefinition = "smallint(5)")
    private Integer filmId;

    @Column(name = "title", nullable = false, columnDefinition = "varchar(255)")
    private String title;

    @Column(name = "description", nullable = false, columnDefinition = "text")
    private String description;

    @ManyToMany(cascade = CascadeType.ALL)
    @JoinTable(
        name = "film_actor",
        joinColumns = @JoinColumn(name = "film_id") ,
        inverseJoinColumns = @JoinColumn(name = "actor_id") 
    )
    private List<Artist> actors;
    ...
}

public class LongestFilms  {

    public static void main(String[] args) {

        final JPAStreamer jpaStreamer = JPAStreamer.of("sakila");

        jpaStreamer.stream(Film.class)
            .sorted(Film$.length.reversed())
            .limit(15)
            .map(f -> String.format("%3d %s", f.getLength(), f.getTitle()))
            .forEach(System.out::println);

        jpaStreamer.close();
    }
 }

185 SOLDIERS EVOLUTION
185 GANGS PRIDE
185 SWEET BROTHERHOOD
185 CHICAGO NORTH
185 HOME PITY
185 POND SEATTLE
185 CONTROL ANTHEM
185 DARN FORRESTER
185 WORST BANGER
184 SMOOCHY CONTROL
184 SONS INTERVIEW
184 SORORITY QUEEN
184 MOONWALKER FOOL
184 THEORY MERMAID

select
    film0_.film_id as film_id1_1_,
    film0_.length as length4_1_,
    film0_.title as title10_1_,
    /* more columns */
from
    film film0_ 
order by
    film0_.length desc limit ?

StreamConfiguration configuration = StreamConfiguration.of(Film.class)
    .joining(Film$.actors)
    .joining(Film$.language);

jpaStreamer.stream(configuration) 
    .filter(Film$.rating.in("G", "PG"))
    .forEach(System.out::println);

select
    Film 
from 
    Film as Film 
left join 
    fetch Film.actors as generatedAlias0 
left join 
    fetch Film.language as GeneratedAlias1
where 
    Film.rating in (
        :param0, :param1
    )

jpaStreamer.stream(User.class)
    .filter(User$.firstName.startsWith(”A”)
    .sort(User$.lastName.reversed())
    .limit(10)
    .forEach(System.out::println);

JPAStreamer jpaStreamer = JPAStreamer.of("db-name"); 

jpaStreamer.stream(User.class); 

List<String> users = jpaStreamer.stream(User.class)
    .filter(User$.age.greaterThan(20))
    .map(u -> u.getFirstName() + ” ” + u.getLastName())
    .collect(Collectors.toList); 
long count = jpaStreamer.stream(User.class)
    .filter(User$.country.equal(”Germany”).and(User$.firstName.equal(”Otto”))
    .count(); 

[
   {
      "actorId": 1,
      "firstName": "PENELOPE",
      "lastName": "GUINESS"
   },
   {
      "actorId": 2,
      "firstName": "NICK",
      "lastName": "WAHLBERG"
   },
   {
      "actorId": 3,
      "firstName": "ED",
      "lastName": "CHASE"
   },
... TRUNCATED ... 
]

FROM openjdk:11-slim-buster

EXPOSE 8080

ARG JAR_LOCATION=target
ARG JAR_NAME=speedment-spring-app
ARG JAR_VERSION=1.0.0

ADD ${JAR_LOCATION}/${JAR_NAME}-${JAR_VERSION}.jar app.jar

ENTRYPOINT ["java", "-jar", "app.jar", "--spring.speedment.host=sakila"]

apiVersion: apps/v1
kind: Deployment
metadata:
  name: sakila
  labels:
    storage: sakila
spec:
  replicas: 1
  selector:
    matchLabels:
      storage: sakila
  template:
    metadata:
      labels:
        storage: sakila
    spec:
      containers:
      - name: sakila
        image: restsql/mysql-sakila
        ports:
        - containerPort: 3306

apiVersion: v1
kind: Service
metadata:
  name: sakila
  labels:
    storage: sakila
spec:
  selector:
    storage: sakila
  ports:
  - name: database
    port: 3306
    targetPort: 3306

apiVersion: apps/v1
kind: Deployment
metadata:
  name: speedment-spring-app
  labels:
    app: speedment-spring-app
spec:
  replicas: 1
  selector:
    matchLabels:
      app: speedment-spring-app
  template:
    metadata:
      labels:
        app: speedment-spring-app
    spec:
      containers:
      - name: speedment-spring-app
        image: speedment-spring-app-example
        ports:
        - containerPort: 8080

apiVersion: v1
kind: Service

metadata:
  name: speedment-spring-app
spec:
  selector:
    app: speedment-spring-app
  ports:
  - name: http
    port: 8080
    targetPort: 8080
  type: NodePort

apiVersion: skaffold/v2alpha3
kind: Config
build:
  artifacts:
  - image: speedment-spring-app-example
    docker:
      dockerfile: Dockerfile
deploy:
  kubectl:
    manifests:
    - k8s/app/*
    - k8s/storage/*

Value	Char	Frequency
0..32		0
32	[space]	181605826
33		0
34		0
35		0
36		0
37		0
38		0
39	'	12104970
40	(	2422734
41	)	2422734
42	*	0
43	+	0
44	,	4845007
45	-	14533546
46	.	9686377
47	/	0
48	0	0
49	1	0
50	2	0
51	3	0
52	4	0
53	5	0
54	6	0
55	7	0
56	8	0
57	9	0
58		0
59		0
60		0
61		0
62		0
63		0
64		0
65	A	82,312,102
66	B	99,268,251
67	C	94,430,447
68	D	70,215,892
69	E	14,526,993
70	F	16,957,547
71	G	33,899,137
72	H	50,855,379
73	I	12,102,959
74	J	26,634,842
75	K	53,262,607
76	L	72,646,247
77	M	99,292,126
78	N	55,695,661
79	O	31,475,791
80	P	79,892,902
81	Q	0
82	R	29,051,519
83	S	99,282,847
84	T	72,636,263
85	U	4,842,032
86	V	33,897,984
87	W	21,796,354
88	X	2,421,477
89	Y	19,373,396
90	Z	7,262,683
91	[	0
92	[back slash]	0
93	]	0
94	^	0
95	_	0
96	`	0
97	a	1,125,887,562
98	b	147,711,179
99	c	106,528,787
100	d	130,753,955
101	e	438,247,893
102	f	16,946,953
103	g	191,298,907
104	h	200,950,920
105	i	547,230,962
106	j	38,740,836
107	k	157,372,493
108	l	314,764,906
109	m	154,957,248
110	n	559,330,103
111	o	535,145,324
112	p	77,480,229
113	q	12,101,467
114	r	411,634,242
115	s	336,573,182
116	t	317,179,382
117	u	334,155,534
118	v	72,645,180
119	w	43,579,677
120	x	19,362,397
121	y	92,009,532
122	z	43,584,786
123	{	0
124	[bar]	0
125	}	0
126	~	0
127	[del]	0
128		0
129		0
130		4842294
131		2422260
132..152		0
153		2422260
154		0
155		0
156		2420126
157		0
158		0
159		0
160		0
161		0
162		0
163		0
164		0
165		0
166		0
167		0
168		4843781
169		29054839
170		0
171		0
172		0
173		2422053
174		0
175		0
176		2421256
177		0
178		0
179		0
180		0
181		0
182		0
183		0
184		2421983
185		0
186		0
187		0
188		4840135
189		0
190		0
191		0
192		0
193		0
194		0
195		46002917
196		4843516
197		4842294
198		0
199		0
200		2422260
201…256		0

Char	Frequency
a	1,208,199,664
b	246,979,430
c	200,959,234
d	200,969,847
e	452,774,886
f	33,904,500
g	225,198,044
h	251,806,299
i	559,333,921
j	65,375,678
k	210,635,100
l	387,411,153
m	254,249,374
n	615,025,764
o	566,621,115
p	157,373,131
q	12,101,467
r	440,685,761
s	435,856,029
t	389,815,645
u	338,997,566
v	106,543,164
w	65,376,031
x	21,783,874
y	111,382,928
z	50,847,469

s = size(S)	poc(s) (permutations of combinations)
0	1
1	2
2	5
3	16
4	65
5	326
6	1,957
7	13,700
8	109,601
9	986,410
10	986,4101
11	108,505,112

Legend (JVM Variant)	Detail
OpenJDK 8	1.8.0_322, vendor: Temurin
OpenJDK 11	11.0.14.1, vendor: Eclipse Adoptium
OpenJDK 17	17.0.2, vendor: Eclipse Adoptium
Graal VM CE 17	17.0.2, vendor: GraalVM Community
Graal VM EE 17	17.0.2, vendor: Oracle Corporation

Percentile	# Messages
0% (all)	10,000,000
50% (“Median”, used below)	5,000,000
99%	100,000
99.9%	10,000
99.99% (used below)	1,000
99.999%	100

JDK Variant	Median	99.99%
OpenJDK 8	280	3,951
OpenJDK 11	370	4,210
OpenJDK 17	290	4,041
GraalVM CE 17 (*)	310	3,950
GraalVM EE 17 (*)	270	3,800

Instances	Median Latency	99%	99.9%	CPU Utilisation
1	0.9	19	30	670%
2	16	72	106 (*)	700% (saturated)

Operation	HashMap	IntMapper
put/add	O(1) < x < O(log(N)) (*)	O(1) (**)
sort	-	O(log(N))
get	O(1) < x < O(log(N)) (*)	O(log(N))

Size	Frequency
3	2
4	18
5	42
6	87
7	66
8	57
9	47
10	26
11	26
12	17
13	9
14	4
15	2
16	6
17	1
18	1
24	1
26	1

Operation	HashMap	IntMapper
put/add	48N (***)	0 (***)
get	16N (***)	0

Minborg's Java Pot

Statistics for the 1BRC Java Challenge

Introduction

Name Statistics

The names are UTF-8 coded and can occupy up 100 bytes of space.Looking at the file, one can conclude that there are 413 distinct names. So, the number of stations is actually relatively limited. Holding these in hash tables or the like is not anticipated to take up a significant amount of memory.

Length

The corresponding cumulative frequency chart looks like this:Chart 2, Shows the cumulative frequency of various name byte sizes.As can be seen, considering only the first 16 bytes will cover more than 99% of the names in the file.

Appearance

The number of times a station appears in the file is very similar ranging from "Roseau" with 2,417,155 to "Chihuahua" with 2,425,899 appearances.

Characters

Chart 3, Shows a character frequency table.

Temperature Statistics

The temperatures appear in the interval [-67.3, 84.9] and have the following distribution:

Chart 4, shows the temperature distribution.I hope this can spark some ideas in the community. Good luck with your challenge entry!

My Presentations at Devoxx BE 2023

Presentations at Devoxx BE October 2023

Java Records are "Trusted" and Consequently Faster

Java Records are "Trusted" and Consequently Faster

Background

Setup

Benchmarks

Under the Hood

Conclusion

Java 22: Panama FFM Provides Massive Performance Improvements for Native Strings

Java 22: Panama FFM Provides Massive Performance Improvements for Native Strings

C and Java Strings

Benchmarks

Diagrams

Future Improvements

JDK Early-Access Builds

Resources

Acknowledgments

Java: New Draft JEP: "Computed Constants"

Java: JEP Draft: "Computed Constants"

Background

Computed Constant

Benchmarks

Collections of ComputedConstant

Benchmarks Collections

Instance Performance

Where is it?

Resources

Acknowledgments

Video: Project Panama and the Foreign Function and Memory API

JDK 21: Image Performance Improvements

Introduction

Background

Improvements in JDK 21

Affected Classes and Impact

Benchmarks

Actual Application Performance Increase

Java 21: Performance Improvements Revealed

Background

Improvements in Java 21

Affected Classes and Impact

Raw Benchmarks

Serialization Benchmarks

Future Improvements

Actual Application Performance Increase

Java 20: An Almost Infinite Memory Segment Allocator

Arena

The InfiniteArena Class

Using the InfiniteArena

Inspecting the Files

Drawbacks and Future Improvements

What’s Next?

Java 20: Colossal Sparse Memory Segments

Mapped Memory

Setting up a Mapped Memory Segment

Sparse Files

Setting up a Sparsely Mapped Memory Segment

Going Colossal

Unrelated Observations on Thread Confinement

What’s Next?

Java 20: A Sneak Peek on the Panama FFM API (Second Preview)

Getting familiar with the FFM

Short Summary

MemorySegment

SegmentScope

Perpetual Global Scope Allocation

The names are UTF-8 coded and can occupy up 100 bytes of space.
Looking at the file, one can conclude that there are 413 distinct names. So, the number of stations is actually relatively limited. Holding these in hash tables or the like is not anticipated to take up a significant amount of memory.

The corresponding cumulative frequency chart looks like this:
Chart 2, Shows the cumulative frequency of various name byte sizes.
As can be seen, considering only the first 16 bytes will cover more than 99% of the names in the file.

Chart 4, shows the temperature distribution.
I hope this can spark some ideas in the community. Good luck with your challenge entry!

Conclusion
About four hundred times more applications can be run on the same hardware if a switch is made from Kafka to Chronicle Queue for specific latency-sensitive EDA applications.

Chart 2, Shows normalised cost vs queue type (less is better)

Resources