fastjson_benchmark

1. 测试环境

1.1 各个库的版本

fastjson2 2.0.12
fastjson1 1.0.83
jackson 2.13.3
gson 2.9.1
kryo 5.3.0
hessian 4.0.66

1.2 服务器环境

测试的服务器基于阿里云最新代ECS，包括x64架构的Intel和AMD处理器，ARM架构的AltraMax和阿里云平头哥的倚天处理器。

cs.c7a.xlarge
ecs.c7a.xlarge
ecs.c6r.xlarge
ecs.g8m.xlarge

ecs.c7.xlarge

这个是阿里云当前代标准型ECS，处理器型号 Intel Xeon(Ice Lake) Platinum 8369B，4核，8G内存

ecs.c7a.xlarge

这个是阿里云当前代标准型ECS，处理器型号 AMD EPYC™ Milan 7T83，4核，8G内存

ecs.c6r.xlarge

这个是阿里云上售卖ARM处理器ECS，处理器型号 Ampere Altra / AltraMax，4核，8G内存

ecs.g8m.xlarge

这个是阿里云上售卖ARM处理器，处理器型号 Yitian 710，4核，16G内存。这个是阿里云平头哥的倚天710处理器，需要联系客服才能购买。

1.3 JDK版本

基于Oracle最新版本的Linux x64/aarch64的JDK版本，下载地址 https://www.oracle.com/java/technologies/

oracle-jdk1.8.0_341
oracle-jdk-11.0.16
oracle-jdk-17.0.4
oracle-jdk-18.0.2

1.4 测试代码以及运行方式

代码路径 https://github.com/alibaba/fastjson2/tree/2.0.12/benchmark/src/main/java/com/alibaba/fastjson2/benchmark/eishay
运行方式测试代码方式

git clone https://github.com/alibaba/fastjson2
cd fastjson2
git checkout 2.0.12
mvn clean install -Dmaven.test.skip
~/Install/jdk-1.8.0_341/bin/java -cp ~/git/fastjson2/benchmark/target/fastjson2-benchmarks.jar com.alibaba.fastjson2.benchmark.eishay.Eishay
~/Install/jdk-11.0.16/bin/java -cp ~/git/fastjson2/benchmark/target/fastjson2-benchmarks.jar com.alibaba.fastjson2.benchmark.eishay.Eishay
~/Install/jdk-17.0.4/bin/java -cp ~/git/fastjson2/benchmark/target/fastjson2-benchmarks.jar com.alibaba.fastjson2.benchmark.eishay.Eishay
~/Install/jdk-18.0.2/bin/java -cp ~/git/fastjson2/benchmark/target/fastjson2-benchmarks.jar com.alibaba.fastjson2.benchmark.eishay.Eishay

2. 测试数据&结果分析

原始数据 https://github.com/alibaba/fastjson2/blob/2.0.12/docs/benchmark/benchmark_2.0.12_raw.md

2.1 EishayParseString

这个是最常用的场景，将JSON格式字符串反序列化为Java对象，这个场景在fastjson中的代码如下：

String str = "...";
Bean bean = JSON.parseJSONObject(str, Bean.class);

这个场景是fastjson1最强的场景；

在ecs.c7.xlarge-oracle-jdk1.8.0_341_x64环境中，只相当于只相当于fastjson2的76.16%;jackson则只相当于fastjson2的41.62%; gson是fastjson2的34.44%.

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > fastjson1 > jackson > gson
100%        76.16%      41.62%    34.44%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	1280.79	975.417 (76.16%)	533.05 (41.62%)	441.098 (34.44%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	1156.746	923.166 (79.81%)	484.755 (41.91%)	432.261 (37.37%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1271.058	1245.138 (97.96%)	488.057 (38.4%)	458.173 (36.05%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1285.745	1162.7 (90.43%)	497.201 (38.67%)	416.679 (32.41%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	1232.166	997.019 (80.92%)	516.422 (41.91%)	454.289 (36.87%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	1074.775	928.247 (86.37%)	496.175 (46.17%)	468.162 (43.56%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	1237.042	1323.952 (107.03%)	529.334 (42.79%)	536.419 (43.36%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	1226.619	1192.121 (97.19%)	481.629 (39.26%)	467.084 (38.08%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	843.629	719.256 (85.26%)	383.128 (45.41%)	313.66 (37.18%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	828.013	734.647 (88.72%)	377.617 (45.61%)	318.57 (38.47%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	853.772	1020.291 (119.5%)	379.463 (44.45%)	318.711 (37.33%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	849.58	995.946 (117.23%)	362.179 (42.63%)	303.018 (35.67%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	917.802	737.61 (80.37%)	370.102 (40.32%)	353.573 (38.52%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	1026.274	802.58 (78.2%)	408.008 (39.76%)	376.903 (36.73%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	1030.558	1026.0 (99.56%)	392.415 (38.08%)	244.263 (23.7%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	1089.054	999.211 (91.75%)	396.401 (36.4%)	268.559 (24.66%)

2.2 EishayParseStringPretty

这个场景是将格式化过的JSON格式字符串反序列化为Java对象，这个场景fastjson1的parse算法不擅长的。这个场景fastjson2表现仍然非常好。这个场景在ecs.c7.xlarge-oracle-jdk1.8.0_341_x64环境下，fastjson1的性能是fastjson2的27.29%，jackson的性能是fastjson2的50.56%，gson是fastjson2的43.01%.

这个场景在fastjson中的代码如下：

// 这里输入的是格式化过后的json字符串
String str = "{\n" +
        "\t\"id\":123\n" +
        "}";
Bean bean = JSON.parseJSONObject(str, Bean.class);

性能排序分别如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > gson  > fastjson1
100%        50.56%    43.01%  27.29%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	963.195	262.851 (27.29%)	486.976 (50.56%)	414.293 (43.01%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	872.125	241.835 (27.73%)	477.263 (54.72%)	404.389 (46.37%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	955.065	299.972 (31.41%)	480.246 (50.28%)	420.784 (44.06%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	955.217	289.455 (30.3%)	462.899 (48.46%)	391.052 (40.94%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	915.908	280.182 (30.59%)	459.843 (50.21%)	426.466 (46.56%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	828.309	250.989 (30.3%)	461.404 (55.7%)	407.729 (49.22%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	849.51	301.561 (35.5%)	462.203 (54.41%)	485.299 (57.13%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	831.064	296.826 (35.72%)	468.978 (56.43%)	432.39 (52.03%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	669.478	208.992 (31.22%)	348.335 (52.03%)	277.598 (41.46%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	644.651	212.15 (32.91%)	341.676 (53%)	296.356 (45.97%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	655.857	241.67 (36.85%)	350.738 (53.48%)	293.984 (44.82%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	652.962	229.365 (35.13%)	332.02 (50.85%)	278.116 (42.59%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	714.792	214.501 (30.01%)	349.959 (48.96%)	323.449 (45.25%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	752.608	231.712 (30.79%)	383.271 (50.93%)	347.639 (46.19%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	784.183	259.68 (33.11%)	387.016 (49.35%)	317.655 (40.51%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	782.402	247.16 (31.59%)	359.573 (45.96%)	323.368 (41.33%)

2.3 EishayParseTreeString

这个场景是将JSON格式字符串反序列化为JSONObject，这也是一个最常见的场景之一，这个场景在fastjson中的代码如下：

String str = "...";
JSONObject jsonObject = JSON.parseJSONObject(str);

fastjson2在这个场景也表现了远超fastjson1/jackson/gson的性能。在不同的环境中，fastjson1和jackson都只能相当于fastjson1性能的40%~60%，gson则更差一些。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > fastjson1 > jackson > gson
100%        55.39%      54.29%    36.53%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	933.434	517.022 (55.39%)	506.806 (54.29%)	340.962 (36.53%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	774.516	394.826 (50.98%)	440.933 (56.93%)	322.93 (41.69%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1024.532	583.808 (56.98%)	505.817 (49.37%)	320.666 (31.3%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1067.726	529.394 (49.58%)	471.898 (44.2%)	324.342 (30.38%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	880.183	474.715 (53.93%)	519.486 (59.02%)	362.992 (41.24%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	797.78	403.234 (50.54%)	484.84 (60.77%)	374.36 (46.93%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	979.102	536.398 (54.78%)	564.281 (57.63%)	397.783 (40.63%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	1010.951	521.952 (51.63%)	543.735 (53.78%)	396.25 (39.2%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	576.868	305.566 (52.97%)	356.642 (61.82%)	256.349 (44.44%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	533.525	319.366 (59.86%)	355.745 (66.68%)	260.384 (48.8%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	611.576	357.592 (58.47%)	385.607 (63.05%)	257.244 (42.06%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	612.693	340.3 (55.54%)	375.034 (61.21%)	255.753 (41.74%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	551.643	308.559 (55.93%)	320.324 (58.07%)	260.041 (47.14%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	655.95	361.855 (55.17%)	382.319 (58.28%)	306.29 (46.69%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	787.272	412.435 (52.39%)	426.889 (54.22%)	245.45 (31.18%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	793.481	368.496 (46.44%)	414.609 (52.25%)	279.315 (35.2%)

2.4 EishayParseTreeStringPretty

这个场景在fastjson中的代码如下：

// 这里输入的是格式化过后的json字符串
String str = "{\n" +
        "\t\"id\":123\n" +
        "}"; 
JSONObject jsonObject = JSON.parseJSONObject(str);

fastjson2在这个场景也表现了远超fastjson1/jackson/gson的性能。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > fastjson1 > gson
100%        63.68%    60.43%      43.28%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	744.185	449.72 (60.43%)	473.933 (63.68%)	322.106 (43.28%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	613.163	344.72 (56.22%)	417.245 (68.05%)	309.322 (50.45%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	814.548	483.549 (59.36%)	454.505 (55.8%)	301.853 (37.06%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	835.176	479.556 (57.42%)	459.29 (54.99%)	301.324 (36.08%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	705.798	423.443 (59.99%)	475.629 (67.39%)	333.679 (47.28%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	611.819	356.926 (58.34%)	441.875 (72.22%)	344.588 (56.32%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	752.479	472.299 (62.77%)	483.349 (64.23%)	371.936 (49.43%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	781.802	465.518 (59.54%)	504.176 (64.49%)	345.81 (44.23%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	493.539	267.924 (54.29%)	330.741 (67.01%)	239.599 (48.55%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	457.78	274.412 (59.94%)	326.581 (71.34%)	246.084 (53.76%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	501.855	311.258 (62.02%)	350.552 (69.85%)	245.072 (48.83%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	500.836	292.854 (58.47%)	337.513 (67.39%)	240.535 (48.03%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	465.583	286.793 (61.6%)	308.676 (66.3%)	248.297 (53.33%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	542.036	283.319 (52.27%)	355.39 (65.57%)	288.205 (53.17%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	618.605	367.727 (59.44%)	375.298 (60.67%)	265.321 (42.89%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	606.432	337.089 (55.59%)	355.232 (58.58%)	250.433 (41.3%)

2.5 EishayParseUTF8Bytes

这个是将UTF8格式的byte数组反序列化为Java对象，这个场景在缓存和RPC场景中常用。

这个场景在fastjson中的代码如下：

byte[] utf8Bytes = ...;
Bean bean = JSON.parseJSONObject(utf8Bytes, Bean.class);

这个场景fastjson2同样表现出了卓越的性能；在JDK8下，fastjson1和jackson的性能分别之后fastjson2的50.49%和63.22%; gson不直接支持输入utf8Bytes，需要先构造字符串，性能只有fastjson2的35%。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > fastjson1 > jackson > gson
100%        71.36%      57.27%    29.08%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	1083.498	773.228 (71.36%)	620.496 (57.27%)	315.12 (29.08%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	1110.444	755.773 (68.06%)	548.41 (49.39%)	306.689 (27.62%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1207.359	951.476 (78.81%)	567.635 (47.01%)	316.653 (26.23%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1121.7	893.757 (79.68%)	563.522 (50.24%)	319.072 (28.45%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	916.079	807.567 (88.15%)	605.357 (66.08%)	328.311 (35.84%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	1083.919	721.734 (66.59%)	587.903 (54.24%)	357.015 (32.94%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	1135.129	906.3 (79.84%)	619.861 (54.61%)	385.474 (33.96%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	1091.062	902.03 (82.67%)	584.882 (53.61%)	383.314 (35.13%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	723.162	563.093 (77.87%)	455.927 (63.05%)	239.648 (33.14%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	743.157	592.675 (79.75%)	442.215 (59.5%)	254.261 (34.21%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	766.054	758.582 (99.02%)	429.762 (56.1%)	253.173 (33.05%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	715.554	742.419 (103.75%)	417.566 (58.36%)	252.146 (35.24%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	806.11	658.904 (81.74%)	423.583 (52.55%)	240.131 (29.79%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	924.035	683.072 (73.92%)	478.147 (51.75%)	284.194 (30.76%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	975.486	767.593 (78.69%)	440.031 (45.11%)	276.31 (28.33%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	922.588	789.418 (85.57%)	453.087 (49.11%)	283.509 (30.73%)

2.6 EishayParseUTF8BytesPretty

这个是将格式化过后的UTF8格式的byte数组反序列化为Java对象，这个场景在缓存和RPC场景中常用。

这个场景在fastjson中的代码如下：

// 这里输入的是格式化过后的json字符串
String str = "{\n" +
        "\t\"id\":123\n" +
        "}";
byte[] utf8Bytes = str.getBytes(StandardCharsets.UTF_8);
Bean bean = JSON.parseJSONObject(utf8Bytes, Bean.class);

这个场景fastjson2同样表现出了卓越的性能；在JDK8下，fastjson1和jackson的性能分别之后fastjson2的53.72%和70.22%， gson不直接支持输入utf8Bytes，需要先构造字符串，性能只有fastjson2的41.25%。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > gson > fastjson1
100%        67.78%    37.4%  31.93%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	786.524	251.142 (31.93%)	533.112 (67.78%)	294.143 (37.4%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	834.493	225.711 (27.05%)	498.434 (59.73%)	303.7 (36.39%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	905.466	271.031 (29.93%)	508.311 (56.14%)	303.584 (33.53%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	872.907	258.943 (29.66%)	490.634 (56.21%)	292.248 (33.48%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	755.152	248.451 (32.9%)	537.181 (71.14%)	302.875 (40.11%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	796.538	228.723 (28.71%)	491.373 (61.69%)	342.609 (43.01%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	828.639	274.785 (33.16%)	557.601 (67.29%)	341.393 (41.2%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	818.144	262.646 (32.1%)	524.572 (64.12%)	351.046 (42.91%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	599.255	191.374 (31.94%)	416.723 (69.54%)	214.748 (35.84%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	590.414	193.621 (32.79%)	406.777 (68.9%)	240.051 (40.66%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	606.763	218.096 (35.94%)	395.387 (65.16%)	240.114 (39.57%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	573.736	208.397 (36.32%)	380.389 (66.3%)	234.282 (40.83%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	654.577	204.519 (31.24%)	372.8 (56.95%)	227.904 (34.82%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	703.777	152.967 (21.74%)	431.272 (61.28%)	275.286 (39.12%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	739.801	237.956 (32.16%)	400.672 (54.16%)	278.539 (37.65%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	702.402	218.875 (31.16%)	410.706 (58.47%)	271.897 (38.71%)

2.7 EishayParseTreeUTF8Bytes

这个是将UTF8格式的byte数组反序列化为JSONObject. 这个场景在fastjson中的代码如下：

byte[] utf8Bytes = ...;
JSONObject jsonObject = JSON.parseJSONObject(utf8Bytes);

这个场景fastjson2同样表现出了卓越的性能；在JDK8下，fastjson1和jackson的性能分别之后fastjson2的50.49%和63.22%; gson不直接支持输入utf8Bytes，需要先构造字符串，性能只有fastjson2的35%。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > fastjson1 > gson
100%        63.22%    50.49%      35.5%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	882.065	445.319 (50.49%)	557.669 (63.22%)	313.124 (35.5%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	654.899	360.26 (55.01%)	514.301 (78.53%)	315.332 (48.15%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1044.966	478.063 (45.75%)	568.644 (54.42%)	321.917 (30.81%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1019.586	475.741 (46.66%)	554.104 (54.35%)	313.421 (30.74%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	724.629	426.142 (58.81%)	572.952 (79.07%)	328.595 (45.35%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	721.44	360.332 (49.95%)	565.566 (78.39%)	360.34 (49.95%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	852.787	452.351 (53.04%)	606.124 (71.08%)	392.469 (46.02%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	880.7	472.351 (53.63%)	597.554 (67.85%)	374.988 (42.58%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	532.788	249.575 (46.84%)	441.72 (82.91%)	240.525 (45.14%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	446.808	287.303 (64.3%)	413.758 (92.6%)	254.144 (56.88%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	590.05	317.036 (53.73%)	456.912 (77.44%)	253.741 (43%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	558.045	301.757 (54.07%)	434.725 (77.9%)	252.7 (45.28%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	501.696	253.89 (50.61%)	372.582 (74.26%)	242.819 (48.4%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	530.747	337.552 (63.6%)	437.781 (82.48%)	297.201 (56%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	771.518	348.721 (45.2%)	507.337 (65.76%)	275.168 (35.67%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	714.17	325.797 (45.62%)	477.229 (66.82%)	241.458 (33.81%)

2.8 EishayParseTreeUTF8BytesPretty

这个是将格式化过后的UTF8格式的byte数组反序列化为JSONObject，这个场景在fastjson中的代码如下：

// 这里输入的是格式化过后的json字符串
String str = "{\n" +
        "\t\"id\":123\n" +
        "}";
byte[] utf8Bytes = str.getBytes(StandardCharsets.UTF_8);
JSONObject jsonObject = JSON.parseJSONObject(utf8Bytes);

这个场景fastjson2同样表现出了卓越的性能；在JDK8下，fastjson1和jackson的性能分别之后fastjson2的53.72%和70.22%， gson不直接支持输入utf8Bytes，需要先构造字符串，性能只有fastjson2的41.25%。

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > fastjson1 > gson
100%        70.22%    53.72%      41.25%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	716.744	385.002 (53.72%)	503.301 (70.22%)	295.637 (41.25%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	601.404	314.029 (52.22%)	490.292 (81.52%)	295.215 (49.09%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	851.286	411.4 (48.33%)	512.209 (60.17%)	293.605 (34.49%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	840.136	401.867 (47.83%)	506.222 (60.25%)	291.464 (34.69%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	635.907	366.566 (57.64%)	516.876 (81.28%)	301.97 (47.49%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	582.825	312.162 (53.56%)	488.51 (83.82%)	332.122 (56.98%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	682.302	381.257 (55.88%)	547.474 (80.24%)	364.789 (53.46%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	749.046	389.891 (52.05%)	541.245 (72.26%)	357.097 (47.67%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	466.952	220.237 (47.16%)	402.868 (86.28%)	220.594 (47.24%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	389.837	243.253 (62.4%)	391.982 (100.55%)	240.36 (61.66%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	499.026	273.415 (54.79%)	406.731 (81.5%)	236.56 (47.4%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	474.922	264.985 (55.8%)	396.4 (83.47%)	235.098 (49.5%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	471.287	234.086 (49.67%)	344.737 (73.15%)	230.781 (48.97%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	470.475	231.123 (49.13%)	421.706 (89.63%)	286.665 (60.93%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	622.918	250.837 (40.27%)	441.649 (70.9%)	279.948 (44.94%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	583.691	209.574 (35.9%)	440.884 (75.53%)	277.102 (47.47%)

2.9 EishayWriteString

这个是Java对象序列化成字符串，这个是最常用的场景之一。这个场景在fastjson中的代码如下：

Bean bean = ...;
String str = JSON.toJSONString(bean);

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > fastjson1 > gson
100%        54.03%    37.61%      20.69%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	1581.037	594.658 (37.61%)	854.22 (54.03%)	327.195 (20.69%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	1343.414	595.695 (44.34%)	906.331 (67.46%)	275.727 (20.52%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1712.266	613.577 (35.83%)	1001.97 (58.52%)	205.201 (11.98%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1699.625	593.921 (34.94%)	916.976 (53.95%)	205.046 (12.06%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	1624.636	659.362 (40.59%)	904.775 (55.69%)	364.295 (22.42%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	1442.604	593.853 (41.17%)	916.225 (63.51%)	287.247 (19.91%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	1841.055	643.602 (34.96%)	925.489 (50.27%)	338.389 (18.38%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	1688.356	629.806 (37.3%)	870.935 (51.58%)	325.734 (19.29%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	1033.834	446.877 (43.23%)	578.324 (55.94%)	278.484 (26.94%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	1046.255	424.717 (40.59%)	608.651 (58.17%)	248.206 (23.72%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	1163.12	433.589 (37.28%)	637.242 (54.79%)	188.469 (16.2%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	1136.576	381.31 (33.55%)	597.276 (52.55%)	188.559 (16.59%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	1221.452	502.207 (41.12%)	643.246 (52.66%)	328.688 (26.91%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	1167.871	424.902 (36.38%)	650.535 (55.7%)	255.595 (21.89%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	1372.859	417.568 (30.42%)	664.706 (48.42%)	204.356 (14.89%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	1314.683	441.703 (33.6%)	564.61 (42.95%)	200.301 (15.24%)

2.10 EishayWriteUTF8Bytes

这个是Java对象序列化成UTF格式的byte数组，这个是最常用的场景之一，在缓存和RPC场景常用。gson不直接支持，需要先序列化为String再转换为UTF8格式的byte数组。

这个场景在fastjson中的代码如下：

Bean bean = ...;
byte[] utf8Bytes = JSON.toJSONBytes(bean);

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2 > jackson > fastjson1 > gson
100%        59.13%    37.99%      19.52%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2	fastjson1	jackson	gson
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	1504.496	571.509 (37.99%)	889.578 (59.13%)	293.743 (19.52%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	1592.624	507.917 (31.89%)	854.436 (53.65%)	276.88 (17.39%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	1856.092	570.574 (30.74%)	969.387 (52.23%)	202.302 (10.9%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	1835.109	548.702 (29.9%)	938.845 (51.16%)	203.566 (11.09%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	1487.078	582.897 (39.2%)	836.831 (56.27%)	333.118 (22.4%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	1642.28	543.887 (33.12%)	830.435 (50.57%)	273.111 (16.63%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	1769.397	576.354 (32.57%)	895.627 (50.62%)	338.069 (19.11%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	1809.864	554.232 (30.62%)	865.073 (47.8%)	320.201 (17.69%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	1111.549	407.663 (36.68%)	564.422 (50.78%)	233.167 (20.98%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	1167.558	387.93 (33.23%)	563.532 (48.27%)	245.837 (21.06%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	1165.606	385.793 (33.1%)	602.722 (51.71%)	188.328 (16.16%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	1178.795	350.563 (29.74%)	581.995 (49.37%)	187.292 (15.89%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	1329.046	437.516 (32.92%)	658.158 (49.52%)	289.968 (21.82%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	1232.089	399.648 (32.44%)	634.515 (51.5%)	266.952 (21.67%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	1360.522	356.383 (26.19%)	678.117 (49.84%)	197.732 (14.53%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	1412.507	401.57 (28.43%)	602.449 (42.65%)	195.807 (13.86%)

2.11 EishayWriteBinary

这个场景是将Java对象序列化为二进制的byte数组，是缓存序列化和RPC场景使用序列化协议的典型场景。

fastson2内置支持二进制格式jsonb，jsonb是fastjson2为了RPC场景设计的序列化和反序列化设置的高性能序列化协议。

jsonb有两种映射方式，将Java对象映射为类似JSONObject的KV格式，也可以映射类似JSONArray的数组格式。

fastjson2JSONBArrayMapping的代码如下：

Bean bean = ...;
byte[] jsonbBytes = JSONB.toBytes(bean, JSONWriter.Feature.BeanToArray);

fastjson2JSONB的代码如下：

Bean bean = ...;
byte[] jsonbBytes = JSONB.toBytes(bean);

fastjson2UTF8Bytes的代码如下：

Bean bean = ...;
byte[] utf8Bytes = JSON.toJSONBytes(bean);

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2JSONBArrayMapping > fastjson2JSONB > kryo  >  fastjson2UTF8Bytes > hessian > javaSerialize
165.59%                      100%             90.73%   71.59%               16.05%    10.37%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2JSONB	fastjson2JSONBArrayMapping	fastjson2UTF8Bytes	hessian	javaSerialize	kryo
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	2109.521	3492.913 (165.58%)	1510.194 (71.59%)	338.479 (16.05%)	218.692 (10.37%)	1913.902 (90.73%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	1832.584	3305.27 (180.36%)	1587.198 (86.61%)	327.303 (17.86%)	204.381 (11.15%)	1914.781 (104.49%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	2398.784	3445.758 (143.65%)	1870.477 (77.98%)	323.179 (13.47%)	223.274 (9.31%)	1875.871 (78.2%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	2631.992	4095.755 (155.61%)	1821.928 (69.22%)	317.838 (12.08%)	241.661 (9.18%)	1807.544 (68.68%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	1977.848	3205.618 (162.08%)	1491.867 (75.43%)	414.534 (20.96%)	228.309 (11.54%)	1761.061 (89.04%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	1986.387	3064.238 (154.26%)	1696.878 (85.43%)	389.741 (19.62%)	242.035 (12.18%)	1981.922 (99.78%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	1974.045	3129.304 (158.52%)	1716.659 (86.96%)	411.243 (20.83%)	264.826 (13.42%)	1790.338 (90.69%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	2385.79	4073.694 (170.75%)	1742.115 (73.02%)	409.094 (17.15%)	260.802 (10.93%)	1721.088 (72.14%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	1331.057	1885.63 (141.66%)	1108.585 (83.29%)	297.788 (22.37%)	176.367 (13.25%)	1107.134 (83.18%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	1476.657	2319.398 (157.07%)	1159.41 (78.52%)	322.343 (21.83%)	189.048 (12.8%)	1296.507 (87.8%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	1443.992	2228.329 (154.32%)	1173.612 (81.28%)	317.746 (22%)	181.799 (12.59%)	1230.239 (85.2%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	1608.314	2627.964 (163.4%)	1165.379 (72.46%)	312.327 (19.42%)	195.064 (12.13%)	1282.308 (79.73%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	1577.12	2362.358 (149.79%)	1330.904 (84.39%)	349.276 (22.15%)	204.709 (12.98%)	1431.816 (90.79%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	1774.645	2726.764 (153.65%)	1360.397 (76.66%)	340.869 (19.21%)	218.516 (12.31%)	1555.83 (87.67%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	1732.468	2463.208 (142.18%)	1381.276 (79.73%)	331.195 (19.12%)	207.308 (11.97%)	1439.65 (83.1%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	2004.688	3157.957 (157.53%)	1417.265 (70.7%)	332.99 (16.61%)	224.267 (11.19%)	1454.408 (72.55%)

2.12 EishayParseBinary

这个场景是将二进制的byte数组反序列化为Java对象，是缓存序列化和RPC场景使用序列化协议的典型场景。

fastjson2JSONBArrayMapping的代码如下：

byte[] jsonbBytes = JSONB.toBytes(bean, JSONWriter.Feature.BeanToArray);
Bean bean = JSONB.parseObject(jsonbBytes, Bean.class, JSONReader.Feature.SupportArrayToBean);

fastjson2JSONB的代码如下：

byte[] jsonbBytes = JSONB.toBytes(bean);
Bean bean = JSONB.parseObject(jsonbBytes, Bean.class);

fastjson2UTF8Bytes的代码如下：

byte[] utf8Bytes = JSON.toJSONBytes(bean);
Bean bean = JSON.parseObject(utf8Bytes, Bean.class);

性能排序如下：

// ecs.c7.xlarge-oracle-jdk1.8.0_341_x64

fastjson2JSONBArrayMapping > kryo   >  fastjson2JSONB > fastjson2UTF8Bytes > hessian > javaSerialize
153.96%                      100%.25%  100%             67.16%               16.83%    2.84%

下面是不同CPU不同JDK下的性能比较表格，第一列是阿里云机器规格和JDK的版本信息，第二列到第五列是JMH的Throughput数据，越大越好。第三列到第五列中的百分比是Throughput相比fastjson2的比，比如第三列的百分比是fastjson1/fastjson2的百分数表示。

	fastjson2JSONB	fastjson2JSONBArrayMapping	fastjson2UTF8Bytes	hessian	javaSerialize	kryo
ecs.c7.xlarge-oracle-jdk1.8.0_341_x64	1648.648	2538.226 (153.96%)	1107.184 (67.16%)	277.395 (16.83%)	46.794 (2.84%)	1652.73 (100.25%)
ecs.c7.xlarge-oracle-jdk-11.0.16_x64	2178.064	3546.334 (162.82%)	1141.561 (52.41%)	248.849 (11.43%)	48.114 (2.21%)	1706.163 (78.33%)
ecs.c7.xlarge-oracle-jdk-17.0.4_x64	2210.446	3581.414 (162.02%)	1240.231 (56.11%)	257.332 (11.64%)	52.042 (2.35%)	1499.663 (67.84%)
ecs.c7.xlarge-oracle-jdk-18.0.2_x6	2208.152	3733.627 (169.08%)	1201.981 (54.43%)	246.276 (11.15%)	51.45 (2.33%)	1633.479 (73.97%)
ecs.c7a.xlarge-oracle-jdk1.8.0_341_x64	1293.382	1745.828 (134.98%)	929.428 (71.86%)	255.801 (19.78%)	43.298 (3.35%)	1446.488 (111.84%)
ecs.c7a.xlarge-oracle-jdk-11.0.16_x64	2067.736	3332.012 (161.14%)	1058.228 (51.18%)	243.973 (11.8%)	47.149 (2.28%)	1554.486 (75.18%)
ecs.c7a.xlarge-oracle-jdk-17.0.4_x64	2223.355	3609.628 (162.35%)	1120.472 (50.4%)	250.796 (11.28%)	53.126 (2.39%)	1623.24 (73.01%)
ecs.c7a.xlarge-oracle-jdk-18.0.2_x64	2114.786	3522.422 (166.56%)	1126.199 (53.25%)	246.965 (11.68%)	51.824 (2.45%)	1644.455 (77.76%)
ecs.c6r.xlarge-oracle-jdk1.8.0_341_aarch64	1256.386	2110.098 (167.95%)	730.412 (58.14%)	190.884 (15.19%)	36.785 (2.93%)	1300.187 (103.49%)
ecs.c6r.xlarge-oracle-jdk-11.0.16_aarch64	1418.018	2542.738 (179.32%)	756.803 (53.37%)	179.716 (12.67%)	37.508 (2.65%)	1242.747 (87.64%)
ecs.c6r.xlarge-oracle-jdk-17.0.4_aarch64	1303.159	2187.211 (167.84%)	761.374 (58.43%)	208.406 (15.99%)	39.672 (3.04%)	1223.11 (93.86%)
ecs.c6r.xlarge-oracle-jdk-18.0.2_aarch64	1271.578	2163.511 (170.14%)	719.966 (56.62%)	187.4 (14.74%)	37.463 (2.95%)	1189.698 (93.56%)
ecs.g8m.xlarge-oracle-jdk1.8.0_341_aarch64	1270.914	2272.559 (178.81%)	816.239 (64.22%)	200.373 (15.77%)	42.276 (3.33%)	1191.146 (93.72%)
ecs.g8m.xlarge-oracle-jdk-11.0.16_aarch64	1811.452	3101.861 (171.24%)	982.786 (54.25%)	200.661 (11.08%)	40.634 (2.24%)	1413.636 (78.04%)
ecs.g8m.xlarge-oracle-jdk-17.0.4_aarch64	1644.118	2692.082 (163.74%)	989.303 (60.17%)	156.648 (9.53%)	40.901 (2.49%)	1090.731 (66.34%)
ecs.g8m.xlarge-oracle-jdk-18.0.2_aarch64	1630.171	2673.894 (164.03%)	938.366 (57.56%)	201.279 (12.35%)	42.244 (2.59%)	1222.099 (74.97%)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fastjson_benchmark

1. 测试环境

1.1 各个库的版本

1.2 服务器环境

ecs.c7.xlarge

ecs.c7a.xlarge

ecs.c6r.xlarge

ecs.g8m.xlarge

1.3 JDK版本

1.4 测试代码以及运行方式

2. 测试数据&结果分析

2.1 EishayParseString

2.2 EishayParseStringPretty

2.3 EishayParseTreeString

2.4 EishayParseTreeStringPretty

2.5 EishayParseUTF8Bytes

2.6 EishayParseUTF8BytesPretty

2.7 EishayParseTreeUTF8Bytes

2.8 EishayParseTreeUTF8BytesPretty

2.9 EishayWriteString

2.10 EishayWriteUTF8Bytes

2.11 EishayWriteBinary

2.12 EishayParseBinary

Clone this wiki locally