三里屯摇滚 发表于 2012-02-16 19:30

Redis-Sampler:深入了解你的Redis存储

Redis-Sampler:深入了解你的Redis存储








redis-sampler 是Redis作者@antirez 同学开发的一个ruby 小工具,用于对Redis存储概况进行抽样检测并给出分析结果。

项目地址:https://github.com/antirez/redis-sampler

使用方式:

下载源码,执行下面命令:

./redis-sampler.rb
host,port和db三个参数都不用解释(db默认为0),samplesize 就是这个抽样检测的样本数量。脚本输出如下例:

Sampling localhost:6379 DB:4 with 1000000 RANDOMKEYS

TYPES
===== zset: 873268 (87.33%)    string: 124995 (12.50%)set: 1022 (0.10%)
hash: 576 (0.06%)      list: 139 (0.01%)      

STRINGS, SIZE OF VALUES
=======================
6: 61222 (48.98%)      7: 17056 (13.65%)      13: 8274 (6.62%)
15: 7991 (6.39%)         5: 4629 (3.70%)          31: 3263 (2.61%)
20: 2670 (2.14%)         2: 2518 (2.01%)          27: 1675 (1.34%)
42: 1270 (1.02%)         159: 893 (0.71%)         1: 705 (0.56%)
47: 641 (0.51%)          34: 594 (0.48%)          41: 521 (0.42%)
38: 493 (0.39%)          28: 413 (0.33%)          22: 406 (0.32%)
139: 351 (0.28%)         29: 343 (0.27%)          83: 337 (0.27%)
(suppressed 172 items with perc < 0.5% for a total of 6.98%)
Average: 15.97 Standard Deviation: 26.52
Min: 0 Max: 1123

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 8: 82913 (66.33%)   <= 16: 16789 (13.43%)    <= 32: 9571 (7.66%)
<= 64: 6232 (4.99%)      <= 128: 3333 (2.67%)   <= 256: 2682 (2.15%)
<= 2: 2518 (2.01%)       <= 1: 740 (0.59%)      <= 4: 199 (0.16%)
<= 512: 14 (0.01%)       <= 1024: 3 (0.00%)       <= 2048: 1 (0.00%)      

LISTS, NUMBER OF ELEMENTS
=========================
2: 28 (20.14%)         5: 18 (12.95%)         9: 10 (7.19%)
8: 9 (6.47%)             11: 9 (6.47%)            13: 7 (5.04%)
12: 7 (5.04%)            14: 6 (4.32%)            15: 6 (4.32%)
4: 4 (2.88%)             16: 4 (2.88%)            21: 4 (2.88%)
27: 3 (2.16%)            7: 3 (2.16%)             10: 3 (2.16%)
19: 2 (1.44%)            1: 2 (1.44%)             25: 2 (1.44%)
41: 1 (0.72%)            3: 1 (0.72%)             17: 1 (0.72%)
(suppressed 9 items with perc < 0.5% for a total of 6.47%)
Average: 10.58 Standard Deviation: 8.58
Min: 1 Max: 42

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 16: 52 (37.41%)       <= 8: 30 (21.58%)      <= 2: 28 (20.14%)
<= 32: 17 (12.23%)       <= 64: 5 (3.60%)         <= 4: 5 (3.60%)
<= 1: 2 (1.44%)         

LISTS, SIZE OF ELEMENTS
=======================
7: 106 (76.26%)          6: 33 (23.74%)
Average: 6.76 Standard Deviation: 0.43
Min: 6 Max: 7

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 8: 139 (100.00%)   

SETS, NUMBER OF ELEMENTS
========================
1: 216361 (24.78%)       2: 106871 (12.24%)       3: 67648 (7.75%)
4: 48207 (5.52%)         5: 36085 (4.13%)         6: 29597 (3.39%)
7: 23765 (2.72%)         8: 22549 (2.58%)         9: 20143 (2.31%)
10: 18069 (2.07%)      11: 16387 (1.88%)      12: 15009 (1.72%)
13: 13869 (1.59%)      14: 12683 (1.45%)      15: 12319 (1.41%)
16: 10794 (1.24%)      17: 10068 (1.15%)      18: 8925 (1.02%)
19: 8007 (0.92%)         20: 7618 (0.87%)         22: 7240 (0.83%)
21: 7055 (0.81%)         23: 5973 (0.68%)         24: 5771 (0.66%)
25: 4934 (0.57%)
(suppressed 2061 items with perc < 0.5% for a total of 15.72%)
Average: 1.24 Standard Deviation: 2.34
Min: 1 Max: 62

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 1: 975 (95.40%)       <= 2: 21 (2.05%)         <= 8: 10 (0.98%)
<= 4: 9 (0.88%)          <= 16: 5 (0.49%)         <= 32: 1 (0.10%)
<= 64: 1 (0.10%)      

SETS, SIZE OF ELEMENTS
======================
19: 871 (85.23%)         3: 66 (6.46%)            4: 65 (6.36%)
5: 11 (1.08%)            2: 9 (0.88%)
Average: 16.71 Standard Deviation: 5.50
Min: 2 Max: 19

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 32: 871 (85.23%)      <= 4: 131 (12.82%)       <= 8: 11 (1.08%)
<= 2: 9 (0.88%)         

SORTED SETS, NUMBER OF ELEMENTS
===============================
1: 216361 (24.78%)       2: 106871 (12.24%)       3: 67648 (7.75%)
4: 48207 (5.52%)         5: 36085 (4.13%)         6: 29597 (3.39%)
7: 23765 (2.72%)         8: 22549 (2.58%)         9: 20143 (2.31%)
10: 18069 (2.07%)      11: 16387 (1.88%)      12: 15009 (1.72%)
13: 13869 (1.59%)      14: 12683 (1.45%)      15: 12319 (1.41%)
16: 10794 (1.24%)      17: 10068 (1.15%)      18: 8925 (1.02%)
19: 8007 (0.92%)         20: 7618 (0.87%)         22: 7240 (0.83%)
21: 7055 (0.81%)         23: 5973 (0.68%)         24: 5771 (0.66%)
25: 4934 (0.57%)
(suppressed 2061 items with perc < 0.5% for a total of 15.72%)
Average: 25.47 Standard Deviation: 110.64
Min: 1 Max: 7018

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 1: 216361 (24.78%)    <= 16: 119273 (13.66%)   <= 4: 115855 (13.27%)
<= 8: 111996 (12.82%)    <= 2: 106871 (12.24%)    <= 32: 92814 (10.63%)
<= 64: 52715 (6.04%)   <= 128: 29286 (3.35%)    <= 256: 13988 (1.60%)
<= 512: 7098 (0.81%)   <= 1024: 4762 (0.55%)    <= 2048: 1840 (0.21%)
<= 4096: 393 (0.05%)   <= 8192: 16 (0.00%)   

SORTED SETS, SIZE OF ELEMENTS
=============================
6: 710230 (81.33%)       5: 75292 (8.62%)         4: 68661 (7.86%)
3: 17136 (1.96%)         2: 1412 (0.16%)          9: 253 (0.03%)
1: 76 (0.01%)            8: 74 (0.01%)            7: 39 (0.00%)
21: 6 (0.00%)            30: 6 (0.00%)            20: 6 (0.00%)
23: 6 (0.00%)            26: 5 (0.00%)            34: 5 (0.00%)
24: 4 (0.00%)            27: 4 (0.00%)            18: 4 (0.00%)
15: 4 (0.00%)            38: 3 (0.00%)            43: 3 (0.00%)
(suppressed 24 items with perc < 0.5% for a total of 0.00%)
Average: 5.69 Standard Deviation: 0.77
Min: 1 Max: 63

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 8: 785635 (89.96%)    <= 4: 85797 (9.82%)      <= 2: 1412 (0.16%)
<= 16: 262 (0.03%)       <= 1: 76 (0.01%)         <= 32: 58 (0.01%)
<= 64: 28 (0.00%)      

HASHES, NUMBER OF FIELDS
========================
1: 301 (52.26%)          12: 177 (30.73%)         11: 95 (16.49%)
13: 2 (0.35%)            14: 1 (0.17%)
Average: 6.09 Standard Deviation: 5.34
Min: 1 Max: 14

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 1: 301 (52.26%)       <= 16: 275 (47.74%)   

HASHES, SIZE OF FIELDS
======================
17: 301 (52.26%)         22: 179 (31.08%)         13: 95 (16.49%)
12: 1 (0.17%)
Average: 17.89 Standard Deviation: 3.11
Min: 12 Max: 22

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 32: 480 (83.33%)      <= 16: 96 (16.67%)      

HASHES, SIZE OF VALUES
======================
13: 116 (20.14%)         3: 103 (17.88%)          410: 44 (7.64%)
409: 38 (6.60%)          408: 27 (4.69%)          14: 22 (3.82%)
407: 17 (2.95%)          395: 12 (2.08%)          406: 12 (2.08%)
392: 11 (1.91%)          396: 11 (1.91%)          393: 11 (1.91%)
4: 10 (1.74%)            12: 10 (1.74%)         411: 9 (1.56%)
5: 7 (1.22%)             376: 7 (1.22%)         405: 6 (1.04%)
349: 5 (0.87%)         347: 5 (0.87%)         359: 5 (0.87%)
(suppressed 43 items with perc < 0.5% for a total of 15.28%)
Average: 207.90 Standard Deviation: 193.12
Min: 1 Max: 416

Powers of two distribution: (NOTE <= p means: p/2 < x <= p)
<= 512: 298 (51.74%)   <= 16: 151 (26.22%)      <= 4: 113 (19.62%)
<= 8: 10 (1.74%)         <= 2: 3 (0.52%)          <= 1: 1 (0.17%)

冰释一片天 发表于 2012-02-17 22:33

谢谢分享
页: [1]
查看完整版本: Redis-Sampler:深入了解你的Redis存储