最近看到redis4支持内存碎片清理了, 之前一直期待有这么一个功能, 因为之前遇到内存碎片的解决办法就是重启, 现在终于有了优雅的解决方案.\^o^/, 这个功能其实oranagra 在2017年1月1日已经提交pr了, 相关地址: https://github.com/antirez/redis/pull/3720
版本说明:
- Redis 4.0-RC3 以上版本才支持的
- 需要使用jemalloc作为内存分配器(默认的)
功能介绍:
- 支持在运行期进行自动内存碎片清理 (config set activedefrag yes)
- 支持通过命令 memory purge 进行清理(与自动清理区域不同)
功能验证流程:
(1) 首先需要拉取4.0-RC3之后的版本代码, 编译
(2) 启动时限定内存大小为1g并启动lru, 命令如下:
./src/redis-server --maxmemory 1gb --maxmemory-policy allkeys-lru --activedefrag no --port 6383(3) 构造大量数据并导致lru, 这样可以触发内存碎片, 命令如下:
redis-cli -p 6383 debug populate 7000000 asdf 150(4) 查看当前的内存使用情况, 会发现有200多万的数据被清理掉了
$ redis-cli -p 6383 info keyspace
# Keyspace
db0:keys=4649543,expires=0,avg_ttl=0(5) 查看当前的内存碎片率, 这时碎片率(mem_fragmentation_ratio)很高 : 1.54, 意味着54%的内存浪费
$ redis-cli -p 6383 info memory
# Memory
used_memory:1073741736
used_memory_human:1024.00M
used_memory_rss:1650737152
used_memory_rss_human:1.54G
used_memory_peak:1608721680
used_memory_peak_human:1.50G
used_memory_peak_perc:66.75%
used_memory_overhead:253906398
used_memory_startup:766152
used_memory_dataset:819835338
used_memory_dataset_perc:76.41%
total_system_memory:67535904768
total_system_memory_human:62.90G
used_memory_lua:37888
used_memory_lua_human:37.00K
maxmemory:1073741824
maxmemory_human:1.00G
maxmemory_policy:allkeys-lru
mem_fragmentation_ratio:1.54
mem_allocator:jemalloc-4.0.3
active_defrag_running:0
lazyfree_pending_objects:0(6) 看看内存分配的详细情况, 这个地方看不懂可以看看: 科普文, 关键的是util指标, 指的是内存利用率, 最大的bins内存util是0.661, 说明内存利用率不高
$ echo "`redis-cli -p 6383 memory malloc-stats`"
___ Begin jemalloc statistics ___
Version: 4.0.3-0-ge9192eacf8935e29fc62fddc2701f7942b1cc02c
Assertions disabled
Run-time option settings:
opt.abort: false
opt.lg_chunk: 21
opt.dss: "secondary"
opt.narenas: 48
opt.lg_dirty_mult: 3 (arenas.lg_dirty_mult: 3)
opt.stats_print: false
opt.junk: "false"
opt.quarantine: 0
opt.redzone: false
opt.zero: false
opt.tcache: true
opt.lg_tcache_max: 15
CPUs: 12
Arenas: 48
Pointer size: 8
Quantum size: 8
Page size: 4096
Min active:dirty page ratio per arena: 8:1
Maximum thread-cached size class: 32768
Chunk size: 2097152 (2^21)
Allocated: 1074509704, active: 1609732096, metadata: 41779072, resident: 1651118080, mapped: 1652555776
Current active ceiling: 1610612736
arenas[0]:
assigned threads: 1
dss allocation precedence: secondary
min active:dirty page ratio: 8:1
dirty pages: 393001:24 active:dirty, 0 sweeps, 0 madvises, 0 purged
allocated nmalloc ndalloc nrequests
small: 1006565256 28412640 9802493 35714594
large: 835584 20 11 20
huge: 67108864 1 0 1
total: 1074509704 28412661 9802504 35714615
active: 1609732096
mapped: 1650458624
metadata: mapped: 40202240, allocated: 491904
bins: size ind allocated nmalloc ndalloc nrequests curregs curruns regs pgs util nfills nflushes newruns reruns
8 0 1992 319 70 357 249 1 512 1 0.486 7 8 1 0
16 1 148618896 14110300 4821619 14310175 9288681 55119 256 1 0.658 141103 48825 55119 5
24 2 112104360 7200400 2529385 7300348 4671015 14064 512 3 0.648 72004 25881 14064 1
32 3 288 112 103 7003270 9 1 128 1 0.070 3 7 1 0
40 4 360 109 100 171 9 1 512 5 0.017 3 7 1 0
48 5 1248 112 86 63 26 1 256 3 0.101 2 5 1 0
56 6 896 106 90 16 16 1 512 7 0.031 2 6 1 0
64 7 128 64 62 5 2 1 64 1 0.031 1 3 1 0
80 8 880 106 95 7 11 1 256 5 0.042 2 4 1 0
96 9 9120 212 117 97 95 1 128 3 0.742 4 6 2 1
112 10 336 109 106 2 3 1 256 7 0.011 3 6 3 0
128 11 640 40 35 4 5 1 32 1 0.156 3 4 2 0
160 12 740617440 7000148 2371289 7000001 4628859 54688 128 5 0.661 70271 24334 54689 4
192 13 768 68 64 1 4 1 64 3 0.062 2 4 2 0
224 14 4683616 100000 79091 99946 20909 781 128 7 0.209 1000 1641 782 0
256 15 0 16 16 4 0 0 16 1 1 1 3 1 0
320 16 5120 64 48 16 16 1 64 5 0.250 1 3 1 0
384 17 768 33 31 2 2 1 32 3 0.062 1 3 1 0
448 18 28672 64 0 0 64 1 64 7 1 1 0 1 0
512 19 1024 10 8 4 2 1 8 1 0.250 1 2 2 0
640 20 0 32 32 1 0 0 32 5 1 1 3 1 0
---
896 22 48384 85 31 50 54 2 32 7 0.843 2 3 2 0
1024 23 3072 10 7 3 3 1 4 1 0.750 1 2 3 0
1280 24 20480 16 0 0 16 1 16 5 1 1 0 1 0
1536 25 15360 10 0 0 10 2 8 3 0.625 1 0 2 0
1792 26 28672 16 0 0 16 1 16 7 1 1 0 1 0
2048 27 4096 10 8 2 2 1 2 1 1 1 2 5 0
---
3584 30 35840 10 0 0 10 2 8 7 0.625 1 0 2 0
---
5120 32 250880 49 0 49 49 13 4 5 0.942 0 0 13 0
---
8192 35 81920 10 0 0 10 10 1 2 1 1 0 10 0
---
large: size ind allocated nmalloc ndalloc nrequests curruns
16384 39 16384 2 1 2 1
20480 40 40960 2 0 2 2
---
32768 43 32768 1 0 1 1
40960 44 40960 10 9 10 1
---
81920 48 81920 1 0 1 1
---
131072 51 131072 1 0 1 1
163840 52 163840 1 0 1 1
---
327680 56 327680 1 0 1 1
---
1048576 63 0 1 1 1 0
---
huge: size ind allocated nmalloc ndalloc nrequests curhchunks
---
67108864 87 67108864 1 0 1 1
---
--- End jemalloc statistics ---(7) 开启自动内存碎片整理
$ redis-cli -p 6383 config set activedefrag yes
OK(8) 等会儿再看看, 发现内存碎片降低了
$ redis-cli -p 6383 info memory
# Memory
used_memory:1073740712
used_memory_human:1024.00M
used_memory_rss:1253371904
used_memory_rss_human:1.17G
used_memory_peak:1608721680
used_memory_peak_human:1.50G
used_memory_peak_perc:66.74%
used_memory_overhead:253906398
used_memory_startup:766152
used_memory_dataset:819834314
used_memory_dataset_perc:76.41%
total_system_memory:67535904768
total_system_memory_human:62.90G
used_memory_lua:37888
used_memory_lua_human:37.00K
maxmemory:1073741824
maxmemory_human:1.00G
maxmemory_policy:allkeys-lru
mem_fragmentation_ratio:1.17
mem_allocator:jemalloc-4.0.3
active_defrag_running:0
lazyfree_pending_objects:0(9) 可以再看看内存利用率, 可以看到已经上升到0.82
$ echo "`redis-cli -p 6383 memory malloc-stats`"
___ Begin jemalloc statistics ___
Version: 4.0.3-0-ge9192eacf8935e29fc62fddc2701f7942b1cc02c
Assertions disabled
Run-time option settings:
opt.abort: false
opt.lg_chunk: 21
opt.dss: "secondary"
opt.narenas: 48
opt.lg_dirty_mult: 3 (arenas.lg_dirty_mult: 3)
opt.stats_print: false
opt.junk: "false"
opt.quarantine: 0
opt.redzone: false
opt.zero: false
opt.tcache: true
opt.lg_tcache_max: 15
CPUs: 12
Arenas: 48
Pointer size: 8
Quantum size: 8
Page size: 4096
Min active:dirty page ratio per arena: 8:1
Maximum thread-cached size class: 32768
Chunk size: 2097152 (2^21)
Allocated: 1074509800, active: 1307602944, metadata: 41779072, resident: 1512247296, mapped: 1652555776
Current active ceiling: 1308622848
arenas[0]:
assigned threads: 1
dss allocation precedence: secondary
min active:dirty page ratio: 8:1
dirty pages: 319239:39882 active:dirty, 4878 sweeps, 6343 madvises, 33915 purged
allocated nmalloc ndalloc nrequests
small: 1006565352 35456589 16846439 45126633
large: 835584 24 15 24
huge: 67108864 1 0 1
total: 1074509800 35456614 16846454 45126658
active: 1307602944
mapped: 1650458624
metadata: mapped: 40202240, allocated: 491904
bins: size ind allocated nmalloc ndalloc nrequests curregs curruns regs pgs util nfills nflushes newruns reruns
8 0 1992 319 70 357 249 1 512 1 0.486 7 8 1 0
16 1 148618896 17658482 8369801 17858357 9288681 44332 256 1 0.818 141103 48825 55119 26364
24 2 112104360 8897298 4226283 8997246 4671015 11525 512 3 0.791 72004 25881 14064 6205
32 3 384 115 103 9371363 12 1 128 1 0.093 4 7 1 0
40 4 360 109 100 171 9 1 512 5 0.017 3 7 1 0
48 5 1248 112 86 63 26 1 256 3 0.101 2 5 1 0
56 6 896 106 90 16 16 1 512 7 0.031 2 6 1 0
64 7 128 64 62 5 2 1 64 1 0.031 1 3 1 0
80 8 880 106 95 7 11 1 256 5 0.042 2 4 1 0
96 9 9120 212 117 97 95 1 128 3 0.742 4 6 2 1
112 10 336 109 106 2 3 1 256 7 0.011 3 6 3 0
128 11 640 40 35 4 5 1 32 1 0.156 3 4 2 0
160 12 740617440 8788058 4159199 8787911 4628859 44056 128 5 0.820 70271 24334 54689 26488
192 13 768 68 64 1 4 1 64 3 0.062 2 4 2 0
224 14 4683616 110956 90047 110902 20909 467 128 7 0.349 1000 1641 782 105
256 15 0 16 16 4 0 0 16 1 1 1 3 1 0
320 16 5120 64 48 16 16 1 64 5 0.250 1 3 1 0
384 17 768 33 31 2 2 1 32 3 0.062 1 3 1 0
448 18 28672 64 0 0 64 1 64 7 1 1 0 1 0
512 19 1024 10 8 4 2 1 8 1 0.250 1 2 2 0
640 20 0 32 32 1 0 0 32 5 1 1 3 1 0
---
896 22 48384 85 31 50 54 2 32 7 0.843 2 3 2 0
1024 23 3072 10 7 3 3 1 4 1 0.750 1 2 3 0
1280 24 20480 16 0 0 16 1 16 5 1 1 0 1 0
1536 25 15360 10 0 0 10 2 8 3 0.625 1 0 2 0
1792 26 28672 16 0 0 16 1 16 7 1 1 0 1 0
2048 27 4096 10 8 2 2 1 2 1 1 1 2 5 0
---
3584 30 35840 10 0 0 10 2 8 7 0.625 1 0 2 0
---
5120 32 250880 49 0 49 49 13 4 5 0.942 0 0 13 0
---
8192 35 81920 10 0 0 10 10 1 2 1 1 0 10 0
---
large: size ind allocated nmalloc ndalloc nrequests curruns
16384 39 16384 2 1 2 1
20480 40 40960 2 0 2 2
---
32768 43 32768 1 0 1 1
40960 44 40960 14 13 14 1
---
81920 48 81920 1 0 1 1
---
131072 51 131072 1 0 1 1
163840 52 163840 1 0 1 1
---
327680 56 327680 1 0 1 1
---
1048576 63 0 1 1 1 0
---
huge: size ind allocated nmalloc ndalloc nrequests curhchunks
---
67108864 87 67108864 1 0 1 1
---
--- End jemalloc statistics ---(10) 别急, 还有一个大招: 手动清理
$ redis-cli -p 6383 memory purge
(11) 再次查看内存使用情况: 发现碎片率降到1.04, 内存利用率到0.998, 内存碎片基本上消灭了^_^
$ redis-cli -p 6383 info memory
# Memory
used_memory:1073740904
used_memory_human:1024.00M
used_memory_rss:1118720000
used_memory_rss_human:1.04G
used_memory_peak:1608721680
used_memory_peak_human:1.50G
used_memory_peak_perc:66.74%
used_memory_overhead:253906398
used_memory_startup:766152
used_memory_dataset:819834506
used_memory_dataset_perc:76.41%
total_system_memory:67535904768
total_system_memory_human:62.90G
used_memory_lua:37888
used_memory_lua_human:37.00K
maxmemory:1073741824
maxmemory_human:1.00G
maxmemory_policy:allkeys-lru
mem_fragmentation_ratio:1.04
mem_allocator:jemalloc-4.0.3
active_defrag_running:0
lazyfree_pending_objects:0
$ 配置说明:
# Enabled active defragmentation
# 碎片整理总开关
# activedefrag yes
# Minimum amount of fragmentation waste to start active defrag
# 内存碎片达到多少的时候开启整理
active-defrag-ignore-bytes 100mb
# Minimum percentage of fragmentation to start active defrag
# 碎片率达到百分之多少开启整理
active-defrag-threshold-lower 10
# Maximum percentage of fragmentation at which we use maximum effort
# 碎片率小余多少百分比开启整理
active-defrag-threshold-upper 100
# Minimal effort for defrag in CPU percentage
active-defrag-cycle-min 25
# Maximal effort for defrag in CPU percentage
active-defrag-cycle-max 75总结:
从测试的结果看, 效果还是非常不错的, 另外在配置中我们可以看到如下一段声明:
说明现在这个功能还是实验性质的, 对应的命令在官方文档中都没有看到. 但是它也说经过了压力测试, 而且现在也一年多了, 经受了一些考验, 可以尝试小流量上线观察
TODO redis 内存碎片整理实现
来源：oschina
链接：https://my.oschina.net/u/871213/blog/1620666