内存泄漏问题 about tairhash

[chenyang8094]

1. 模拟数据 插入以下结构的大约数据 1万个
   exhset us:126902 session:dea155d63828f002484fb2d67a879c81 "" EX 60
2. 查看每个数据占用内存情况 (大约152)
   127.0.0.1:6379> memory usage us:105681
   (integer) 152
3. 但总内存占用 100M
   127.0.0.1:6379> info memory

Memory

used_memory:105388744 used_memory_human:100.51M 问题：按道理算 10000 * 152 = 1.449M 那为什么会占用 100M内存

测试了三种模式，都是占用这么多内存，希望得到回复

Memory

used_memory:105388744 used_memory_human:100.51M used_memory_rss:118505472 used_memory_rss_human:113.02M used_memory_peak:105447208 used_memory_peak_human:100.56M used_memory_peak_perc:99.94% used_memory_overhead:1407072 used_memory_startup:929088 used_memory_dataset:103981672 used_memory_dataset_perc:99.54% allocator_allocated:105497072 allocator_active:105697280 allocator_resident:109481984 total_system_memory:16573603840 total_system_memory_human:15.44G used_memory_lua:31744 used_memory_vm_eval:31744 used_memory_lua_human:31.00K used_memory_scripts_eval:0 number_of_cached_scripts:0 number_of_functions:0 number_of_libraries:0 used_memory_vm_functions:32768 used_memory_vm_total:64512 used_memory_vm_total_human:63.00K used_memory_functions:184 used_memory_scripts:184 used_memory_scripts_human:184B maxmemory:0 maxmemory_human:0B maxmemory_policy:noeviction allocator_frag_ratio:1.00 allocator_frag_bytes:200208 allocator_rss_ratio:1.04 allocator_rss_bytes:3784704 rss_overhead_ratio:1.08 rss_overhead_bytes:9023488 mem_fragmentation_ratio:1.12 mem_fragmentation_bytes:13137400 mem_not_counted_for_evict:0 mem_replication_backlog:0 mem_total_replication_buffers:0 mem_clients_slaves:0 mem_clients_normal:1928 mem_cluster_links:0 mem_aof_buffer:0 mem_allocator:jemalloc-5.3.0 active_defrag_running:0 lazyfree_pending_objects:0 lazyfreed_objects:0

1、请描述下你的redis版本
2、请测试前flushall redis，然后info memory看下内存初始大小，然后在灌入数据后，再次Info memory对比下
3、memory usage只是在算这个tairhash key相关的数据开销，info memory看到的是整个redis进程的内存开销（包括了redis自身元数据、各种Buffer以及内存碎片开销等）