赵琦
2014-01-13 06:41:21 UTC
I'm using leveldb with large LRUcache in multithread environment.
Db is started in the main thread. Then I started a new thread and scan some
data into LRUCache in the thread using seek() and next() api. Without the
former thread exit, I do repeatelly starting a new thread and loading the
same data to LRUCache. The werid thing is the memory of the process never
stop increasing, and far exceeded the capcity of LRUCache i had set. BTW, I
had closed the mmap by modify src of leveldb. So mmap is not account for
ceaseless incremental of memory usage.
Can any one tell me what happend?
I have been bothered by this problem for weeks...
Db is started in the main thread. Then I started a new thread and scan some
data into LRUCache in the thread using seek() and next() api. Without the
former thread exit, I do repeatelly starting a new thread and loading the
same data to LRUCache. The werid thing is the memory of the process never
stop increasing, and far exceeded the capcity of LRUCache i had set. BTW, I
had closed the mmap by modify src of leveldb. So mmap is not account for
ceaseless incremental of memory usage.
Can any one tell me what happend?
I have been bothered by this problem for weeks...
--
You received this message because you are subscribed to the Google Groups "leveldb" group.
To unsubscribe from this group and stop receiving emails from it, send an email to leveldb+***@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
You received this message because you are subscribed to the Google Groups "leveldb" group.
To unsubscribe from this group and stop receiving emails from it, send an email to leveldb+***@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.