On Sun, 17 Jun 2007, Iain Wade wrote: > With tugela/BDB, when you ask to fetch a key it > single-threaded/synchronously fetches the data from it's store (fast > if in cache, slower if disk) into a freshly malloc'd buffer where it > is handed off to the same event based flushing code as in memcached.
So basically you block. Gross. > In my view, the blocking nature of the item access could be directly > compared to accesing a memory page on a non-mlock'd or over-committed > memcache instance that would need to be paged in from swap. Running a memcached instance on a machine without enough memory and swapping is terribly broken too, just like Tugela. Unless Tugela can do BDB API calls in other thread, it's not suitable for mainline inclusion. I would, however, be interested in a set of patches to make the storage system pluggable. An updated (threaded) Tugela could then later implement that interface. - Brad
