Right now, the library is thread-safe (supposed to be) and it does this by creating new requests each time one is needed per thread. Speed comes through the many threads.
Requests can be pooled and batched and then collected. This might increase the speed quite significantly. There is a decent test-framework for this piece now which will help us.
Right now, the library is thread-safe (supposed to be) and it does this by creating new requests each time one is needed per thread. Speed comes through the many threads.
Requests can be pooled and batched and then collected. This might increase the speed quite significantly. There is a decent test-framework for this piece now which will help us.