I'm not aware that there is a way to do this, but I'm fairly certain that it wouldn't make a big difference. HTTP connections (in contrast to
JDBC connection) are not kept open -I don't think HTTP keep-alive applies to this situation-, so the savings you get by using a pool is mainly the object instantiation. That used to be a costly operation, but JVMs have gotten much better at it. At any rate, in most if not all circumstances the time needed for instantiation pales in comparison to the time used for making the actual network connection, so I wouldn't think that this is worthwhile persuing.