Your performance issues probably have more to do with the algorithms you implement than with the operations you use.
Posting the code might be useful. Telling us something about what you're trying to do, and what algorithm you're using would also help. As it is, your question doesn't provide enough information for anyone to give you a helpful answer.
Other people recommended gprof - I second that, if you're interested in profiling your code. I've also used VTune before, and liked it. But first make sure you understand your code and what it does, and that the algorithm you're implementing is time efficient when dealing with the size of data you expect it to handle.
As an aside, using C doesn't mean that your code would automatically run faster. I/O bound code would see no performance improvement, typically. UI heavy code might not benefit from using a low level language. Typically, C is a better implementation language where you need low level access, when interfacing with hardware or low level operating system services, or if you have very specific and stringent performance requirements that would be difficult to meet in a high level, garbage collected language. Or if you happen to like C, but that's obviously a subjective matter.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…