Because your kernel is general-purpose and has been tuned with certain balance between latency, throughput, and memory usage (memory management takes memory!) If your application deals with things < 4kb in size, has different performance needs &c, then it may achieve higher performance than just leaving everything up to the kernel.
Actually, MongoDB does leave memory management to the kernel by using MMAP to access data files. Most kernels will allocate a lot of memory to the disk cache, which can make it look like mongo is eating all of your ram.
