efficiency of records and dictionaries

[Irene Langkilde-Geary]

Hi Mozart Hackers,

I first posted this question to the Users list, but I haven't received any answer.  I thought maybe it belonged better here on the hackers list.

How efficient in terms of speed is the lookup function in Mozart records and dictionaries?  I need a static hash table to store integer counts of English words, which means 40,000-400,000 entries, optimized for speedy lookup.  Would it be significantly faster to code it in C++?

Thanks in advance,

Irene