Re: ODBC - some observations

In article <55sflj$31t@reader1.reader.news.ozemail.net>, Robert Lore <rlore@ozemail.com.au> writes >Hello, I am posting this for people who are interested in using ODBC >for Informix. > >First some golden truths.. >1. Forget about Informix I-Net product. Its useless (sorry). It >has a large memory footprint, is difficult to set up and is slow >and is expensive. And also it only works with Informix! > Agreed. >2. For get about Intersolvs Q+E or whatever it is called. Its >bloody expensive and performance is slow and also uses lots of >memory. I could continue here but I tried this product a few years ago >and threw it out way back when we were in the evaluation stage. > >There are two products which I know of, both by English companies >(co-incidence ?). One is called SQL-Retreiver, by VisionWare, >(who also do a very good X-windows on Microsoft Windows), and >OpenLink by OpenLink Software. We use SQL-Retriever. I reviewed >OpenLink and decided against it. > We used SQL Retreiver at my last place and found it gave very very good performance. We were storing Word Documents as blobs on the server and gets performance which amazed everyone there. We had used other databases E.g. Oracle 7.?.0.16 and Informix/SQL-Retreiver blew my managers mind. He had could not believe the performance. After >10 years as an IT manager (>5 in document managment) he though he had seen it all. The only improvement I could see Informix doing would be to integreate the server portion of sat SQL_Retriever into the engine. I.e. have 'retriever' threads running on 'NET VPs'. The server portion of client/server being multi-threaded! -- David Williams