Re: Standard Engine and OnLine - Which one and When?
>Subject: Standard Engine and OnLine - Which one and When? >X-Informix-List-Id: <newsgate.623> >From: cs_e266@kingston.ac.uk (Edwin Yau) >Message-ID: <1992Jan14.134405.8825@kingston.ac.uk> >Date: 14 Jan 92 13:44:05 GMT >Sender: phyd@cix.compulink.co.uk > > Standard Engine and OnLine - Which one? - and When? > >We're a small software house based in the south of England developing >Informix-SQL and 4GL based systems. Recently we've been hearing >more about the OnLine product and have had a few customers who want to know >more about the difference between it an the Informix `Standard Engine' (SE). > >We know that OnLine can handle Binary Large OBjects (BLOBs) >and that it moves faster but what is the primary difference between it and SE? >Is the underlying software the same or is OnLine the result of a >complete re-write? OnLine uses raw discs. SE doesn't. Internally, the RSAM used to manage the data in the database has a lot in common with C-ISAM (in particular, it uses B+Trees to organise the indexes), but there are all sorts of major differences too. OnLine is a complete rewrite. >It's difficult to know quite when to propose OnLine and Standard Engine >for a customer, besides the obvious Air-line booking systems and >Stock-Exchange applications it's difficult to know where to draw the line. > >Although Informix have been helpful in supplying a list of differences >between the two products I have not had a definitive answer on the situtations >that would demand either SE or OnLine. I was recently told SE would happily >support around 200 users (concurrently) given a decent processor (such as >a Sparc Station) to drive it. When to use OnLine: * 24 hours a day, 7 days a week access required. * Multi-database updates (2 Phase Commit and all that jazz. Version 5.0) * OnLine database archiving -- while the database is active. * Large numbers of users (>20 in my book). * Large databases (say >100k rows in largest table). * Performance When to use SE * When the client doesn't know much about Unix * When the database is small * When the database is lightly loaded * When the machine is small (286 PC) These are guidelines, and judgement is called for. >Also, has anybody used OnLine, or better still - has anyone used BOTH products? >If you have then did you notice any great improvement or difference between >them? > >Aside a little now, if OnLine supports BLOBs and SE doesn't then what's to >stop users of SE just adding a column into a table giving a filename pointer >to the disc storage of a BLOB and accessing it that way? It would only take >around twenty to thirty good C routines to drive it. Twenty to thirty? Read, Write, Open, Close, ... A Dozen at the outside. But more to the point, you cannot access an OnLine BLOB except via OnLine, because you don't know how OnLine organises its raw discs, which is completely and utterly different from a Unix file system. You cannot access OnLine database via the back door period. (OK: you *can* read the raw disk, and maybe you could even make sense of some of it, but if you wrote anything on the raw disk, you would be unsupported immediately and permanently and deservedly -- don't do it.) >Please e-mail me not at the above address but at - > > phyd@cix.compulink.co.uk > >Many thanks for your time and help, >Brian Leake. > >/-- Brian Leake, e-mail replies to : phyd@cix.compulink.co.uk ONLY please -\\ >| Tel: Home (+44) 276 23662 Work (+44) 734 776144 Fax (+44) 734 776055 | >\\------------- "No problems, Only Solutions" - K.Flynn, Tron --------------/ Yours sincerely, Jonathan Leffler (johnl@obelix.informix.com)