Re: Configuation of Front End.. choosing a database

If your running Access your database needs aren't very high, and expectations are quite well managed. The learning curve with Oracle or Informix is high, especially since the Windows 95 version of Informix is no longer available, even setting up the NT version requires some mastery of the concepts of User versus Root authority. Based on what little info i can see you might do well to learn first about UNIX through say Linux, running MySQL first on Windows or Linux, starting on Windows and "gradually" moving to Linux. BY and large all databases engines follow the same model: 1. a file server OS and fileserver host name 2. a database server PROCESS (-running program) and database server name 3. a choice of reusing the existing OS user accounts for authentication, or using an internal "database" set of accounts independent of the OS user accounts. 4. databases within the database concept.. terminology varies.. but in general the "datastore" encompasses all databases that the database server can serve over a network.. where serving can mean (A) running a simple SQL language query or (B) providing remote access to the SQL query interface. 5. a command line client tool like DOS "edit" that presents a full screen editor for (a.) accessing a local or remote database server (b.) creating and/or editing SQL question files (c.) submitting and receiving the question files and results. This last item is the subject of much axe grinding. Where the (isql, sqlplus, mysql) universal command line tool.. they all work very similar.. is easy to use cross database platform.. many people prefer a Windows wysiwyg (IMHO.. messy) buffer of "windows of windows" littering the screen, and drag and drop of the results to autoconvert the results into an native apps format.. ala' DDE, OLE, COM, CORBA, DCOM, SOAP Lesser used, these same tools will accept single commandline submissions and controls, via a single script file for completing a task.. VBA is the closest analogy. As per your question about how to speed things up, indexes or views generally tend to work.. at a cost. Databases store information efficiently, but also broker a users specialized view of that information.. the cost being generally disk space and up-to-dated-ness of the view... realtime actually costs real-hardware.. windows, no matter the vintage, can't simulate it and get you realtime stock quotes without the big iron. If a view is used often enough an index makes sense and reorganizes a snapshot of your data in a different configuration that serves it up quickly.. but usually at the expense of duplication.. and timeliness.. Starting with MySQL just makes common sense as it costs nothing, comes in Windows InstallShield varities and MyODBC drivers are available for windows apps to use.. not much different from Access.. in fact you can use Access to design the front ends. However, UNIX is more stable, fortunately MySQL is there too and not just as an afterthought port.. but a concurrent port.. identical feature set. Now MySQL does not have transactions or a long term support infrastructure hence it has its limitations, but most Access users don't know what a transaction even is.. so no problem there. As sophistication grows migrating to a cost effective database server like Informix would make sense.. its far more economical than Oracle.. and .. keep this under your hat.. but for the majority of data types.. all of the commercial database servers are "identical" feature wise. Training-wise you loose nothing at this low level of understanding either.. SQL is SQL.. VB is VB as it were.. And VC is VC.. only when you start getting to the level of depending on a specialized function API.. as in the MFC with VC.. do you pass the point of no return.. and by then your the size of Computer Associates.. or history. Last time I looked, Oracle was monolithic.. the clunk.. and hence had severe growth problems.. unless you shelled out big bucks.. it had to be that way to support all of the features you don't normally use. Informix on the other hand went modular, rather like the Borg.. and spread out its tools so you only learned, used, or had to deal with the few pieces that you actually "used" at a time. Oracle costs a lot.. but is no different than any other choice at the low end.. same for Informix.. same for MySQL, PostgreSQL, or .. Access... opinion for thought .. At 11:00 AM 9/26/00 +0000, you wrote: > > > > >I'm trying to build a front end for my IDS 7.23 running on SCO UNIX. I have >tried using MSAccess with ODBC connection and find that many queries and >reports are very slow. I want to leave the data on the server as many of >the clients will be accessing the database with just 56K dialup. >Unfortunately access pulls all the data across the network before running >the query. I have also tried some cognos tools such as impromptu. This >product is good but the learning curve is high for most of the clients. >Basically, I'm looking for a product that may be able to help me. If anyone >has any suggestions please let me know. >Thanks, >Jerry > > > > > > =================================================== John Willis Phone: 936.788.7823 Web : www.txucom.com Email: john.willis@txucom.net ===================================================