Review of DBA/Query for Windows
by
Shawn M. Gordon
President
S.M.Gordon & Associates
Introduction
With Client/Server data retrieval tools all the rage these days, I ran across an interesting product called DBA/Query from LeeTech software. DBA/Query is a Client/Server product that provides you with a set of tools for manipulating data and for quickly finding and displaying information found in ALLBASE/SQL and IMAGE/SQL databases. Although DBA/Query uses Structured Query Language (SQL) to access the databases, you don’t really need to be familiar with SQL to use it.
DBA/Query has two active software programs. The client runs on a standard PC that has a LAN or WAN connection to the host (in this case an HP 3000). The server runs on the HP 3000 that contains the databases that you wish to access. The client software communicates with the server using DBA/Query’s built in communications, no external, or add-on communications or terminal emulators are required. The client software is a standard MS Windows application, and has so special hardware requirements, other than the network connection mentioned previously.
So that’s what it is, and what it requires, now let’s get into using the software.
Features
Once you run the Client software, the first most important thing you will do is connect to the server. Figure 1 has an example of the server connection screen. Once you are all set up, you can now start working with the data. As mentioned in the opening, you can query data as well as update, insert (add), and delete rows. The ability to make physical changes to data is configurable, so you don’t have to worry about someone messing your database up.
DBA/Query uses a visual environment to walk you through each step of setting up a query and executing. First you will want to define what data you are going to do your retrieval against, figure 2 shows an example of how easy this is to do. This can be a single table, or it could be multiple tables linked together with a sophisticated join, figure 3 shows an example of a join screen. As you can see, it is quit simple to select tables and fields to link in the join. The only lack here that I noticed was that the software won’t try to guess on what the field will be for the join. It seems they could look for the same field name and offer that as the default for you.
Once these selections have been made, you now need select output columns (see figure 4). You can then specify selection criteria (the WHERE clause), row sort order (ORDER BY clause), data grouping (GROUP BY clause), and group criteria (HAVING clause). See figure 5 for a screen full of everything. It is important to note that all of these windows of criteria were build with Windows pick lists, by just clicking on each parameter that you want. Figure 6 has an example of the default data returned by the parameters in figure 5. That’s really all there is to selecting and reporting data.
DBA/Query doesn’t pretend to be a report writer, what they are trying to be is a data extraction tool. From this point you can save the data out in various formats and pull it into something like Lotus 123, or Microsoft Excel spreadsheets for doing graphing, or reporting. You may want to import the data into a database for more long term manipulation and reporting. You can do anything you want with it at this point. The only real drawback I see here is that a user could potentially manipulate the data on the Client side to scew a report to their advantage. Hopefully that won’t be a problem, but anytime you give a user the ability to manipulate data after it has been disconnected from it’s repository, it’s a little scary. I should mention that you can indeed print your query results in DBA/Query, but it’s a pretty straight forward print.
What’s nice is when you are selecting fields for retrieval, it is very easy to move them around in the pick list. Simply hi-lite them and use the up and down arrow to shuffle the field around. This feature is true in any portion that you are dealing with fields. You can also substitute a variable in a query so the user is prompted for it when the query runs. This keeps your from having to change the query every time the user runs it because you have a hard coded selection criteria like date or product code or something.
So what else can you do with DBA/Query? Well as I mentioned, you can also configure it so users can add, delete, and update data in the tables, figure 7 is a good screen showing the update ability. What is nice is the option to convert the data presentation from horizontal to vertical. Spreadsheets tend to present their data in a horizontal format, and databases tend to present in a vertical format, personally I like the vertical format better unless I am browsing data. In DBA/Query you can do both. Adding data is simply a matter of filling in the fields on the table, deleting data relies on you doing a selection, and then a deletion of the appropriate data.
There is a very neat option in DBA/Query, and that is the ability for the software to generate and evaluate the SQL statements that you built. This will show you where you are getting indexed reads, and where you are getting serial reads. This helps you to potentially optimize your query, and I thought it was pretty darn neat. I have never seen this option in a software package before. You also have the ability to enter in your own SQL statements in the Query Free Form window. This will do syntax checking for you on the client so you don’t wast the host CPU cycles on it. All and all a very robust SQL environment.
Usability (also installation)
The product is very simple to use once it is set up, the setup can be a bit of a challenge however, this is due to the nature of using PC’s on a network. The PC software uses the standard Windows SETUP program, and the HP host software is installed via the standard tape, restore job. Your biggest hurdle will be configuring the client network connection, making sure that you have all the files set up correctly with the right addresses
The ability to save queries is obviously a huge boon, this allows you to actually build queries for users if you don’t want them to even be doing that. The extended description field also makes it easy to fully describe a query, rather than having to do it in just eight characters.
Reliability
Well I reviewed LeeTech’s middleware, Client Server Foundation (CSF) recently, and that is what is used for DBA/Query to connect to the host. The CSF software is solid as a rock, and the middleware is usually your biggest concern when dealing with Client/Server reliability. The DBA/Query software itself seems to work effectivly as well. No GPFs, and no problems with lockups. You shouldn’t have any problem letting the end-users loose in this package.
Performance
Well the software performance seems to be very quick. As I have said before, this is hard to judge, since there is no real way to measure performance when you have a client application, accessing a middleware application, that is accessing a host application that sends the data back down. Which part is fast, and which part is slow, I don’t know, but the perceived response time in DBA/Query was very fast, so I guess that’s all that is really important.
Supportability (including Doc)
Support from LeeTech is generally very good, however at times it can be a bit tough to actually get a live person on the phone. The documentation is mercifully thin, but seems to cover everything that needs to be covered. My only real complain here is that the table of contents is kind of strange, and there is no index. From what I understand, new versions of the manual will be available before you read this review, so I doubt you will see what I saw.
Summary
In a field that is quickly becoming crowded, PC based data retrieval, DBA/Query seems to set itself apart in a few areas. The query analyzer is a nice tool for understanding what you are doing, as well as enhancing your ability to fine tune queries. The ability to change data on the host is both a convenience, and a little frightening, however, since you can configure what options are available to the user, it will be your own fault if damage is done to the data. I prefer to have the option, rather than not have the ability at all. I have a slight cosmetic complaint, and that is that some of the buttons used in the product are awfully garish , and a little hard to make heads or tales of. An emerging standard is to have a description of the buttons function display after you let the mouse rest on it for a second, this would have also been a nice feature. One last complaint, and that is it would be nice to have the software prompt you for connecting to the host if you haven’t already done it, when you try to perform a function.
DBA/Query isn’t exactly cheap, but it seems to be priced competitively for what it does. As more of us become proficient at writing Windows based code, and network software, you might be inclined to try to write this stuff yourself, but I doubt you could write it quick enough to make it cost effective. The fact that the middleware under DBA/Query is running on so many platforms, makes the odd’s very good that you will be able to maintain your investment in DBA/Query over a pretty long time. So the bottom line is this, the software works, it’s easy to use, and it’s fairly priced. If you have a need, or your users are bugging you for too many reports, check out DBA/Query.
At-a-Glance box
DBA/Query version A.01.00
LeeTech
20410 Town Center Ln. Suite 220
Cupertino, CA 95014
800-995-1987
Fax 408-253-4008
Call, write, or FAX to purchase or demo. The price of DBA/Query is $2,500 for Image and Allbase under MPE, and $1,000 for Allbase under UNIX. Each client connection is $99. Support is $250 per year for MPE, $100 for UNIX and $30 for each client. Oracle support will be available about the time you read this.