| Author |
 Topic  |
|
pdrg
Support Moderator
United Kingdom
2897 Posts |
 Posted - 25 January 2007 : 08:00:06
|
Rui, you're right, of course...and I am not impartial, I admit...
Back in the day, I was a private beta for SQL7, and managed to get a syntax enhancement, and was just wowed by how much better it was than everything else out there (with all the extra free stuff like self-tuning, full-text searches, etc). But I'd grown up with databases from dbase2 and all the incarnations through dbase3,4, foxpro for Dos, FoxPro 2.6, Access 2, 95 (a duff one), 97 (a goodun), 2000 (whatever) plus its variants (and the Rushmore optimisations stripped from the fox engine).
Then I heard a bunch of people screaming about how brilliant MySQL was back in v2-ish, so I thought I'd take a look, and it was naff. Then around v3-ish, and I thought I'd take a look, and it was naff. Then around v4-ish, and I took a look, and guess what... In the end I just considered they were 'crying wolf', the people who were wetting themselves with excitement were clearly people who'd never used a better dbengine.
Actually, I really hope MySQL does flourish and become useable and fast - it's never going to challenge MSSQL or Oracle, but certainly has a place in the world. That's purely because MySQL hasn't got the money and thousands of hardcore academic research hours into optimisations that the big boys have, not through poor will or intent.
Anyone interested in a more hardcore free, open DBEngine, Postgres is a belter http://www.postgresql.org/about/
So, whenever people hear me being grumpy about MySQL, I don't really mean to knock such magnificent efforts from so many people, I've just heard so much hype around it when there are some seriously better for-free and for-cash alternatives! |
 |
|
|
ruirib
Snitz Forums Admin
Portugal
26364 Posts |
Posted - 25 January 2007 : 08:37:59
|
Paddy,
I did feel the same way when I first looked at MySQL 3.x. I even spent some $50 on a book to learn how to use it and I remember when I read about the lack of referential integrity and the excuses made for that, even stating that MySQL was faster because of it (what did they expect!!!). I still feel a lot of contempt for MySQL as a result of those statements. I feel the need to use different engines just to get transactional support is absurd and getting a MySQL server optimized for anything other than the basic MyISAM format is very hard.
Anyway, for Snitz, I've learned that MySQL can be a performant option. Of course, painfully so at times, but it can handle the load of several tens of users gracefully. Without transactional support, of course, which make things like archiving errors a big issue (I've rewritten our archive routine for transactional support on SQL Server and it avoids quite a few common duplication issues).
Anyway, as the saying goes, you get what you pay for, but MySQL can be a valid option in certain circumstances, as long as you can cope with the features it lacks. And I won't even go into the issue of handling an upgrade option from Access, cause I know how many headaches this has given me in the past...
|
Snitz 3.4 Readme | Like the support? Support Snitz too |
|
|
|
AnonJr
Moderator
United States
5768 Posts |
Posted - 25 January 2007 : 09:17:34
|
Since we've got some experts here discussing the relative merits of DB Engines, let me ask this (slightly off-topic) question:
A while back there were some brief discussions about FirebirdSQL (based on the code for Boreland's InterBase). Curious I started to look at it, but to be quite honest I just don't have the knowledge to make an intelligent and meaningful analysis of its relative merits when compared to SQL Server, MySQL, and PostgreSQL, etc.
So how does FirebirdSQL fit in to this picture? What should I be looking at when evaluating a database system?
I know I generally prefer SQL Server, but that's mostly because that's what we worked with at FTCC so I'm familiar with it and tend to run into fewer problems. I know what I've read, but I'd like to be able to better evaluate what I read and be more objective in my reading. |
|
|
|
pdrg
Support Moderator
United Kingdom
2897 Posts |
Posted - 25 January 2007 : 09:59:02
|
Some of the most important things to look at when evaluating a dbengine are
*price
*support
*speed
*capability (can it do what you need? eg full-text searching on MySQL vs the same on Postgres)
*capacity (10 users or 10,000)
*experience (what do you already know?)
*maintainability (does it self-tune, for example, or will it run slow if not manually tweaked?)
Funnily enough, to my mind, price is often less important than experience, support and capability. Spending hours hammering against a problem, or having a slow-running db because the indexes are suboptimal is a false economy.
For an enterprise db, it'd be MSSQL for me, followed by postgres (both proven in multi-terabyte, massive-user situations)
for anything smaller, it's going to be a call on what you're familiar with. Again, my first would be MSSQL, or MSSQL Express as I've invested a lot in understanding the internal workings and the self-tuning means you need less DBA time.
As it happens, I was once at an MDSN MSSQL 2000 launch event where there was a particularly scruffy-looking beardy guy with sandals. He was an Oracle DBA (and playing to stereotypes, too!). He was religeously anti-MSSQL, so I had a good chat about why. He had doubts about its capacity (unfounded, when we checked), its speed (again, unfounded), reliability (again, unfounded) and eventually we got round to a level platform. I asked him how much time it took to tune his Oracle DB a day - and he said he only spent 20-30 mins a day on tuning, analysing optimiser stats, etc. Then we did the sums - he spent over 3 weeks a year doing what MSSQL does rather elegantly right out the box. Several grand a year (Oracle DBA's are expensive) pays for a lot of MSSQL licenses. Then we discussed the add-ons you have to buy to use Oracle - the full-text searching was another expensive attachment (free in MSSQL), and so on. In the end, we were both convinced, but agreed to keep quiet so he could keep his job! |
|
|
|
JohnC
Junior Member
215 Posts |
Posted - 26 January 2007 : 09:47:08
|
This is turning out to be a lot more confusing then I first thought... what happened to my topic?  |
|
|
|
AnonJr
Moderator
United States
5768 Posts |
Posted - 26 January 2007 : 09:59:10
|
Sorry... 
In summary, it looks like most of the replies are recommendations for either SQL Server Express 2005 or SQL Server 2000 since you have access to some good admin tools. |
|
|
|
Davio
Development Team Member
Jamaica
12217 Posts |
Posted - 26 January 2007 : 10:22:52
|
I use SQL Server Express to test the forum on local machine. It runs fine. I don't know the limitations to it, but I'm sure one of these fine MS SQL fans that posted before me have already mentioned that.  |
Support Snitz Forums
|
|
|
|
pdrg
Support Moderator
United Kingdom
2897 Posts |
Posted - 26 January 2007 : 12:21:37
|
Ahhh yes John - use MSSQL Express in my opinion.
hth |
|
|
|
JohnC
Junior Member
215 Posts |
Posted - 27 January 2007 : 13:21:19
|
Dare I ask for feedback on the best tool to use to migrate my Access DB to MS SQL Express?  |
|
|
|
ruirib
Snitz Forums Admin
Portugal
26364 Posts |
Posted - 27 January 2007 : 14:25:42
|
MS SQL Server Enterprise Manager / Management Studio. There is post on the way to migrate that you should follow carefully.
I don't know if the online tools are any good importing data. |
Snitz 3.4 Readme | Like the support? Support Snitz too |
|
|
|
TonyB7
Junior Member
USA
267 Posts |
Posted - 28 January 2007 : 14:35:47
|
You never said what the size limit is on that single MS SQL db is. With my current host, it's 50mb. After that, another 50mb will cost me $60/mo.
I'm in the same boat as you, with my Access db at 50mb and climbing, but since MYSQL is free and apparently limited by my disk space limits, I'm trying to convert to it (with no luck so far).
|
|
|
|
JohnC
Junior Member
215 Posts |
Posted - 30 January 2007 : 10:33:21
|
My hosting company has the following size limits:
MSSQL Express, MSSQL 2000 - 170 MB
MySQL 5 - 200 MB
At $60/mo extra, I would definitely be looking for a new host.
Even though everyone is saying go to MS SQL 2000 or Express, I'm also kind of leaning toward MySQL because the admin tools are much cheaper (free) and more readily available. I also can have 8 MySQL DB's with my current host.
I've searched Snitz and most everyone that wanted to migrate from Access to MySQL has either went off-line or went with another forum software. I'm still waiting for more folks to chime in here though. |
|
|
|
AnonJr
Moderator
United States
5768 Posts |
Posted - 30 January 2007 : 10:55:48
|
Last I checked, the Admin tools for MS SQL Server were free too, just not all hosts allowed access with them. Since it seems your host is allowing that, I don't see where the problem is.
As for the size bit, I'd ask some of those who are already on SQL Server how much space they are using in relation to how many topics/etc. they have to get a good rule of thumb for what your requirements are and may be in the next few years. |
|
|
|
pdrg
Support Moderator
United Kingdom
2897 Posts |
Posted - 30 January 2007 : 11:38:34
|
MSSQL Server WILL handle whatever load you throw at it, and scale just fine, import like a breeze etc, but you require a host who allows you to use it properly - some don't.
Maybe try the SQL Express upgrade, and help out all round (you can always delete it and go MySQL if you must) - there's a new thread here http://forum.snitz.com/forum/topic.asp?TOPIC_ID=63884 which you can help beta test the instructions :) |
|
|
|
JohnC
Junior Member
215 Posts |
Posted - 24 February 2007 : 17:44:51
|
| FYI: This topic continues in the topic that is linked in pdrg's post above. |
|
|
|
Topic |
|
|
|
Topic Locked
