Slashdot Mirror
IBM, MS Critique MySQL
magellan writes "InfoWorld has an article reporting how both IBM and Microsoft are dissing MySQL. While it is understandable from Microsoft, it is interesting that IBM, who often claims to be a defender of Open Source Software, would be so negative. Sun Microsystems and Yahoo are quoted as providing positive opinions on MySQL." On the credit site for MySQL, though, Bingo Foo writes "MySQL has finally answered its detractors who complained about its lack of transactions. A press release today reveals that InnoDB is now fully integrated with the stock MySQL product, allowing ACID-compliant transactions, rollback, and crash recovery. Let the religious wars begin!"
Except for the fact that IBM has a competing product. IBM can like open source all they want, but they would be stupid to promote something that does for free, what they sell a product to do.
This is an obvious attempt by IBM, Microsoft, and Sun to discredit open source while promoting closed source products. I find this very disturbing, especially in light of the fact that IBM and Sun are supposed to be open soure allies.
If you don't know what Zoo Blacklisting is, click here.
I can understand why IBM would bash MySQL, though. After all, they are selling DB2 -- so they have to compete with a free SQL db. Remember to keep in mind that both IBM and MS are primarily interested in keeping up revenues from their commercial DB products when reading their complaints on MySQL.
I don't suppose it ever occured to the submitter of the story that IBM might have taken an objective look at MySQL and formulated it's remarks based on first-hand observations, instead of the rehearsed and oft-repeated rhetoric of open-source fanaticism.
"Ask not what your country can do for you." --John F. Kennedy
I would not fault MySQL for this, though, since after all it was designed and still mainly used as an SQL wrapper for flat file data, and this is why it's usually much faster than full-featured RDBMSes. The problem is with mindless open-source advocates who try to pump up MySQL as the be-all, end-all database solution. For a personal website or small business, MySQL is more than adequate, but its lack of higher-end SQL features make it a poor fit for large, distributed, mission-critical corporate or university data storage.
IBM and Microsoft's customers are generally in this higher end of the database spectrum, where Oracle or DB2 makes much more sense. It's no surprise that they would want to put MySQL in its place as an entry-level database system, where it belongs, and I fail to see how this story qualifies as news.
If you look at the complaints in the article
Open-source databases "don't support as many users, they don't support as much data, and you don't have as many connectivity options," said Jeff Jones, director of strategy for data management solutions at IBM. "They lack some key functionality and lack the scalability and performance, which keeps them out of the enterprise," Jones said.
All of that is absolutely true. The MySQL response was also true the missing functionality in products such as MySQL is not needed by some companies
One of the posters below made the comparisons between a landrover and 747. You don't need a 747 for all tasks and there is nothing wrong with making cars and not planes. MySQL is a really nice alternative to both flatfiles and overkill complex databases for departmental servers. Oracle and DB2 still have major features for enterprise servers that MYSQL and Postgres just don't have. Especially DB2 where the advantages of I-OS and Z-OS over Unix can be felt. In some ways Oracle probably has the most to fear because with Sun backing MySQL there is a good chance that as far as Unix systems go within 5 years it could be comparable.
In terms of ease of use for departmental level servers SQL Server beats out MySQL. An open source project like Access would do a ton to close this gap.
Besides the FUD (both MS and IBM are sellers of massively overpowered SQL servers), there's good reason to warn customers about MySQL. It's still a toy -- it can't scale for shit, has trouble optimizing queries and setting up multiple indexes and transactions are a huge performance hit. This is the point at which someone mentions a good DBA is already optimizing the queries. I would like to point out that a company trying to avoid paying $2-$20,000 on a SQL server license don't have the $60k+ to pay a good DBA.
Some consultants are no doubt going to tout to companies the impressiveness of MySQL and hook them on it for its value without telling them that it's not as scalable. Which would mean costly conversion in the future to one of the other database systems, which could have been avoided by just using them in the first place.
I am not a fan of MS, but SQL Server is an impressive piece of software. I've dealt with it my entire career, while running mysql and postgres at home. I would never deliver a product based on MySQL to an F500, or any company that's going to do more with their database than manage a small ebusiness server.
Postgres, on the other hand, is very full featured and a joy to work with.
This isn't to say that MySQL doesn't work for your web log, your cd database, your employee info database or your company wide contact system with SOAP front end. It's to say that I wouldn't trust it with any data I needed 100% responsive and 100% reliable.
Hey freaks: now you're ju
a pocket knife is a lousy axe......
love is just extroverted narcissism
Comment removed based on user account deletion
"it is interesting that IBM, who often claims to be a defender of Open Source Software, would be so negative. "
Just because you support Open Source, doesn't(or shouldn't) mean you blindly like something just because it is open source.
Would you really want your multi-terabyte real time database to be MySQL?
The Kruger Dunning explains most post on
Let's not forget two things here:
:-)). Now if they start dissing PostgreSQL, which I stake my job and reputation on the reliability of, then I will begin to reel off the reasons why I parted ways with IBM, and would never go back...
1) IBM sells a very capable (as in, "Runs the UPS package-tracking system, at 15TB the largest publicly disclosed OLTP database in the world") system called DB2, and they make money doing so.
2) MySQL has only recently included transactions in the base package. They still do not handle subselects or foreign keys, both of which become very useful when dealing with large databases.
Why on earth *wouldn't* IBM recommend against MySQL for their enterprise customers? IGS does not service the sorts of customers that are typically suited to using MySQL (US Census Department excluded
Hint: it's not because IGS technical people are anything less than world-class. Management is another story. But don't think IBM engineers don't know what they're doing. They're damn good.
Remember that what's inside of you doesn't matter because nobody can see it.
And they were not wrong at all. At this time, DB2/6000 was missing some exotic features Oracle was having. Anyway, some peoples decided to adopt DB2 and IBM continued to improve DB2.
I don't see the point about a lack of endorsement of OSS by IBM because they are just saying some features, they think are required by enterprises class architectures, are missing by MySQL. It's just plain truth, MySQL is missing some features. It's up to the customer to decide if these features are required or not.
In short, a storm in a glass of water...
Achille Talon
Hop!
>IBM has made a ton of money on databases
That's about right. DB2 is an integral part of IBM's solutions platforms, and they make a lot of money on that software. You'll never hear an IBM "consultant" say, "Oh yeah, to reduce cost, I recommend this free database technology over buying 250 seats for our DB2 platform." Three reasons:
1) Licensing fees for DB2 are a significant source of revenue.
2) Professional installation of DB2 on IBM hardware is a significant source of revenue.
3) Custom integration of DB2 is a significant source of revenue.
IBM can make extremely good arguments about why DB2 is technologically superior product to MySQL, but it's extremely difficult to argue its cost-effectiveness, which they never did in that article (that I could see).
Besides, when you're talking about providing solutions, open-source often fights uphill battles against the "more expensive is better" and "we must spend our entire budget" attitudes.
-AAAWalrus
Back about a year ago /. experienced a major outage. Roblimo at the time noted, "By 7 a.m. it was obvious that this was not a typical, easily-fixed, reboot-the-database problem."
Can anybody imagine an SQL Server or DB/2 customer being satisfied with that solution? That's what IBM/MS is saying.
Of course IBM and MS will bash Mysql, it's to be expected.
But before you think that the bashing is solely due to marking ploys, think again. Mysql IS lacking many of the features that "real" databases have. In the past MySQL was unabashadly vocal in not including these features as they could be programmed around, and would slow down the database.
Remember, we're not talking about the latest SQL extensions, but common things like FOREIGN KEY, and stored procedures.
MySQL rocks when you don't need the extras, but that's not a reason to migrate to MySQL, or to ignore that MySQL lacks features common to general-purpose databases. Just because MySQL is my favorite "quick and dirty" db, does not make it the best tool for all jobs. That's like arguing that MSAccess is most popular db (by having the most installations) therefore it's the most technically advanced/secure/useful/etc.
Microsoft (surprisingly) nor IBM seemed terribly negative on OSS databases, including MySQL.
What I came away with was that they think databases like MySQL don't have some key features that are important to enterprises like supporting a massive number of concurrent users or a price tag with a non-zero integer followed by lots of zeros (US$).
Which sorta implies MySQL, Postgres, etc. are great for the other 99.9% of database applications.
We have one quote on MySQL from IBM, from Jeff Jones. In the end, he's saying the MySQL isn't an enterprise level product because of some missing functionality, and scalability and performance.
Considering that I've seen much harsher comments on MySQL here at /. during the database wars, this doesn't seem that off base.
It's the Microsoftie, Sheryl Tullis, who describes MySQL as a niche player and then disses open-source in general with her "you're not going to get a platform that's as reliable or scalable or as secure as what you're going to get with a leading vendor".
This is typical MS FUD, and should be dismissed as worthless. Since the "leading vendor" is MS, and since their products are NOT reliable, scalable or secure, her commentary is laughable.
It's also quite possible that MySQL isn't the answer to every customers wants and needs. Because it's open source doesn't make it the holy grail of DB products.
Companies that produce those products spend many years and a lot of money optimizing their products.
What I really dislike about slashdot is that every time somebody even takes a quip at an open source project it's dismissed as FUD (a word which gets thrown around so often that its meaning isn't even clear anymore).
Has anybody ever stopped to consider that many, many closed source projects are just better than their open source counterparts? Open source doesn't automatically mean better (hell, in my experience, open source usually implies an inferior product).
I'd choose IBM DB2 and MS SQL Server over MySQL any day of the week.
To get a little off topic I'd also choose MS Office over kOffice, Photoshop over GIMP, etc...
It's not that all open source projects are bad, it's just that a lot of them don't even begin to compare to the software that they are trying to emulate.
"we must spend our entire budget"
Where can I find such a company?
That's a common dilema in government and other pissing contest situations where managers are more interested in defending their turf than providing value to their employers. It comes from the situation where, if you didn't spend all of last year's budget, you obviously didn't need it all, leading to a cut in the following year. And we can't have that, now, can we?!!
IBM *is* exactly right about MySQL, and for that matter most Open Source databases. It takes years to mature a major product like AIX and DB2, and the GPL competitors (which I love and use daily) do *not* have the same functionality. They are *lite* versions of the real deal. The two most important features of Open Source products is they are 1) Free, 2) come with source code. It is *not* their functionality!!!
IBM and Borland will do *allot* to improve these GPL products and all the grousing by people with little to no expereince in the *real world* won't change that.
sight of the fact that MySQL was never intended as an "Enterprise" DBMS?
Look at the very top of the O'Reilly book. What does it say there?
"Databases for Moderate-Sized Organizations & Websites"
Please note that it doesn't say:
"This is a free product that kicks Oracle's ass"
It is explicitly intended to be, and I quote:
"Inexpensive, lightweight and fast."
To accomplish this they restrict themselves to a subset of the SQL language.
Why do you think that *adjustable* wrenches come in different sizes? If they're adjustable wouldn't you just get the biggest one and use it for everything? That philosophy might seem like a good idea, until you try to turn a 6mm *aluminum* nut with a 14" wrench!
It's OK for tools to come in different sizes and types. Pick (are you ready for it?) the *right tool for the right job.*
Does MySQL suck? For many particular jobs, sure, but that's the fault of the person who attempted to use it for those jobs. Conversely, there are situations and jobs for which it is everything *but* MySQL that sucks.
I just don't get the *one true DBMS* holy wars. Diversity be good. Monolithism be bad. Get with the program.
KFG
I agree that MySQL is perfectly suitable for many tasks, but I am not sure I agree that it is "much easier to use than fully-grown RDBMSs." SQL Server has a very easy-to-understand interface for creating a database in Enterprise Manager.
And in any case, dismissing all criticisms as anti-OS propaganda is not constructive. The Open Source movement does not have a future if its adherent cop a "The Emperor Cannot Be Naked" attitude.
If MySQL wants to make serious inroads and be considered a decent database for business it needs things that other primetime databases has like:
Stored Procedures
A good gui (MyCC is a good start - not a web one)
Functions
Replication
I know most of these things are in development or are in beta but lets get moving folks. Sure MySQL is fast and all but it can't hold a candle to MS SQL Server for ease of use and features. It's free which is great but it really can't compete in business yet. From that context it deserves to get dissed. Both dissers, by the way, sell their own databases so don't be too surprised by the negative press.
Have a Happy.
Let's start by pulling our heads from our asses and think about this subject from an intelligent viewpoint. MySQL is NOT the right tool for every freakin' site. I like MySQL for my little sites that get like 100 hits a day and store very little data. MySQL sucks major ass when you attempt any enterprise application use. It fails horribly with large tables and is extremely difficult to manage. Postgresql is the best Open source database, period. It is fast, reliable, and has all the features you will need and is rock solid.
Use MySQL for your little web site and use Postgresql for the real stuff. As for DB2, Oracle, MSSQL, I don't see the value in any of these over Postgresql.
LoRider
A database that can only handle one subset of our customers per database instance is too amateurish to consider for much beyond managing a Christmas card list. And, come to think of it, without Unicode it couldn't even handle my Christmas card list.
I can imagine some niche uses, but I would never consider it for a general-purpose database platform for a company with international aspirations.
"Those who have never entered upon scientific pursuits know not a tithe of the poetry by which they are surrounded."
Doesn't IBM realize that, by being a proponent of open source, they're obligated to say only nice things about every single open source project under the sun?
Doesn't anyone see the irony in the slashdot blurb? "It's surprising to see IBM diss MySQL; In other news, MySQL just got transactions!"
MySQL is a fine departmental database, but a lousy enterprise database. That will probably change, of course, but for the time being both Microsoft and IBM are right.
Cheers
-b
Fact is, BDB!=MySQL!=PostgreSQL!=DB2!=Oracle
BDB (Berkeley DataBase) is a simple, DBM-style database that only has key/value pairs. Notwithstanding this simplicity, a company (Sleepy Cat) has been making a profit with this product! It's used extensively by OpenLDAP, which is an "enterprise ready" application, capable of scaling to handle every single person or thing on the face of the earth today.
MySQL is not "under par" or "substandard", it is written to perform simple queries rapidly.
PostgreSQL is not "a toy", it's designed to be a feature-complete, modest SQL engine, with features over performance. (Though performance gains of recent have been quite staggering)
Oracle is the "nut buster" of a database. Based on code now some 20 years old, they've had the time (and the money) to make a truly upwardly scalable application. For those to whom the tens of thousands of dollars price is not a problem, Oracle is it.
For those who want high performance and database replication for simple databases for cheap, MySQL is it.
For those who need to build complex datastructures and access them on a budget, PostgreSQL is the one.
For those who want a very simple values-container, BDB is what you want.
The scale is not linear, with "bad" on one side, and "good" on the other.
I would not even consider BDB for most of my mid-tier web-based software. Nor would I consider Oracle. Postgres fits just about perfectly - I need transactions, and frequently have to perform nested outer joins and subselects in a single statement.
On the other hand, the LDAP-based network I manage runs just fine on BDB, and one of my recent projects (a large database of registration information) works best on MySQL.
Which is better - a sledgehammer or a screwdriver? They're both tools that get a job done. Don't call a screwdriver "deficient" because you wouldn't want to crush a brick wall, and don't call a sledgehammer a "Piece of Sh--" because you can't turn a screw with it!
I have no problem with your religion until you decide it's reason to deprive others of the truth.
it is interesting that IBM, who often claims to be a defender of Open Source Software, would be so negative.
I know a little (and nto much more) about databases, being more a sysadmin than DBA.
IBM's claim seem reasonable - that MySQL isn't suited for extreme high end use. This seems reasonable. Just because IBM chooses to advocate quality Open Source tools where they see fit doesn't mean that IBM must think every Open Source app is quality.
Maybe IBM just used MySQL and found it lacking.
Then why is it that at least once a month I end up getting the static page instead of the (MySQL-backended) dynamically generated one?
It is obviously possible to use MySQL to backend a site such as Slashdot, since they're clearly doing it, but that doesn't mean that MySQL is better suited to the task than any of the alternatives (free or otherwise). You can compensate for a lot of software deficiencies with sufficient sweat and a bit of hardware. I know, I've done it.
And despite Slashdot's success with MySQL, Sourceforge (owned by the very same people that own Slashdot, even!) was essentially forced to convert to PostgreSQL due to the limitations of this very same engine that Slashdot successfully uses.
The real question isn't whether or not MySQL can function adequately for Slashdot right now. It clearly can, with sufficient effort (and don't mistake me: other, more powerful database engines would require effort, too, but I suspect it wouldn't be quite as much). The real question is whether or not MySQL will continue to function adequately for Slashdot in light of Sourceforge's experience with the very same product.
My bet is that it will, but only because of the amount of effort or money they're putting into keeping it going.
It's possible that PostgreSQL wouldn't be able to handle acting as the backend for Slashdot, but given the experience Sourceforge has had with it, I doubt it would be much of a problem. The one thing lacking in PostgreSQL right now, replication, is the one thing they'd have to throw resources/money into. Otherwise it's more capable and faster under load than MySQL, at least if Tim Purdue's experience with Sourceforge is any indication -- so it should prove less expensive to work with in the long run.
Nobody's questioning the possibility that you can run a large, heavily-loaded site under MySQL. They're merely questioning the wisdom of doing so given the alternatives available today (there's no question that the Slashdot guys made the right choice initially).
Even so, there's a lot to be said for getting in there and actually deploying something. My experience with MySQL is that it's very limited in what it can handle, but that experience is based on the MyISAM backend, not the InnoDB one. Slashdot is using the latter, and that probably makes all the difference.
There's no way you'd get me to argue that Slashdot should switch over to PostgreSQL without a lot of testing first. Chances are, it would be more trouble than it's worth, but that's only for now. Experience shows that requirements change over time. Slashdot may eventually need exactly what PostgreSQL's got. So one can only hope that MySQL keeps up with their needs.
I'd still be very interested in how well PostgreSQL would work (compared with MySQL + InnoDB) as a Slashdot backend...
Use 'slashdot stuff' in the subject line in any email you send me if you want to get past the spam filter.
Open-source databases "don't support as many users, they don't support as much data, and you don't have as many connectivity options," said Jeff Jones, director of strategy for data management solutions at IBM. "They lack some key functionality and lack the scalability and performance, which keeps them out of the enterprise," Jones said. That's absolutely true: MySQL is well-suited to the needs of small and middle sized businesses (say search engines and invoice databases), but not for the biggest enterprise-level tasks out there (say for the IRS or Social Security). "So far, I still see MySQL and some of the other open-source databases as really niche players," said Sheryl Tullis, product manager for the SQL Server database at Microsoft. Now that's just wrong. The number of companies that can use a middle sized database greatly outnumber those that need big iron. Oracle, IBM, and so on are the "niche players" here. Sure they make money hand over fist, but it's for support contracts and so on for fewer (but massive) clients. Companies will pick the software that best meets their needs. Open source is cheap, easy, and supports the majority of cases. Big iron databases are powerful but expensive to buy and maintain, and only those who can afford them will use them. In this case, Microsoft is simply trying to convince people that bigger (and more expensive) is always better.
Matt Slot / Bitwise Operator / Ambrosia Software, Inc.