Slashdot Mirror
Should Dual Cores Require Dual Licenses?
sebFlyte writes "The multi-core debate continues. HP and Intel have laid into Oracle and (to a lesser extent) BEA over their their treatment of multi-core processers. Oracle's argument that 'a core is a CPU and therefore you should pay us all your money' isn't a popular one, it would seem. What does Oracle's stubbornness imply for the industry as a whole, with multicore chips coming to the fore so strongly?"
What about licensing per machine instead of per CPU core?
I'm not paying for any "processers"!
Oracle's stubborness says, time to start looking at DB2.
I was thinking of the immortal words of Socrates, who said: "I drank what?" - Chris Knight (Val Kilmer)- Real Genius
So people will move to competition if the competition is more cost effective for them.
I thought that they just turned you upside down and saw how much money fell out of your pockets.
I'm kinda torn on this one...
on one hand, a person with a dual core chip is likely to get slightly better performance than 2 actual chips.
on the other, if everything goes to dual core, then we've just handed Oracle, MS, et al. double (or more?) profits on their products. Support costs will remain somewhat constant, so wtf?
I dunno... it's a hard nut to crack
Dual core chips are sold in the "CPU" section of stores I'm going to consider them singular.
Central Processing Unit.
Theres no 's' on the end.
"Get The MOney, Get The Fucking Money"
"brxref
HP and Intel should manage their own business, and leave Oracle to mismanage theirs.
What have we come to that companies write open letters to themselves, using public opinion to try to damage competitors or enhance their own position... and the public eats it up and supports it?
Intel, this is your problem. Deal with it without whining to the public... or you'll look like whiners. It isn't like the wining is going to actually help your case anyway.
Microsoft of all people did the right thing.. why can't Oracle?
nothing like billionairs sitting around trying to figure out how to charge even more money. We need OPENSOURCE HARDWARE!!!
Let them be stupid...the market will correct them.
--Mike--
I wonder what happens if/when many, many nodes of a cluster could/would be considered "a core".
Obviously Oracle is going to price their product in whatever way will earn them the most money. Their pricing schemes are very opaque. The customer really doesn't know how much they are going to be charged until they see the price quote from Oracle. I think this is going to hurt them much more in the end.
I feel that Oracle flexes semantics to justify greater fees. A two core system uses the same amount of Oracle's software and a one core system does.
> What does Oracle's stubbornness imply for the
> industry as a whole, with multicore chips coming
> to the fore so strongly?"
PostgreSQL is coming along nicely...
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
i just hope that isp's dont start charging double if you have multiple computers connected to the same connection. just like the software, your not paying per processor, its buy machine.
This is sort of scam is used on pricing for mainframes all the time. One place where I worked used this as an excuse to (finally!) dump some crappy and archaic Computer Associates products when they started charging us double for a dual processor, even though one processor was partitioned to another OS that didn't run any of their products.
"I'm a scientist! I don't think, I observe!" - Dr. Clayton Forrester
I've never understood why licensing for some software is based on the number of processors. Why not base it on the number of installs? One installation, one license, period.
The company that I work for has never had that policy. We have products for AIX, Linux and Solaris; while we charge per processor, it's never been our policy to charge per core. We had to tweak things recently for our Linux products to understand about multi-core processors. Before we did that, we'd issue the users licenses that would be double the number of processors if they were using hyperthreading.
If you don't use the processor as the defining quality of "one computer" for the purpose of licensing, then what do you use?
For ages, for the purpose of licensing "one computer" has meant "a box with one cpu in it"; now that we have dual cores what qualityshould the license writers use to distinguish between a license for one computer, and (say) a site license?
I charge on a per logic gate basis.
.. but hey .. who says i have to care about that?
I was going to charge on a per transistor basis but decided against it.
Yeah I realize I wont be utilizing all the logic gates per transaction
Vendors charge what the market will bear. Buyers pay the least they can for value. Charging per-processor, or any other basis, is just a way to negotiate prices without saying "how much have you got?", which would make the buyer more resistant. It's arbitrary, except as a way of measuring buyer's willingness to pay. Trying to derive finer-grained sense from per-processor licenses to per-core licenses is treating the price model with more respect than it deserves, so no wonder it breaks down quickly.
--
make install -not war
I mean, it's like 1 1/2 cores, so how do they handle that?
Why can't Intel/HP make the mutipule cores seen as a single cpu to the os. Like hardware SMP. -Joey
with respect to Oracle: a tad of refactoring
I've always found Oracle's licensing to be pretty wrong-headed at every turn. You can sense that they really don't feel they need to compete on price, which is usually the ultimate undoing of an overly arrogant company.
My sense of things, though, is that to move from one database technology to another is a massive undertaking. You fight with these tools so much that you become an expert with them... warts and all... and even if someone else has a better and cheaper mouse-trap, mission-critical stuff just refuses to budge off the old workhorse.
The dual-core problem is just a new flavor of the Oracle licensing problem. It will be interesting to see if they budge.
David Whatley
If you really have to ask? The answer is ...
NO! Of course....
(unless you're Larry Ellison)
"When the president does it, that means it's not illegal." - Richard M. Nixon
I can live with dual-BSD or dual-GPL licenses. Won't slow me down a bit.
In the real world, while Oracle is a top player they are not alone. This is a perfect opportunity for DB2, Sybase and *shudder* Microsoft's SQL Server to make a serious play for marketshare.
That knocking sound you hear is opportunity.
-Charles
Learning HOW to think is more important than learning WHAT to think.
> So people will move to competition if the competition is more cost effective for them.
Hah, hah! This is sooo funny!
If the world worked like that, we'd all be running Linux and GPL software.
Do you have any idea how painful it is for enterprise users to change their database vendor?
Even upgrading to the latest version is a nightmare.
Just keep two copies of the GPL on your system.
Sheesh, evil *and* a jerk. -- Jade
Just imagine how much we'd have to pay for Cell's multiunit design...
It means Oracle are greedy bastards and they should be boycotted. Liscencing schemes have always been ridiculous, to base a sftware price on hardware is just completely silly. You might as well complain that some single cpus might outperform dual core (at different clock speeds of course) and pay more for a lower return in productivity. Then if you base price on hardware, make brackets on the number of operations possible in a given time frame. . . . charge by cpu cycle while you're at it.
Unless Oracle matches this policy, they run the risk of losing sales to Microsoft's SQLServer product.
I'm not usually an MS fanboy, but I'm rooting for them this time.
The language of future licensing is likely going to be per thread of main functionality running on a system at once. Multiple CPUs won't help you unless you can make the system branch out into multiple threads cooperatively. The marketing of such systems would be interesting too - a nice attractive initial price, but with extra costs multiplied by the resources you want to take advantage of.
Ryan Fenton
Let them.
I'll keep using postgresql and smirking.
Oracle's stubborness says, time to start looking at DB2.
Absolutely. But how many can easily switch?
For a long time I have had (occasionally heated) arguments with SQL addicts who insist that almost everything about an application should be coded in SQL and stored procedures. Meanwhile I have been moving all my logic away from the database engine, using APIs such as Java Data Objects, which makes my code very rapidly portable between databases. Now I am in a position to switch my code (and data) easily between different database vendors if there is a licensing or price issue.
I strongly believe we should start to think of databases simply as engines for storing and retrieving inter-related objects and not as platforms for writing applications.
If a vendor is going to charge per CPU, then they should charge per CPU whether the multiple processors are bonded into the same die, or assembled in multi-die packages, or in multiple packages.
Having run between 6 and 20-way SMP for ten years, I find per-CPU licenses distasteful and eschew products under such licenses. Such are market forces.
If a technological leap makes putting multiple processors into a machine suddenly affordable by having multiple cores on a die, that is a good thing. It will just force users to apply their force to the market that much sooner.
How does the Slashdot Effect happen given that no slashdotters ever RTFA?
Wait until Cell processors become the norm... when you have a process that runs around your network looking for resources to run on.... Oracle's sales reps are going to have a field day with that one!
Due to greed and stagnancy, Oracle has maybe 5 years left before the "smell of rot" is all pervasive. When MySQL and PostgreSQL become so common place (think Apache on the net today vs. Netscape's web server from the mid to late '90s), Oracle will be lucky to be a million dollar company.
If you doubt my words, think of what MySQL and PostgreSQL were just a year ago. Then think "What will they be like with 5 more YEARS of development?". Then realize that they are free to everyone and you'll see why Oracle is doomed.
Of course, Microsoft will claim it as their victory, but you, me and everyone else not running SQL Server will know better.
"To make a mistake is only human; to persist in a mistake is idiotic." Cicero
Don't patronize companies with licensing/pricing that you don't like.
Specifically, there are a number of alternatives for Oracle, both freeware and commercial. I haven't spec'ed Oracle for a client deployment in years thanks to DB2/Sybase/Postgresql.
Cheers,
Costs per use of a program should be based upon the users and not what it runs on. If somebody wants to run Oracle on a single 900 cpu box then they should pay X amount. If two users access the database at a time on a single CPU box it should be twice X (although sliding sacale for colume usage). More users == more strain and support variations than a single user on some uber nutter box. Maybe when it comes to support a small premium per cpu given the slight increased coding and as such more code to go wrong due to process interaction across CPU's, beyond that the bigist fault in any error on a computer is always the user. Lets face it without the user would there be an error (think tree falling in an empty forrest ;).
So no too per CPU costing and YES to per user costing.
If you have a car that has an electric and internal combustion engine do you have to pay double the tax or car parking fees? No that would be silly. Unfortunatly in the State of Oracle there is muchos sillyness, alas no Ministry of Silly Walks =(
When Microsoft licenses its OS to large universities, they have to buy a license NOT for every computer, but one for every person who could possibly use a computer. Thus universities still have to buy licenses for students and janitors who would never use a computer.
Could you imagine if TVs were sold that way: "Sorry, kids, you can't use the TV. I could only afford once license."
Sellers of software can put any asinine restriction they want on their products. Consumers are free to accept or to say no.
If someone says he and his monkey have nothing to hide, they almost certainly do.
and that includes the seat or user licensing schemes as well. Server based software should be based on a maximum transaction rate which would be more realistic.
"I feel that Oracle flexes semantics to justify greater fees. A two core system uses the same amount of Oracle's software and a one core system does."
There is a precedent in the mainframe world for what Oracle is doing. IBM charges per a CPU, and so do others. And after all a multi-core is basically two (or more) Processing Units* in one package.
*A computer is basically the "PU" in CPU. The "C" isn't really needed.
Why debate the interaction between groups of hypotheticals?
Practically nobody uses dual-core chips now, so it doesn't make any difference what oracle says right now, does it? If the chips start getting widespread use, Oracle can just turn around and change their minds at any time.
Intel says dual-cores are the future, and AMD says dal-cores are the future, but they have been wrong plenty of times in the past, and it's entirely possible that dual-cores will be as short-lived as the Itanium, and this wont be an issue for anyone in the first place. Intel and their partner (HP) just have their PR machine going full-tilt, to hype their none-existant, but supposedly forthcomming dual-core processors.
Bah! There are many, many things of real importance that could be discussed rather than this.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Switching to dual cores is a rewrite for most software. Basically software will have to save the chip manufacturer's bacon: the only reason dual core, multiple core, etc is gaining traction is because the chip manufacturers have hit a performance wall. Money is the only way to encourage software vendors to parellilise their software... so whining about it is counterproductive. While video encode and sound apps may see an improvement with dual core, don't expect most programs to easily.
Since I don't have to buy licenses for anything I run, why hell not yes double well why not even triple the price. While you are at it raise the cost of Windows by 200%, sounds really great to me.
Got Code?
The problem is that per CPU price is the same regardless of CPU performace. How about they run a publicly available (GPLed?) becnhmark which reflects all aspects of the system related to database (CPU, disk, memory, bus, etc.) and then base their price on that benchmark. That would seem fair as long as they revise pricing every year due to ever-improving system performance.
"You mortals are so obtuse." -Q
Does that mean I need to pay for four licenses of Donkey Kong on my new Sony game system?!?! Time to yell at my Congressman! :-)
On rare occasions, their secret desire to drive intel sales actually does help the rest of us out.
What does Oracle's stubbornness imply for the industry as a whole, with multicore chips coming to the fore so strongly?
I can't say about the whole thing... What I do know is that because Oracle inflexibility, high pricing and intrusive license-checking they will certainly lose clients on the long run.
And it's not just about multi-core processors...
Let me give an example:
I do work in a Federal University in Brazil, and we don't have exacly much money available.
Several months ago we bought a 4-CPU Sun E450 and we were going to pay for an Oracle license accordingly to that machine (a MHz-based license), it was just a matter of waiting the money to come for that.
In the meantime, Oracle decided to change the license so it's now based on the number of CPUs. When FINALLY the money arrived and we noticed the money wouldn't be enough anymore.
In the end we've got a 1-CPU license and we had to physically remove the other 3 CPUs from the machine.
Because of this and many other things (like a license-monitoring software from Oracle we HAD to install, as if we were some sort of criminals) we're now planning the migration to PostgreSQL and never again to use Oracle.
n/t
Quantum materiae materietur marmota monax si marmota monax materiam possit materiari?
Give me ten programmers and I'll put Oracle out of business.
"Are you absolutely sure? Because legally I am allowed to shake him by ankles and see what falls out. It's established in the case of Lawyers v. Justice...that was a wonderful day for us."
English is easier said than done.
From the article: What does Oracle's stubbornness imply for the industry as a whole, with multicore chips coming to the fore so strongly?
That IBM's DB2 and MS' MSSQL Server will get more customers than Oracle in the future.
Bye bye, Larry.
Oracle -> Postgresql
BEA -> JBoss, Geronimo
Saves you a bunch of money.
my brain is "dual-core," with a "left brain" and a "right brain". I hope the airlines don't find this out and double my ticket price.
No data, no cry
IMHO, Dual-Core should not infer an additional cost. The cost for per-cpu licensing should instead be based upon how many seperate chips are used. (ie. a dual-core dual-cpu rig would be priced as a two-cpu machine.)
I say this because dual-core is the natural evolution of moore's law. Has Oracle (or anyone else) ever charged you for having a faster processor? Or a processor with more advanced features, such as SSE or SIMD type instructions.
We're reaching the end of our abilities to push silicon based processors, and so dual-core is as natural an additional feature as SSE and the like. It doesn't make sense to charge extra for it, and doing so is just flat greedy. It's not like in a couple years there will be a single-core 10 ghz CPU available instead of that dual-core 6 ghz CPU.
Just my 2 cents worth. If anyone has a contrary thought, I'd be happy to hear it.
/dev/random
should 20 billion transistors require 20 billion licences?
Now Intel has a lot of nerve. Their architecture sucked so bad (due to register starvation) that they had to come up with this dual-core kludge to get past the brick wall that their performance hit. They market it to everyone as dual core, like two processors, twice the power etcetera, and then get all sanctimonious when companies that charge per CPU say "Well OK then, that's two CPUs!" Who didn't see that coming?
Methinks they're speaking out of both sides of their mouthpiece.
Everybody's a libertarian 'till their neighbour's becomes a crack house.
... OSS GNU/Linux
tell the ripoff corporate types like Oracle & MSFT to go eat a carrot
If Microsoft can license software based on the computer, the SO that control you keyboard need one license per key on your keyboard.
http://www.michel.eti.br
If it means more income they will charge for it.
Be glad they arent using the old IBM model where they charged per cycle used..
With some luck, as OSS solutions improve, these giants will just price themselves out of the market. Problem solved.
---- Booth was a patriot ----
Wouldn't do it for a database, but for most other stuff, we run many (sometimes 15+) vmware instances on each physical, 4-8 CPU box, and one our vendors was just caught off guard when they confirmed that the pr cpu licensing scheme they stupidly employ ment that we could install their software on each vmware instance, but only pay for 8 licenses... Bad for them, but good for us since they will be supporting 15 "installs". I think they are seriously going to reconsider their licensing in the future..
...why are you still alive? Get as much Oracle for as little money as you possible can.
Negotiate tenaciously. Period.
No.
"same amount of software".. wtf does that mean ? 1000 oracle installations run the same amount of software as a single machin, just different copies of it. And a dual core machine running two different instances may have two different copies in cache.
I dont see what all the fuss is about. The way software is licensed normally is weird.. this is no more weird than normal.
http://rareformnewmedia.com/
Sounds like they have a position.
I always thought that 'simultaneous warm bodies' was the fairest system...
Even in 2000 it was already very bad, with Oracle not so brilliantly (i.e. disastrously) introducing per-MHz pricing./ 001023hndb.html (see the insert towards the bottom of the page titled "Pricing becomes contentious")p oracle_1.html
Take a look at these urls:
http://www.infoworld.com/articles/hn/xml/00/10/23
and:
http://www.infoworld.com/article/02/04/04/020408a
and I quote:
"Meanwhile, Asseily added that the pricing issues surrounding Oracle's database leave lingering questions about its application to Oracle's application pricing.
"I have no faith they won't change the licensing strategy at a moment's notice because it has happened before," Asseily said."
How does Oracle charge for SPARC IV and POWER systems? Those are already Dual Core systems. If they're extending pricing from those established markets I see no foul on the play. If they are changing the pricing scheme then it would also extend to the 'Big Iron' shops with those current Dual Core designs wouldn't it?
What if it is just turtles all the way down?
seems like a transactions per second governor would be the best way to license database software.
It's an "Open Sores" gang rape!!
And it can hardly be argued that it's an issue of chip count, what if I were to take a dozen or more chips (PLAs, slice processors, and other exotic devices) and from these build up a single 386 class CPU? Clearly such a device would only require one license to run software, even though it was made of multiple chips. And since there are already court rulings that instruction sets can not be copyrighted, it is clearly my right to build such a device and software vendors would have no valid reason to keep me from legally buying copies of their software and running it on my creation.
One should also consider that my "single core" desktop computer actually contains at least two significant processors, the CPU and the graphics card (which may very well have more processing power than the CPU). While software like Oracle doesn't take advantage of the processing power of the graphics processor today, if some sophisticated user were to enhance his OS such that some improvements were made that could take some small advantage of the processing power of the graphics card, would this somehow change the processor count as far as Oracle was concerned?
If a 386 computer with a 387 co-processor counts as only one CPU, shouldn't I be able to designate one of two Athlon processors on my dual CPU motherboard as a "co-processor" and pay for only one machine? Sure, each of the Athlon processors is far more powerful that the 396 and 397 combined, but that's not the issue. And if chip count is the issue then the 386 and 387 certainly use as many or more chips (and more support chips).
I'm an American. I love this country and the freedoms that we used to have.
Very serious and documented article.
Beside the name of the "four year old son Tux," the most interesting part is "five more PlayStation 4 consoles".
Since Sony did not even show the PlayStation 3 yet, this is hot! I want one!
So people will move to competition if the competition is more cost effective for them.
Exactly (potentially)...
The original question was, "Should Dual Cores Require Dual Licenses?"
There is no should or shouldn't.
A contract is an agreement between two parties.
One sets forward their terms. The other agrees, steps away, or offers ammended terms for consideration. A license is essentially just a representation of that.
"Should" a dual core require dual licenses? There is no should. Oracle are allowed to consider it essential to them and for them to walk away if they don't get their way - and potential users are allowed to consider it too high a cost and walk away if they don't get their way too. Or they can come to an agreement.
Inevitably, one of three things happen:
Customers walk away, Oracle reconsiders its stance.
Customers suck it up, deciding it's still worth it, if less so. Oracle continues.
Oracle loses overall share but profits per customer are higher, thus they're willing to continue with fewer, more valuable customers.
From Oracle's perspective, why should customers halve their license fees by simply upgrading to dual cores? What happens in a few years when Intel has 8 core CPUs? Do they only get 1/8th revenues? As Oracle sees it, they're right.
From the customer's perspective, all they did was upgrade their hardware with a single piece. As they see it, they're right.
In the end, there's not really the notion of right or wrong. Just two different views. Ultimately, equilibrium will likely settle it somewhere in the middle.
Why charge more for multiple cpu's? They should just keep track of how many [computing power measure]'s that they use to run oracle and charge for that. Forget per machine/cpu licensing, charge for use! Generous discounts for heavy users! No more cheap performance increases from extra RAM!
Nerd rage is the funniest rage.
"What have we come to that companies write open letters to themselves, using public opinion to try to damage competitors or enhance their own position... and the public eats it up and supports it?"
Of course. Vigilante justice is best meted out by the court of public opinion. He who sways the court sways the judgement.
.. is that the licensor can define all the terms (subject to certain restrictions of law) as to how the licensee can use the licensor's property. They could say you "need a license for each left-handed dyslexic that uses our software" and they'd probably be within their legal right to do it.
Laws should not be written based on the assumptions of the writers as to how people will operate within the law... they should be written with the expectation that people will behave in the worst possible way they can within the law. In this way, the concept of "software licensing" is broken. It's too easy to abuse, and we cannot rely on "bandaid" legislation or peer pressure against big business, because neither is even remotely effecitve.
I work for the Department of Redundancy Department.
wow... 6 icons alongside the story... now that is a record :P
Veritas is as bad or worse on "Tiered pricing". In past Oracle was worse and they charged on potential CPU's. If you had a eight CPU server, but only four CPU's installed they still charged for eight CPU's.
This is what drove many Oracle users to Windows, because Intel based servers tend to be smaller.
Oracle came after the place I was working for being out of license by around a million dollars. After a long negotiations Oracle agreed to charge us per installed CPU. So after signing the agreement with started pulling CPU's and max'ing out RAM. We ended up only owing Oracle a few thousand, and maintained performance with the extra RAM.
Veritas NetBackup is the same thing. Explain to me why it cost more to backup a multi-CPU server.
Why not by number of users?
I do not care how many cores, cpus, sockets, etc.
If there is just one user, there should only be one license.
What am I missing here?
wake up and hold your nose
After all, if it's per chip and each core gives the performance equivalent to a single CPU then you will see chip makers cramming many cores onto a single chip.
It's almost like if a company charges per-seat licensing and a company then installs benches in the office as it counts as one seat. Of course that's just terminology.
Shh be very quiet Microsoft will want to do the same thing.
Q: Should Dual Cores Require Dual Licenses?
A: Hell yes they should. All linux kernels should download with at least two copies of the GPL to comply with thie requirement.
Honestly, Oracle can charge whatever they want. Let the market decide if it's a good idea. Gotta wonder what they are going to do when cell style processors start going into app/db servers though.
Davo -- Free speech, free software, AND free beer.
Say Intel or Sun develops a 32-core chip. Should that be licensed as one processor?
I can sort of see where Oracle, et al, are coming from. Not that I think Oracle's per-CPU pricing is completely fair, but I can see where they would want to protect their income. Right now they get a big chunk of money for multi-CPU systems, especially large scale server stuff. A 32 or 128-core chip would destroy them if it counted as one CPU.
Now, that's just how I see where Oracle is coming from. I could care less if they die or not, I don't care much for Ellison or the company.
The ratio of people to cake is too big
Many companies have been attempting to screw people over on this since Intel introduced Hyperthreading. /prco/cpuinfo finds two cpu's you need to license for two proc's. This is one reason a lot of places turn off HT, 'specially if the apps are IO bound. So the question: can you turn off one of the dual cores?
As long as cat
Who cares how many processors ? It's a minor engineering detail. You don't think the radiotelescope http://www.lofar.org/ with its 6144 chips and 12288 CPUS will be paying any per-processor license fees ?
Is processing speed really the most important part of a database? I'd have thought database (i.e. disk) size was a much greater concern, and a perfectly reasonable measure to base the tiered pricing on.
Although it isn't our decision to make, letting Oracle and others know where we stand will help them find a manageable licensing strategy.
With today's heavy per-processor trends in the database world, those at the middle of the market sometimes have to decide between one over priced bleeding-edge single processor system, or a more mature and stable dual processor system. Given no licensing pressure, they would buy the dual system.
Possibly in the next few years, dual core systems will see the same pressure. However, the current trends tend to make people buy an expnsive Intel system to avoid paying Oracle money. Dual cores are a problem for Intel because it will cause consumers to buy older technology instead of jumping on dual core systems. So, of course, Intel is complaining.
PS - this will neither help or hinder Oracle taking more of anybody's money. They will still be subject to market pressure on the overall cost of any given system. If they make you buy more licenses, then they'll each have to be cheaper.
And since there are already court rulings that instruction sets can not be copyrighted
But they can be patented.
In tomorrow's news, Oracle is charging extra for execution on pipelined processors, citing that their software is running on more than one execution unit at a time, and that the user should pay for this extra value their software provides.
This is trivial, but with open source software, there is nothing special for multi-core chips. So Oracle's position indirectly favors opensource software use on multi-processors. This is good for opensource developers.
To me this implies that Oracle's competitors are so far behind that Oracle can get away with charging multi-CPU licenses for dual-core chips.
why not just charge by the number of concurrant users? Or if anything performance related, the raw volume of data or the transaction count.
Oracle does not do more work on a faster cpu. That is like presuming that a sports car gets more miles than a minivan.
No company will double their transaction count by simply upgrading their db server. They need more staff and more customers as well!
If the cpu mattered at all on a RDBMS, it would only mean the same amount of work is done faster. However CPU speed is almost one of the most irrelevant factors in RDBMS performance and throughput.
This is just a lame attempt at grasping at a moral justification for price. Why not just say : we want more money, help us justify why we should get it.
No properly designed database will do twice as much work just because the CPU is dual core, or even 8 cores.
The CPU should not be the bottleneck!
Oracle may as well charge more based on how experienced the SYSDBA and designers are. These guys can make at least an order of magnitude (if not 2 orders) difference in performance.
No one has a right to their *own* opinion. They have a right to the TRUTH.
Only yesterday on /. there was a OS comparison test where MySQL was used as the benchmark. For commercial use where you aren't trying to sell it as a product, it's free, runs on as many cores as you like, the only cost being support (if you need it). If you can support the application yourself, then you don't have to pay a thing (except yourself). But people bellyached hard "oh I like this feature not found, or that obscure feature over there". Fine. No problem. If you really need that obscure feature that you and 4 other people on the planet need, then go ahead, enjoy paying $40,000US per processor ($80,000US for each dual core processor), for your obscure feature. If the boss gets sour over the price, maybe he can take it out of your salary. Have a real good day.
But what wouldn't we do for more money? Why not license software according to the number of pipeline stages too? Or MIPS? Or the hard drive size it's installed on? Imagination is the limit. :D
We had this problem on a Cisco product. The install came out and saw I had "4 CPU's running." It was in fact a dual-hyperthreaded server. I think I offered to open one up and show him what he was seeing. I don't think we had a problem but it shows that some are definitely trying to license this way.
Nowadays DBA groups manage change control, administer the applications running on the database (Peoplesoft, Oracle applications, SAP, etc), set standards and help developers use the database effectively. (+ more)
Usually when a development group asks me why they should move their database support to our group I only need to ask them a couple questions. Shortly after that they become quiet. Regardless of which database you'd like to have you need a REAL dba. Whether its Oracle, DB2, SQL*Server, mySql or postgreSQL. If you don't think so, your only fooling yourself.
As for hordes of overpriced DBA's, its the same as hordes of overpriced developers. Unfortunately in IT today there are tons of people willing to do the job but most of them are average at best. We find it very difficult to hire good DBA's.
"Thanks to the remote control I have the attention span of a gerbil."
Oracle's argument that 'a core is a CPU and therefore you should pay us all your money'
Translated...
HP and Intel: "Somebody set up us the core!"
Oracle: "All your money are belong to us!"
+1 Insightful, -1 Troll. What can I say, I'm an Insightful Troll.
First of all let me say that the question is somewhat silly and very vague. What do we mean 'should' there be a per CPU charge. Does this mean do you like this idea, well of course anyone who has to pay more will say no and anyone who gets money will say yes. Does this mean is it morally correct to charge this amount? Well thats alot like asking is it morally correct to promote a green DVD player rather than a blue one, perhaps there is some little moral overtone (if you do bad your company goes under and your employees get fired) but not much. So we are left with the question of whether this is a good buisness practice.
Personally, i don't see this as much differnt than making people pay for using the product based on processor speed. While it is a little weird I tend to think it is a good buisness practice. Just like with airlines by charging based on processor performance the company can charge those users who have the most resources the most for the product. Just like the crazy expensive last minute fairs target as buisness travel allow the airlines to sell tourists cheap tickets for vacation soo too does a performance based pricing plan allow a company like oracle to sell their product to normal consumers at a reasonable rate while still profiting overall.
This is a general problem for buisnesses that sell a product where the fixed cost overwhelms the marginal cost. It is in everyone's interest for the company to sell cheap units to those who can't afford anything more so long as the cheap units still cost more than the marginal cost (which is near 0 for software). However, if everyone can buy cheap units no one will buy expensive units to offset the large fixed cost (the cost of writting all that code).
So instead of just charging everyone alot and denying the small guy any chance to use the product a smart buisness figures out some way to divide the market. Airlines use weekend stays and when you order your ticket. Software companies use number and type of processors. This is just a refinement of that methodology which benefits both consumers and the producer.
If you liked this thought maybe you would find my blog nice too:
quote "What does Oracle's stubbornness imply for the industry as a whole, with multicore chips coming to the fore so strongly?" It means that there's a huge amount of cash to be made developing open source alternatives. The potential is there. The business model is there to be used. Note, I said used not abused. The huge gravy train days of uber profits with software are coming to a close,you can see that train coming around the bend, but every day meat and potatoes level money will always be there. You actually can sell and service open source. Even if it is freely given away you can still sell service to it and customizatise it and get paid to do that. Some companies are doing that now. That's what it means. If oracle wants to be dinks about it, or any other big company, someone else will step up to the plate. It won't happen overnight but it WILL happen. Software is no longer all that exotic, and we've gone from a few hundred people coding some decades ago to millions now, so industries have to adapt and adopt new business practices or go the way of the dodo. And I say let them.
This is like saying I need two licenses if I have two Word documents open at once. License agreements have always meant "per CPU" in the sense that a CPU is the main box of your computer. Why on earth is there even a question of "should there be multiple licenses for dual corse". No! It's still only ONE computer.
This is what really bothers me about software licenses. They continue to make the software more and more bloated, while charging people with better hardware, which they already payed extra for, have to pay extra to use the software. It's like being penalized for buying better hardware. I can just see it. Vendors will start charging double when they release and AMD64 version because that's twice the bits. Maybe they'll even charge extra for support on certain pieces of hardware like RAID, or SATA because it is faster. It's equivalent to charging owners of fuel efficient cars more because they can get more out of the fuel.
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Let them charge whatever they want. The massive companies with money to burn will still burn it.
The rest of us that would never shelled out for oracle anyways will keep on using postgresql, to our advantage.
This goes beyond simple enterprise databases. Look at spatial databases. In Canada, it costs roughly $50,000 plus $13,000 per year in maintenance fees for an ArcSDE / Oracle based spatial database license.
Or, it costs nothing but your time if you choose to make an equally powerful, easier to use spatial database using PostGIS.
So, you can buy your spatial database, or you can have a database plus (at least in the purchase year) pay for a dedicated person to play with it for you.
George Bush + Linux = "I will not let information get in the way of the fight against Windows"
Now I know why "per CPU", which seemed like slang, was in EULAs...it's a clever way to get MP and MC (oh god, glad i don't own a MP/MC system!) users to pay more.
Big business needs to learn they cant' keep sucking money out of all of us around every corner they can or we're all going to be too broke to afford their products (at which time they'll just "blame open source" or something).
Of course, if FOSS kills them legitimately, they can't sue for giving something away, now can they?
CAn'T CompreHend SARcaSm?
If the Powers That Be insist upon Oracle databases, that's their call; all they have to do is fork out whatever Oracle is demanding (whether it be per CPU, per core, per RAM chip, firstborn son, etc.) If the price Oracle demands is too high, they'll start looking at alternatives: DB2, PostgreSQL, SQL Server being the main ones that spring to mind.
Oracle needs to be careful that they don't price themselves out of the market. Because the simple fact of the matter is, there comes a point where people are simply unwilling to pay the price demanded, and at that point, you'll see a massive exodus to the competition. For some, that price is higher than others. Only the marketplace will determine what is, and is not, reasonable; and that's ultimately the way it should be.
ultrasparc 5 has 32 cores. How much would you pay now ?
I suppose this qualifies as news, but I guess I'm just kind of annoyed by the way it's handled. The copyright owner of software can define their licenses however they want. If Oracles wants to charge for two CPUs for use on dual cores, fine. That's certainly their right. Economics will handle the rest. If Oracle thinks they'll make more money from dual core licenses than they'll lose from people switching to other products, then it's a smart move and more power to them. If they lose more customers than it was worth, then they screwed up.
But to complain that it's unfair or what qualifies as two CPUs is competely irrelevant. It's Oracle's product and their choice how they license.
This reminds me of the Microsoft Windows NT (Dual Processor Edition) license agreement...Circa 1998 or so. They actually put out a special version for the early Pentium-Pro dual-processor systems. There was this thing in the license agreement. It said I was licensed to run it on a single computer, as defined by a single CPU, as defined by a single processor.
I asked MS for clarification, and got laughter.
Andy Out!
If you ignore a very small number of details, there's not even much difference between multi-core and multi-pipeline, which has been around for a very long time, and has been tweaked around in so many ways that it's very difficult to put an integer figure on it for most chips. Athlons, for example have 3 units each for int math, fp math, and memory ops, but bogomips comes out to 2 X clock speed.
From Oracle's perspective, they're charging based on how much someone could actually use their product. Instead of lab-testing every customer's hardware configuration, they charge per processor, because it's a decent heuristic, or at least it has been for a long time. They could get an even better approximation by charging by some multiple of the square root of total on-die cache, but then people would complain that it's too complicated.
The bottom line is that Oracle decided to charge based on performance, and is no more able to reach a consensus on how to measure it than anyone else in the industry is.
WARNING: there is a trojan on your
Oracle tried to charge us a per user license for a EMAIL SERVER. They wanted around a 1.50 per user for a few million users. Crazy...
They didnt want to honor the site license we bought back in the day, after they found what we had in plan.
Licensing gets even worse in corporate companies, after a project is bought and paid for in 1 group, its handed off to another to run, and keep licenses up to date. I've seen the operations group buy licenses they didnt need, and you cant tell them they made a mistake, because its not your department anymore, and your a "Troublemaker"..
Little fiefdoms cost companies millions in license fees, and other expenses. Its amazing how mangelment can spend company money like no tomorrow on big projects without any oversight, tracking.
You think SOX is going to help? Every penny? No, it just tracks money, you can still pay for the same product a dozen times to a vendor.
I will say Oracle wasnt the worst license, but it ranks up there..
Our smaller servers use mysql now, lucky they proven themselves so we can try to use it in larger production boxes. Same with redhat, get in the door, then prove it can handle some tasks..
One thing though, in a major oracle outage, oracle will be on the phone and have a group of people willing to fix the problem, the support has been great. But thats a support contract, not a license.
And that's pretty much the bottom line.
... find another product ... or build our own product (it's really not that hard).
... unwilling to share 10K with a company providing core software!!
... If our software is installed on a "fill in the blank". They should rather say ... We want a cut of your bottom line ... show us your books.
We may not like that people have a right to sell a product for what ever they deem fit. It's their right. And it's our right not to use the product
Individuals have a right to seek what ever compensation they would like.
I know of too many companies pulling in 40M on a software project
I think the water would be less muddy if software companies didn't say things like
Cheers,
--The Dude
As CPUs get faster you get more performance for your licensing dollar... Adding another core to the CPU is no different than adding a few more ALUs or a SIMD unit, or a memory channel, etc... Yet you don't see them charging more for a license on a CPU that performs better than they do for a slow processor.
Why aren't they licensing per database, or per application, or per unit of performance? Or something, anything, that actually has some actual meaning?
Then again, this whole thing is silly. Oracle is just like every other enterprise software company. They send their sales staff out and negotiate a price for each individual customer that is somewhere just below the absolute most that customer is willing to pay without going to another vendor. What they divide that price by on the invoice is just a petty detail.
There are two issues here, not one.
First, should a multi-core processor chip count as more than one CPU. Second, should software be licensed on a per-CPU basis.
I think it's obvious that a multi-core processor chip should count as multiple CPUs. Arguments otherwise seem to equate a "chip" and "CPU", something laughably oversimplified. You can have "processors" that involve no chips at all (remember TTL "CPU boards"?), or that are made up of dozens of chips -- so there is really no inherent relation between "processor" and "chip".
Put a dual-core chip, or a quad-core chip, into your machine, and you have to deal with the same issues as a dual-processor or quad-processor machine.
[I would love to see how many 6502s or 6800s could fit in the space of a "modern CPU" die, possibly with some RAM on-chip for each "core". Play the games with clock speed on top of all that, and it might be something quite interesting to program for.]
The second issue is harder, and shouldn't be allowed to influence the definition of what is or is not a "CPU". If you don't agree with per-CPU licenses, then don't fudge the definition of "CPU", rail against the real grip: per-CPU licenses. If you do claim to agree with per-CPU licenses but are too cheap to actually PAY them once you get a machine with multiple CPUs, stop trying to muddy the water by claiming your multiple-CPU machine really isn't a multiple-CPU machine.
If your concern is that *all* new machines will eventually end up as multiple-CPU machines, that's yet another legitimate concern, but if you chase the bleeding edge, you're going to bleed. Don't pretend to be suprised by it.
Personally, I don't much care for multiple-CPU licenses. I'd rather deal with a per-machine, per-user, per-organization, or site license. But not all businesses want to play that game, and that's okay, so long as there's always an alternative.
Pick One: http://www-rohan.sdsu.edu/~stremler/sigs/sigs.html (Note - disable Javascript first!)
who cares howmany cpus you have if it's only one computer? I know companies do care a lot however the whole licencing based on cpu count rather than computer is just insane. If you have an 8-way server do you actually think that those 8 cpus are there to run 8 different installs of windows or oracle for that matter? I know for a fact that there is only one install of the OS and only one install of oracle. The number of cpus increases because you need more processing power and NOT more installations of the software. And last I checked you can have as many people as you want using the same computer. So why should it be any different for servers of dual cores for that matter. Oh yeah and if Oracle doesn't quit that crap they are in a lot greater risk of looking business to OSS that M$ ever was.
I think that charging more for dual core licenses / multi cpu is very fair, it benefits the small guy / business and its supportable by the market. Here's why in a nutshell....
Development of multi-threaded applications is more expensive. The scope of possible flows of execution is vastly increased. Design, programming and testing time has to be put into achieving a proper interaction between threads. This can add up to a big overhead. However the performance benefits ban be large / necessary.
So if you have a small business / area hobbyist and are therefore not needing massive performance; not making the money from that performance; cannot justify a large expense to get that performance - but still want the functionality. It seems fair - because you're not really fully using the work behind that threadedness; and marketable - you cannot afford a huge license fee, yet you become a customer.
A large business on the other hand is fully taking advantage of the work and benefitting from it financially. Therefore they will pay more, they can afford too and it is right that they should do so.
It's a good kind of metering. There are other kinds of metering in software - like needing a server version of windows to handle more than 10? concurrent connections. Microsoft's counting of CPU counting has simply been a pragmatic one - in the future if the market makes sense they will count the cores.
"What does Oracle's stubbornness imply for the industry as a whole, with multicore chips coming to the fore so strongly?"
more marketshare.
US Citizen living abroad? Register to vote!
Per-processor licensing has always been a shady sham thing anyways. The industry really needs to get the hell away from it. Even before this multi-core monkey-wrench got thrown in, it never made sense. Why should I pay twice as much to run your code on a crappy old quad 200Mhz Pentium Pro as I do to run it on a dual Opteron? Why should I pay more to run on a dual P3-933Mhz than I do to run it on a single P4-3.2Ghz? Some vendors try to bandaid the situation by classifying architectures and machine types into tiered per-processor pricing, but these inequities still abound even then.
If they're really stuck on the asinine idea that I should pay more for their code if I run it on bigger hardware, they could at least rate all the possible platforms they support in terms of MIPS or GFLOPS or something and charge per-performance-unit licensing.
11*43+456^2
Which makes me wonder - how do they rank up against Oracle/SQL Server? (I'm primarily a SQL Server user, followed by PostgreSQL and the odd once in a while MySQL - Access when I'm stuck fixing someone else's crappy work). Any good speed/features/pricing/... reviews out there?
It would also be nice to hear about people who did the switch to DB2 (or other alternatives to Oracle), their main issues and everything.
///<sig
Oracle should charge per MFLOP. That will drive many more people to Postgres and GNU.
Oh well, what the hell...
If the Oracles and BEA's of the world don't change their licensing policies, then Open Source is looming over their heads like a giant wooden caveman club. Before Open Source, these companies could've just told their customers where to go. These days they at least have to think about it.
r ticle.php/3 423971
Microsoft has at least partially come around to understanding this. A couple of months ago, AMD and Sun launched a similar initiative as what HP and Intel did this time.
Single License for Dual Core at Microsoft
http://www.internetnews.com/bus-news/a
-
Using software. If software had no licence, then the default legal position is that you're free to use it as you see fit. There's no need to 'grant' that right in a licence, because you already have it!
-
Secondly, there's copying. As you might expect, this falls under copyright law: you're not allowed to make copies of software (unless you own the copyright, or have a licence from someone who does).
Do you see why different laws apply in the two cases? If you're just using software, then that has nothing to do with copyright law because you're not copying it.Licences like the GPL and BSD ones say nothing about using software, so you're back to that legal position of being able to use it how you like. OTOH, companies like Oracle and Microsoft make you agree to a licence before you can use the software, so they restrict that right. Those licences are contracts, so if there's a problem, it's a matter of contract law.
Proprietary software licences don't usually give you the right to copy software (apart from backup purposes), but the GPL and BSD licences do. So they're adding rights, not taking them away. If you don't comply with them, you're back to the default legal rights, so you're still free to use the software how you like; you just can't copy it (which would violate copyright law).
Ceterum censeo subscriptionem esse delendam.
The problem with per-cpu licensing is that because gaining so much performance in CPU upgradea is going to start to be more difficult in the forseeable future. That's why multi-core is so important; you can't move up anymore so you move horizontally.
The per-cpu licensing scheme has always been flawed.
If they really want to charge based on machine performance, then they should do just that. Calculate how many MIPS the processors can do. Add them together, come up with some metric, and charge based on that. That way, it doesn't matter how many cores or processors you have.
- It's not the Macs I hate. It's Digg users. -
All 3 offer everything I need, but Oracle is like twice the price of SQL Server - or 4x if you're looking at the standard ed (not considering we already have the licenses). And SQL Server performs pretty good (has the top of the TPC-C price/perf - if that matters to you). Same story for DB2, but it's "only" 50% more expensive instead. I'm not a big microsoft fan, but mssql works really well, it's simple, performs well, ... Never had any problems with it really.
And (the enterprise ed) includes OLAP, Data Mining and enterprise reporting. For Oracle/DB2, it costs extra.
Also, another big factor for me is - the SQL server developer edition is only like 50$, so having development servers is really cheap (I run one for developping on a old beige box PC loaded with ram and it works great). Not sure if Oracle/DB2 have similar offerings.
Anyways. Unless I'd have specific needs for Oracle or DB2, I just can't see myself spend the extra $ (a lot of it), plus have to "port" everything to it.
///<sig
This is basically an old question. Back about 15 years ago we were using numerous mini-computers and ran into a similar problem with DEC VAX series machines. The C compiler for the MicroVAX and our "large and powerful" (for that era anyway) VAX 11/780 utilized exactly the same binary on exactly the same media, but the cost difference was something like 3-4X. Seems we could expect much better performance from the "big iron" and had to pay accordingly. Never mind the additional $$$ we spent with DEC to purchase the machine. This is a capitalist marketplace and some one will offer the better solution at the best cost. I have no doubt that this too will sort itself out.
I see java/database arguments and flamewars pop up from time to time. They usually centre on the performance vs. portability argument. I'll admit to some bias towards the smart-database school. Here's my $0.02 worth on this particular debate.
The fallacy of total database independence
For political reasons Sun would like us to treat Java as a platform. Therefore java is designed to be all things to all people, a layer between you and the operating system. This is at odds with what it actually gets used for. The same is arguably true for .NET, though for slightly different reasons.
Neither of these platforms are widely used for developing shrinkwrap software. I'll argue that both platforms are essentially irrelevant to the shrinkwrap software community outside of those companies that make Java or .NET development tools and infrastructure. Most java (or .NET for that matter) development is done for bespoke applications - where the client is paying for all the development and maintenance work. Most importantly, the client also owns and therefore specifies the hardware on which the application will be deployed.
I believe that the key fallacy of the portability argument lies in these four points:
Point 1: M:M application:database relationship
The database is not your personal persistent object store. Other people have to use the data.
Outside the trivial case, databases and applications live in a M:M relationship. The data will be used by more than one application. Tying your data integrity to the middle tier forces you to to go through the middle tier to update it. You have not escaped platform dependence but just moved it up a layer.
Any database has multiple stakeholders, including business users who want to analyse it. Therefore, application developers are not the only stakeholders in a database. Developers often do not realise this until it is pointed out to them.
Point 2: You still need to regression test
Any other platform (hardware, OS etc.) that you intend to support deployment on needs to be regression tested. Even if the application is write-once-run-anywhere portable, you still have to devise, maintain and find the warm bodies to execute a comprehensive regression test. Java's portability is not good enough to avoid this.
Point 3: Java is mostly used for bespoke development
Remember that Java is mostly used for bespoke development. The practical difference between "write-once-run-anywhere" and "relatively easy to port" is not an overwhelming issue on a bespoke project. There is a significant investment in development, and a set of regression tests has to be carried out.
If you have spent $1,000,000 on an application then the difference between $100,000 to regression test it on your new platform and $150,000 to port and regression test it is not such a big deal. The difference between "zero development effort" and "reasonable development effort" is not so significant in this light.
Point 4: O-R mappers are not a substitute for competent developers
If you're developing a database app without reasonable SQL skills on the team you've got bigger problems than portability. The notion of abstracting away data access into a black box is utopian. Data access performance is actually subject to mechanical constraints of disk performance. NOTHING else in the application stack has to actually move chunks of metal and ferrite (or whatever they make GMR heads from these days) around to work.
For this reason, database independence is just as big a myth as network transparency. Hands up those of us who design EJB components without regard to network round trips. You cannot ignore what the database is doing behind the scenes.
My argument is thus:
An application using POJO's and a reasonably competent data layer gives you flexibilit
Who cares what Oracle and M$ say about it? Just use Free Software and you can use as many cores as you want! End of discussion.
Is this karma-whoring or is it a genuine knee-jerk response of the type that gives open-source supporters a bad name?
You know, there might be good reasons for people to prefer Oracle to open source solutions, in spite of its inherent disadvantages. Perhaps it would be better to acknowledge them, and/or provide a persuasive argument in favor of your preferred open-source solution.
But I'll tell you now; open source may have numerous advantages, but Oracle is still way more powerful than current open-source offerings.
That may change, but perhaps you should direct your efforts into improving them, rather than spouting black-and-white zealotry.
Its obvious that Oracle can do whatever they want in terms of their contracts and that users can eventually leave.
The problem is that it exerts an un-natural force on chip development. Sun is about to come out with an 32 core 4 cpu pizza box server next year. In terms of through put this may really be the way to go; but companies will be unable to select it because of the forces acting up stack.
It could lead to commercial applications acting as a disincentive to engineer the next break-through. If that happens the whole industry looses.
Ok, most people here think it is crazy for oracle to change double when you still only have one motherboard. Let me give you a counter story.
We use a product here that was built with allegro. If we want to extend it then we need a license to allegro and they don't come cheap. The nice folks at allegro think that it is crazy to charge two licenses for a single machine and so we went out and bought a SMP box just so we could have multiple developers log in at the same time.
Now, in terms of MIPS/dollar, the SMP box came out quite poorly. But once you factor the license in, it is cheaper for us to buy a 4 way enterprise server than it is to buy four workstations and four licenses.
I think we (legally) swindled the developers of allegro. Of course, at the amount they are charging for once license they deserve to be swindled, but I digress...
That's like saying a hard drive with four platters is really four hard drives, not just one.
As CTO, it came down to doing the math when we were looking at updating the Oracle or starting from scratch on a simlar project but starting on a 4 processor box capable of being updated to 8-way. Instead of Oracle we chose PostgreSQL, even though it will cost us quite a bit of up-front programming costs - the down stream benefits will be enormous as we move to the new multi-core processors (next generation hardware we're already looking at)
Been there, done that, paid for the T-shirt
and didn't get it
from what I have heard, Intel's z-project is developing a cpu with 256 cores. Figure what Oracle would cost on that.
...of one of these dual core things, do I have to pay *double*???
(Stolen sig) Remember: it's a "Microsoft virus", not an "email virus", a "Microsoft worm", not a "computer worm
I'm not surprised in the least that an organization whose sole purpose is to generate profits, is looking at a creative (if rediculous), means to improve their bottom line. That is their job. In fact it's job one, and any CEO that forget's that will soon be looking for work.
That said, it is the purpose of people, the citizenry, the public, you and me, to make certain that when a company attempts to make a profit by paving your and/or anybody elses' ass over, we step up and say 'NO". We do this through legal channels, we do this through regulatory bodies, and we do this with our pocket books.
In the not too distant future, a machine at the center of your home, or your weareble technology, will have a reconfigurable processor perfectly capable of spinning up dozens or even hundreds of cores. One or more may be running proprietary software that some company can claim they should be getting paid for. The point is, that only one customer, is receiving value from their singular operation of a product they purchased for their own personal use.
I'm terribly sad this makes it more complicated for HP and Oracle to charge time against service, but to suggest that they should be getting paid by the core is rediculous... as a response, I'd suggest that if they want to charge by the core, that as users we resond by paying only for the process time alotted. By paying them only so many femtocents per Core cycle, they would suddenly see a significant drop in profits, and would see the err of their ways, hopefully shutting up and thanking their lucky stars that they still have a product and some semblance of a customer base (keep screwing with your patrons and see what that does to your long term profits...)
I don't blame them for money grubbing... it's not pretty, but it's kind of expected. I do blame them for shear stupidity... what makes them think people will just assume the position and take what it is they're trying to sell us... for shame...
Genda
Anyone runing on a dual core architecture should definitely need 2 GPLs :-)
I don't understand why a story about Oracle raising their prices, is thought to be interesting.
And remember: From a user's point a of view, the question isn't "Should dual cores require dual licenses?" The question is, "Should I license proprietary software at all, instead of just buying it and using it under the terms of copyright?" As soon as you start asking questions about the details of software licenses, you've already made a huge consession.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
While Microsoft may not make a distinction based on the number of cores, they have assured themselves a healthy revenue stream based on virtualization - another key technology for the near future. Microsoft requires a license for every virtual machine running under any virtual technology, regardless the number of CPUs in the system. Even if you are running Windows XP in a virtual machine on a physical machine running Windows XP, you must have at least 2 licenses for Windows XP. I would not be surprised to find a lot of VMware and Virtual PC / Server customers that are in violation of this requirement. I could understand requiring a license for Windows Server 2003 running in a virtual machine on that XP box - but only one license for any number of virtual machines. Looks like Microsoft is protecting its revenue stream against any extreemly powerful single processors also. Too bad, machine virtualization is a very good protection against the various memory leaks and lockups caused by Windows applications, and is an effective strategy to consolidate servers for small remote offices.
Every change is not progress, but there is no progress without change.
Per-cpu licensing has been around a long time, and has been a retarded idea a long time. Hopefully, this will spell the death-knell for per-cpu licensing.
Licensing should be per-computer. Or possibly per-user in some cases. The number of cpus/cores in your computer shouldn't drive up the cost of the license.
Oracle, MS, et. al were able to get away with per-cpu licensing as long as the only really multi-cpu machines were corporate 'big-iron' that had 8 processors. But now that cpu makers are trying to shove multiple cores into the cpu, and multi-cpu systems are gradually becoming more mainstream, it's not gonna wash. Home users will never agree to pay microsoft $1200 for licensing an 8-cpu copy of windows for their 8-core computer.
Granted, the immediate thing in question, Oracle, isn't a home-user situation, but even companies are going to get fed up with it. If Oracle gets too greedy, companies may just choose from among a few other alternatives. And if they are really smart, and as long as it can do what they need, they might choose a free software database, like PostgreSQL or the ever-popular-on-slashdot MySQL.
Dual processor (dual socket) motherboards cost a bunch of money... not necessarily because the boards really cost 2X as much to make, but because they simply *can*. Companies that want/need dual/quad/etc machines will pay a premium to get them. Intel knows this, AMD knows this, Sun knows this, software companies know this, everybody knows this. There is a premium on multiprocessor machines and software that is designed to take advantage of multiple processors has a premium attached to it as well. Machines and software usually license by the socket (single-core processor) because they can charge a premium because those that need that functionality are willing to pay premiums to get it. Simply look at traditional costs of things like Windows 2000, Oracle, and plenty other pieces of software that charge by the CPU for licensing.
Now, with dual cores, especially since Intel and AMD seem to be going to push dual-core onto the desktop as the 'norm' instead of as an exception. This means that everyone will be able to get them (probably won't have much of a choice after some point). Now, all of those pieces of software that charged premiums for multiprocessing licenses have a problem. Usually, single processor versions of some software (Oracle, Windows, etc.) were pretty cheap comparitively. Going up to the next licensing stage for multiple processors is a big jump. So, software that normally ran on single processor computers at home and at workplaces will suddently be very expensive if you say that dual-core is the same as dual-processor (which it in all ways that really matter to a programmer and user is). Will home users and businesses with 'normal' computers on lots of desks willingly buy "server" versions of Windows 2000? or Oracle? or whatever? Their $99 piece of software per seat just magically jumped to something like $1500. Would YOU buy Windows XP (assuming you have before) for $499 now because you bought a dual-core machine?
Microsoft and everybody else knows that claiming dual-core is the same as dual-cpu and charging accordingly will not fly. Also, since to tell the difference between a dual-core machine and a 'regular' dual-socket board with single cores requires extra coding (have to look up CPUIDs and the like), how do you make people with dual-socket boards with single-core in each pay different prices than dual-core single-socket users? It's unfair and any company that did that would be blasted to smithereens in the marketplace. The only recourse is to say that licensing will be by the socket, not by the core.
This way, there is still a premium for dual socket boards, for the same reasons that there's a premium on them now. Software licensing and such can likewise retain their premium costs for multiple-socket boards. Single socket boards will still be cheap because the market demands it (even if there are two or more cores on the chip you put into that socket) and the software that runs on the single socket boards will continue to be cheap so that licensing fees do not suddenly get astronomical when dual-/multi-core is the only option.
Now, from the software side, there is very little functional difference between a board with two sockets and single-core processors in each socket and a board with one socket but two cores in that chip. There may be some performance implications but programming the things is still so similar as to require special code just to tell the difference, and then you wouldn't do anything with that knowledge except maybe charge different licensing fees.
Dual-core vs. dual processor, vs. whatever is all just marketing and making prices of software and hardware remain pretty much the same as they are today for those who don't want to pay the premiums for having multi-core machines (multi-core covers having multiple single-core processors or one or more multi-core processors).
The other "benefit" to these licensing schemes is that it will really push multiprocessing out the door into the masses (something Intel has been tryi
I run several fairly sizeable Oracle, MS SQL and even an Informix database with a healthy number of simultaneous users hammering away at each one. On comparable-performing hardware, the Oracle by far supports the greater number of concurrent users. MS SQL is just as fast as Oracle, as long as you only have only a moderate number of users hammering away at it and a fairly simple database structure, but if you have a lot of stored procedures, triggers, and complex queries hitting it, it just doesn't handle the load nearly as well under stress like Oracle can.
Interestingly, Informix handles stress more gracefully and linearly than both Oracle and MS SQL.... up to a point when it then suffers sudden meltdown and crashes the engine.
There are a few priveleged companies that are always on the make and will spare no opportunity to rake off their customers. When the technology on mainframes changed from requiring the use of all the CPU's to do whatever minmal work was required by a software package to one where you could tailor the hardware resources to fit the package, these greedy companies still insisted that if you were using their application only on one CPU, you should pay them for 16 CPU's worth of processing if 16 processors were physically present. About time someone challenged them in court.
has a sound approach to this trade off.
It is encouraged that you look at the generated sql, via HQL you can greatly influence the generated sql yet it is automatically adapted to the current sql dialect.
If you need, you can use direct SQL instead of HQL, e.g. to add optimizer hints. Those parts would have to be ported, but in practice it is only a few spots in an application that are really performance ciritical. Thus you can not 100% eliminate portability issues, but you can reduce them to a very small fraction of the application yet do not compromise performance at all.
As for id-generators: hibernate has about 10 built-in methods, a.o. oracle sequences, but also a table where the application may reserve id-s in (large) blocks thus removing the potential bottleneck with that. (sequences have their drawbacks too). A third way is to generate a unique 16-byte id by mixing in ip address, JVM starting time and a counter (however id's generated in the DB would have to use some other scheme without overlap in this case).
Point of information: AMD announced dual core Opterons months before intel announced dual core Pentiums.
The engineering samples have been out for a long, long time.
The 387 wasn't a seperate processor. It was more like an add-on extention to the microcode ROM of the 386 to implement the floating-point instructions, in addition to a few extra pieces of low-level circuitry, and it also contained the floating-point specific registers.
Please note that "Hyperthreading" processors are not dual-core. They just have hackery to make a single pipeline look like two.
Stick Men
We should make all open source licences this way. Er Humph. CLAUSE XVI This licence is for a single processor core only. By using this code on a machine with multiple procesors, you will have accepted this licence multiple times. Just thought you'd like to know that.
This is not a new thing. Specific machine software licenses have often been tagged to the power of the machine running them -- or the number of users.
Considering just the number of processors is kind of odd, though. When I've seen this done in the past, it's been based on the measure throughput of the application (and usually gave a discount to the higher performance machines).
An engineer who ran for Congress. http://herbrobinson.us
Per-user licensing is roughly the same logic - the more users you have, the more money you have to spend.
I type this from a Windows box because the weekend is over and I don't have time to fix YALP, Yet Another Linux Problem. Oh, sure if I didn't have a family, work and a desire to do other things than tweak and fix basic components of my system then I suppose it would be no problem to fix... unless I had to get something done quickly. Most people reserve an emergency Windows box as a "quick, get something that just works" solution when in need. Consider this a rainy day fund for productivity. Most Linux fans don't have to worry about productivity so this is not a problem I suppose.
While not a definite judgment of zealotry, always look for those "M$" and "Windoze" statements as a sign of the mindless zealot.
I have a nice running truck and a car. Sure, if I wanted to I could start some hobby-car project. I might even join a local HcUG and spend almost as much time working on the car as I spend on various forums and such. I would be a fool to get rid of my current vehicles. (especially my truck... I doubt a hobby hotrod is going to handle hauling 500+ pounds of hay in mud that a man sinks up to the top of his shins in)
I will tinker and toil because my purpose is to tinker and toil. The goal is a tinkered toy that is the result of my own blood, sweat and tears. I just don't think I would be satisfied with anything less.
Lets not pick too much on the Linux zealot... after all even a hobiest's toys can contribute improvements to the auto industry and all users of automobiles, from time to time.
is that it's still Free no matter how many cpus/cores/computers you run it on!
JDO was a good attempt intially at creating this framework as it used ORM as a lower level implementation and had various (some still plans) implements for either dynamically mapping at runtime or just helping a brother out at design and code time. Hibernate kicks ass and I am glad that seems to be the choice as the ORM component for the next round of JDO but JDO seems now overly cumbersome, including what I have seen from its next iteration.
So yes I agree with you fully. You seem to have come to the same conclusion I have based on pain of usage and annoyance at flopped attempts to abstract away from the specifics. If conceptually, a relational database is merely a particular paradigm for storage and retrieval of data then why do we then ironically design data persistence frameworks (and use those terms) that then are married to the relational paradigm?
Let the dbm software do what it does best and leave usage logic to external sources. I however can see a need to start replacing the more general but yet more tightly restrictive (from an implementation detail perspective) stub of "SQL" and "Stored Procedure language X" to a more useful set of implementation, access, and extension stubs for use by whatever system you desire.
Not being either a MS person or a current .NET programmer I can still say that what I hear about the .NET plans of integration and extension of MS's SQL Server 2005 will go a long way in helping these efforts out. If nothing else, it will kick some people in the head who are still stuck in that narrow scope of thinking about current computing paradigms. The kind that if you asked to create logic extensions and perhaps even create a general purpose "Persistence Layer" they would inevitably just wrap some classes around SQL and specifically RDBMS based design. I guess that means that text and binary files, data streams, and transient/volatile data just doesn't ever need to be used. Lets just scrap it all and go with SQL and RDBMS!
Grandparent was making a point about market forces, and how Oracle will have to figure this out for themselves based on what their customer will be willing to pay for.
Parent seems to have misunderstood the point, but still makes a condescending post about terminology to get an argument started. He then admits at the bottom that he doesn't even understand the point of *any* per-CPU licensing, let alone the question under consideration.
Oracle will soon announce their latest license revolution: The Cost of Doing Business Monitor (CDBM). The CDBM is a stand-alone computer that monitors your entire enterprise and charges you for each unit of work in which your enterprise engages.
It is a very sophisticated system that will track every mile driven, key pressed, KW/h of lighting used, oxygen breathed by employees, phone calls made, line of open source code used, etc. and will use electronic funds transfer (EFT) to forward those "license fees" in real-time to Oracle.
What if you don't use Oracle? Have no fear, the CDBM uses Oracle 10g to keep track of your units of work, so you will still owe the same amount if you use Oracle products or not.
Apparently, the philosophy behind this revolution is that anyone who makes money doing anything, most likely owes money to Oracle.