Slashdot Mirror
AMD's 12-Core Chip Cuts Software Licensing Costs
CWmike writes "AMD released on Monday its 12-core chip code-named Magny-Cours, doubling the number of cores over the previous-generation Opteron chip. While a doubling of performance is nice, another key benefit delivered by a chip with a dozen cores may be in reducing software costs. For Matt Lavallee, director of technology at MLS Property Information Network, a company that supplies real estate data, upgrading to the 12-core Opteron chip from his current quad-core chips will allow him to cut the number of servers — and his software licensing fees. While the 12-core chip costs a little more than an eight-core chip, it's 'nowhere near as much as a SQL server costs,' said Lavallee, who has been beta-testing the new chips. MLS operates 60 servers, and Lavallee said he could theoretically cut the number of servers by half but will likely reduce his server count by a third with the chip upgrade."
Reader adeelershad82 adds that AMD is hoping the new Opterons will compete with Intel in the high-volume server market.
These companies will always protect their bottom line. It's an arms race of sorts. Increase power to cut down on the need for servers and licenses, and these companies will change their licensing models / costs to ensure that you are still paying the amount in the end. It's the same with HDDs and other hardware. The price/GB may go down, but the size of the drives just keep going up, ensuring that you still pay $100 to replace your parents' dead HDD, no matter the decade.
Why the heck is he paying anything? Just use MySql and be done with it. It is certainly easier to use/setup/maintain than that crappy SQL Server stuff. And it is free to boot! sheesh.
Really? You mean, as computers get faster you *might* need fewer of them?
No really. Please provide evidence for the thesis that as computers get faster, people need fewer of them.
Second point. It's usually the I/O performance anyway. A 12 core server is unlikely to be able to push as much throughput as 3 quad cores, given the same I/O technology.
Deleted
One of the coolest things about this stuff is that inside of one dual-processor workstation you can set up a whole datacenter worth of VMs, and model how the pieces interact without fiddling with racks and cables. You can build up a redundant database, fileserver or iSCSI server solution (or all three!) and see how it handles failover and failback. The simulated clients that apply stress can be VMs in the same box. You can even float a cloud of routers and see how they handle various BGP commands. Pretty neat stuff.
Help stamp out iliturcy.
And the fee for linux is the cost of the admins -- the people who are good in the environment know they are good and their price goes up every year -- it takes about three years to become really proficient at most MS products - it might take half a year to really understand the linux environment and methodology if you are unfamiliar and then another 4 - 5 to gain that same profiencency equivalent.
Interestingly, I work for a company where the IT-department is a pure Microsoft-shop. Only windows-hosting and almost only C#-development for internal applications. In the technology department however, we operate a bunch of production-system for our customers, running mostly CentOS Linux. Lately, I've discovered that the Linux-admin-staff often writes simple script-solutions with their left hand, with equivalent complexity to systems written by trained developer in the IT department. And even though I personally often would rather see a more structured systematic approach to some problems, when listening to the end-users they almost always perceive they've gotten BETTER support and reliability from those scripts.
Point being that, a Linux-admin MAY cost a bit more than a windows-admin, and the learning-period might certainly be a bit longer, but I see much more productivity coming out of our Linux-crew than the windows-equivalent. More services hosted and administered per admin, and ~10 times the operational availability. Also, when more complex jobs needs being done (configuring network device, someone needs help with a tricky SQL query for a report, or needs someone to mirror a huge chunk of text-files into a searchable DB for performance), they usually come to the Linux-crew than the windows IS/IT.
What I will give the person who goes the linux route is that once you are profiecient in Linux - gaining the same proficeincy in other systems is cake - basically because the they are just easier to use in the first place.
Definitely matter of oppinion. Personally, I've never found anything "easy" about windows. Sure, the very limited amount of things you can do within three button-presses is usually simple enough (interestingly the same goes for modern Linux Desktops/Simple Server Setups). However, once something breaks, or you need/desire to stride outside the comfy gui-box, just forget about it. (IMHO) For example, a standard CentOS5 server install comes with high-availability software that from commercial vendors (IBM and HP, I don't know if Microsoft can even match the fully distributed transactional storage components) START at ~100K euro. For those money, I can let one high-school self-taught Linux hacker spend 2 years in researching and fine-tuning for the JUST the entrance fees of the proprietary variant. How would you estimate my chances of getting some more use out of that admin meanwhile?
SQL server is solid ... Oracle is just as good if not hands down better
Oracle is hands down better - a lot better. But, you go way off into the weeds with stupidity and ignorance.
robustness to their product lines how long will it take for postgress to catch them??
Go read up about what's coming in PostgreSQL 9.0, which is right around the corner; plus its current feature set. Now go read about EnterpriseDB (commercial PostgreSQL offering). Not to mention, according to the FAA's (yes, that FAA) recent talk at PostgreSQL East Conference, PostgreSQL + PostGIS is roughly TWO ORDERS of magnitude faster than Oracle Spacial. Furthermore, recent benchmarks of Enterprise DB squarely pushes far into Oracle's warehousing territory; which also supports things like parallel load/query/indexing, etc. At the FAA, Oracle is being completely usurped by PostgreSQL. All new projects are to use PostgreSQL and/or PostGIS.
PostgreSQL has shown that it easily out scales MySQL for most common loads and its performance now trounces it in most cases. Plus, PostgreSQL's query optimizer makes MySQL's look like a toy. Additionally, even as far back as 2007, PostgreSQL was typically on par with Oracle's performance and scalability. More recently, PostgreSQL is typically on par or exceeds Oracle in performance. When talking about spacial queries, clearly according to the FAA, Oracle isn't even playing in the same park as PostgreSQL. And if you need seriously high end enterprise performance, commercial PostgreSQL offerings, such as EnterpriseDB, is here too.
So next time, before you start mindlessly mouthing off about open source catching up, you might want to put it into your head, its far more common for commercial needing to catch up with open source than the other way around; at least these days. Seriously, take a hard look at PostgreSQL. For the vast majority of users (small to medium size databases), MySQL, MSSQL, and Oracle shouldn't even be considered as PostgreSQL already has you covered. And frankly, MySQL isn't even in the same league as MSSQL, Oracle, and PostgreSQL.
Since multi-threading is so hard to do right, most of what you are going to do is consolidation.
So instead of having 6 separate servers, you just shove buttloads of RAM in a single server and set up a SAN for your data storage, and move all 6 servers to one box. You can even split it up further than that - if you have a couple servers that need to be separate from each other, but don't really need a lot of processing power, you can put those on a single core apiece. So you could potentially consolidate up to 12 servers into one box with virtual servers. More than likely you'll only get 6 or 8 out of it, because dual cores do help a lot, but still there's the potential to turn two racks of servers into one server and a SAN.
You save on space, you save on energy, and you ultimately save on hardware (though SANs are expensive, so if you don't need the speed you could go to a standard NAS setup). To expand your data storage you just need to expand your SAN, so you can add servers and storage independently of each other. All of these are major up sides to going this route.
Going from a 60 server setup to a 10 server setup has a massive potential for savings.
Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
it might take half a year to really understand the linux environment and methodology if you are unfamiliar and then another 4 - 5 to gain that same profiencency equivalent.
Since the sibling posts have been dismantling the rest of your bullshit post I thought I'd attack this. I went from Linux virgin to Linux expert in about 3 years. Now all my home computers run Linux (well my gaming PC dual boots Win7), even my phone runs Linux. I was already a Windows expert but I'd say I know more about Linux now, as I can do things that aren't possible for a Windows end user - Windows feels a bit like a toy now.
"When information is power, privacy is freedom" - Jah-Wren Ryel
No, it's a reflection of the differing architectures. The 6-core Intel processor has about 50% higher clock speed and the new hyperthreading while not perfect, really does improve performance on virtualized workloads. So net, performance should be roughly the same for most things even though one has six cores and the other has 12. And released at the same time. What a coincidence. Each of course has its strong points.
One strong point of the Intel CPU is that you can run the cheaper VSphere Standard edition with six cores, and hyperthreads don't count. From Seven to twelve cores per CPU you need the more expensive Advanced or Enterprise Plus.
Nehalem-EX coming up has up to 8 cores though so we'll have the question again in a few months. To stay at parity AMD will have to ramp their clock. They can't bump the cores to 16 because that falls afoul of VMWare.
Help stamp out iliturcy.
There's nothing that easy about MS technologies. They're superficially easy, sure. But there's quite a lot underneath that.
Well, I certainly agree with you there! But what bothers me more about Windows technology isn't that it's as complex/powerful/intricate as comparable Linux technologies, as much as it's opaque.
You get a binary to install, and there you go. Enjoy, and hope to God that somebody at the other end of the 800 tech support line has mercy on your poor, sorry soul. Because you have virtually no recourse otherwise.
Compare/contrast with more open solutions, which provide options when the chips are down. How many times I've pined for a decent documented config file when rooting thru the menus to fix some obscure problem!
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Which given Oracle's pricing, could be a whole heck of a lot.
I once heard a VP saying something like the following: "Today, the Oracle salesguy is coming to wrap up the licensing. I've cleared my complete schedule for today for the negotiations. It's worth it. I never save so much money on a day as when negotiating with Oracle."
8 of 13 people found this answer helpful. Did you?
I've noticed this with LCD monitors at newegg.com. I bought a 21 inch lcd monitor for around $200 a year ago and now I can get a 23 inch for $200. But what I'd really like is a brand new 21 inch for less than $200 but no such luck, the minimum size available to buy keeps going up.
Time to render 30 minute Video CD image (at VHS resolution) on 1999 mid-high level PC (cost £1200): 10 hours approx, PC effectively unusable for other purposes.
Time to render 2hr DVD image (at std DVD resolution) on 2008 low end PC (cost £350): 30 mins approx, PC also playing music/video, web browsing, ripping CDs etc. at the same time.
The effect of 'bloat' is often overstated.
And if 100% of what was done with computers scaled like that, you'd have made a great point.
Well, I think you've missed an important part of my point.
What I described is a good illustration of a fairly common scenario - one fairly heavy task going on in the background and a variety of less demanding tasks in the foreground. Although other demanding background tasks would not have increased in speed as much as video rendering, the general principle holds good that with a cheap modern PC you can run something really demanding at the same time as using the PC for several other foreground purposes; going back a ten years this was largely impossible even with higher end machines. So bloat is not cancelling out hardware advances.