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.
Has MS updated their licensing to be "per-core" instead of "per-CPU"?
http://www.bynarystudio.com
Fair enough, but my Linux licensing costs won't change!
From hell's heart I fstab at /dev/hdc
Almost all of the enterprise software we buy charges by the CPU and by the seat. For this purpose a CPU core is the same thing as seperately socketed CPU. Whatever about OS savings I think you'd save more in hardware and running costs than you would on software.
I never get used to these constant resurrections
In my experience, it's rare for SQL Servers to be CPU bound, they're almost invariably IO bound, and having more cores won't help you when your disks are the bottleneck. I could see excitement over lowering per-machine costs for something like a renderfarm, but it doesn't seem likely to materialize for Databases.
I read the internet for the articles.
Very clever, AMD. Naming your chip after a location in Europe as usual, but this time making it able to be read as "Many-Cores" (or possibly more accurately "Many-Core", I don't really know how to pronounce French words). Very clever indeed...
Will I need to buy more SCO licenses for this one chip? This could get expensive.
Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
What are these? Is this something that afflicts Windows people still?
Stick Men
Anandtech has an excellent review of the new chip. The AMD chip is compared against the latest Xeon. In some situations such as OLTP and ERP, the AMD offering lives down to it's name Mangy Cores. In HTP and data-mining, Anandtech gives the nod to AMD.
So choose depending on your needs.
Think VM (VmWare/Xen/Solaris Zones) instead of parallel applications...Multi-core CPUs are great for server consolidation. We went from a row of 10 full racks of Sun gear down to 10 T2+ blades + a SAN over the last 18 months. Database / webserver / Java app server, you name it, the T2+ handles it all!
If only "common" sense was actually that common...
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
Sorry, MySQL isn't in the slightest comparible to MSSQL or Oracle. It doesn't have half the features, it's buggy, and it's generally slower. Tooling is also poor in comparison. It's still, unforunately, a toy.
MySQL does well in the web. That's because it's free licence is suited to horizontal expansion - throw lots of cheap servers at it (where such expansion is possible). Tight integration with PHP just puts the icing on the cake. However, compared to other stacks it's poor. Both MSSQL/ASP.net and Oracle/Java-application-server perform significantly better (often factors) than the MySQL/PHP stack.
So buying Microsoft/Oracle might seem expensive, that is often not the case.
But the web isn't the world for databases. There are lots of other usages.
MSSQL is for example is ideal for SMEs, you get a heck of a lot for your money - very well performing database with mature, well integrated and well performing stack. Plus a really nice BI implementation built right in, with nice easy GUIs for dummies / business users.
Oracle's the daddy. It's complex but it's a more capable db than MSSQL. As a developer you have fine grain control over how the engine works. For certain enterprise applications it's the only real option (apart from going to IBM). I've been lead to believe that it's the performance king too.
If you're serious about open source databases, then you need to use a serious open source database as an example. Both Ingres and PostgreSQL are mature, well performing and fully featured databases which are available under an open source license. They're what you should be comparing with SQL Server / Oracle. Not MySQL.
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.
Actually Oracle charges per socket on Standard and Standard One licenses and per Core only on Enterprise licenses.
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
I have an SME ERP background, and while all of what you say is correct, in my fairly extensive experience (since 1987) as a VAR and working on the inside at Sage, it is rare to run across a customer, at least in the mid-market ($5 to $50 million in revenues) that actually needs all the features of MSSQL. Hell, most of them do just fine with c-isam or btrieve style files. Most companies in this segment can do just fine with MySQL. Also, there are lots of tools out there. None quite as good as SQL Studio, I'll give you that, but Navicat, for example, is pretty good and affordable. My biggest issue with MySQL is what Oracle is going to do (or not do) with it...
I still cannot find the droids I am looking for...
We've had a lot of problems with MySQL, especially the InnoDB engine corrupting datatables. It got bad enough during development that after the proof of concept, we ported to PostgreSQL and have been running ever since. And it's been night and day. All our DB's are now postgres save for our billing system, which was written by a 3rd party. PostgreSQL is taking far more traffic than we expected and honestly we were thinking that we'd be needing DB2 or Oracle at this point, but so far PostgreSQL has handled all we've thrown at it and with the new clustering/replication/HA hot-standby features in PostgreSQL 9, it looks like we can put off that large purchase another year or so.
"The problem with socialism is eventually you run out of other people's money" - Thatcher.
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.