Slashback: Vaidhyanathan, Oregon, Opteron

Slashback this evening with a few more notes on AMD's upcoming Opteron processor, Siva Vaidhyanathan (three times quickly), Oregon's open source bill, and more. Read on below for this round of updates and amplifications. That Charlie's no dummy. softwareJoe writes "Tim Bray has come right out saying that if IP chancer Charlie Northrup manages to enforce his most recent claim, the consequences would be 'disastrous.'

'It would become impossible to have Open Source implementations of key pieces of the infrastructure. This would be harmful, perhaps fatal, to the grand plans of those who want to deploy Web services everywhere,' Bray is reported as saying, in XML Industry Newsletter ."

Waiting for the low-power version. Jethro writes "Ace's hardware Opteron review was a very interesting read which shows some real Java webserver benchmarks on SUSE and Debian Linux, and real world database performance in MySQL and MS SQL server 2000. A lot better than those synthetic mysql benchmarks that Tom's hardware served up."

And Distinguished Hero writes "[H]ardocp.com ([H]ardNews 1oth Edition) is reporting that the Opteron processor does not actually have an integrated dual channel controller. This explains why all the Opteron reviews only used a single channel configuration. While the integrated memory controlled is not dual channel, it can be bypassed by an external (Northbridge) memory controller connected to the processor via the HyperTransport bus."

One more: EconolineCrush writes "Yesterday's Opteron launch gave us all glimpse at AMD's new 64-bit platform, but the Opteron is a server and workstation chip that will be out of reach for the majority of consumers. AMD's upcoming Athlon 64, however, will bring 64-bit computing to the desktop. Drawing heavily from what we've seen of the Opteron's performance thus far, Tech-Report has posted its thoughts on what it will take for the Athlon 64 to succeed. It's an interesting read for anyone salivating at the thought of an affordable 64-bit desktop platform."

Ma'am, can you please ask those anarchists in the carrels to pipe down a bit? BrianWCarver writes "Readers may recall a Slashdot interview with Siva Vaidhyanathan, Professor at NYU, and author of Copyrights and Copywrongs. Vaidhyanathan is working on a new book, The Anarchist in the Library, and was interviewed on the blog, Eyeteeth. This is a brilliant and amazing interview where Vaidhyanathan discusses how creative communities share, the DMCA, the American industrial production of culture, the USA Patriot Act, the importance of libraries and librarians, and the policies of the FCC. It is a must-read for those who care about the future of creative and democratic culture."

Technically, Oregon is not Washington. Daniel Phillips is among the many folks who have been following the progress of a bill in Oregon (HB 2892) to encourage open source software, and he points out this Register story (picked up from NewsForge, actually), writing "Apparently, moving Oregon's open source bill forward comes down to convincing the house speaker."

Reader PotatoHead fleshes that out just a bit: " Despite reports detailing the demise of HB 2982, this bill continues to be a topic at the Oregon Legislature. We have broad support for HB 2892, but need everyone to continue showing support in the form of your phone calls, e-mails, faxes and snail-mail to your Oregon Representatives. We have the attention of the Oregon Legislature in a pretty big way and need to keep up the good work if HB 2892 is to move forward against the constant efforts of the usual industry lobbyists. If you don't already know, here is how you contact your representative. Please take a moment --right now-- and show your support for HB 2982. Every contact matters as we continue to move forward with HB 2892!"

Sir, can you direct me to the nearest buggy whip store so I can beat this dead horse? If $98 billion seems to you a bit much for the music cartel to charge students for even the most indiscriminant file swapping, you may be interested in following the chilling effects that it generates, too: PL_2003 writes "A follow up on a previous slashdot article. It really seems like the recording industry is determined to continue its fight.Check this NYTimes article (free reg. required). My Take: Couldn't they use their brains for a better business model?"

OK, here are the rules ... Grub (mentioned previously) is apparently causing consternation among many webmasters. Though they claim the client honors robots.txt , it seems that only the central servers check it (and don't honor it properly) and that grub clients don't don't check it at all. Ooops.

Time to round up and segregate the arrogant. jtheory writes "There's an AP story today here on Yahoo news) that the Justice department has dropped its probe into the recommendation policy of a Texas Tech bio professor. It's encouraging that all he had to do to stop the investigation was make some very minor changes in his policy, but it's still horrifying to me that he got into trouble in the first place. Is it even safe to encourage strict Creationists (or others with strong anti-scientific beliefs) to become doctors? Would they ignore animal research results, etc?"

Comments

Re:Opteron memory controller details

[turm]

The announced Opteron parts do not have dual DDR memory channels

I don't know where this crap is coming from. Certainly not from the document referenced in the parent post.

A DIMM is 64-bits wide. The Opteron has a 128-bit wide memory bus, which means you need to use pairs of DIMMS much like the older P4's with Rambus memory.

There are plenty of pins for this in the 940 package. The block diagram on page 11 of the the data sheet even shows the 128 MEMDATA pins.

The memory controller is configurable to support a 64-bit memory bus (probably for desktop or mobile versions of the part), but in all the systems I've used you can't even boot with an odd number of DIMMS.

Now you can decide for yourself if a 128-bit wide DDR bus is "dual channel" or not. I'm not going to argue semantics. I am, however, going to do the math and tell you that the Opteron paired with DDR333 provides 128*333/8 = 5328 MB/s of some seriously low-latency bandwidth. Oh yea and it scales with the number of processors too.

DISCLAIMER: I work at AMD but I am not speaking on behalf of the company.

Opteron's ECC supports scrubbing!

[Eric Smith]

Most if not all prior x86 chipsets that supported ECC did a fairly half-assed job of it. They didn't report the location of a detected error (whether correctable or not), but instead only the base address of a block of memory which contained the error.

It appears that the Opteron can report the actual address of any detected errors. Plus, it can report details of ECC errors in its caches.

But the coolest feature is that it supports memory scrubbing, a feature I'd previously not seen in a microprocessor or chipset since the iAPX 432 memory controller back in 1983.

When a SEU causes a single-bit error in a word of memory, the ECC is capable of correcting it when the word is read. But if that word doesn't get read again for a long time, it's possible that a second SEU might happen in the same word, which would then be an uncorrectable error. With memory scrubbing, the memory controller uses a small portion of the memory bandwidth to scan the entire memory, correcting any single-bit errors that are found, so that the probability of a two-bit (or more) uncorrectable error is greatly reduced.

My last several computers (including a dual Athlon using the 760MPX chip set, and a DEC Alpha) had ECC, but not scrubbing. I considered writing a Linux program to scrub the memory by direct access to /dev/mem, but this has the disadvantage of thrashing the processor's caches. By implementing scrubbing in hardware, the Opteron avoids that problem.

The Opteron has a Scrub Control Register that is used to enable or disable scrubbing and control the rate. There are independent scrubbing controls for the L1 data cache, L2 cache, and main memory.

Those of us that want high reliability really welcome this feature. Well done, AMD!

By the way, it should be noted that it is typical for a PC with 128 megabytes of memory to get a single bit error several times a week. On my Alpha, I routinely saw corrected error log messages in the syslog, which gave me much more confidence in the system than the way that most PCs simply fail to even detect memory errors, let alone correct them. The log messages are also useful in that you can determine whether you have some memory that is getting marginal. For instance, at one point I started getting a much higher rate of corrected errors on one particular SIMM. There may have been a slight amount of oxidation or corrosion on the contacts, or they may have just worked themselves loose a bit. Cleaning the contacts and reseating the SIMM solved the problem with only a few minutes of down time, instead of what probably would have been hours of down time had the errors gone unnoticed.

The results of an undetected error vary considerably; it may be in memory that is not in use at the time, or it could be in the midst of the operating system, an application, or user data.