RAM Parallelism: Single And Dual Channel

Ad

Some Memory History

Developments in system main memory (also known as RAM - random access memory) had been mostly linear until AMD and Intel introduced dual channel controllers in 2003. In the server space, you can also find Xeon platforms (Bensley or the latest Stoakley platform for 45 nm processors), which utilize a quad channel memory controller.

Memory modules, as opposed to installing individual memory chips, were introduced to facilitate memory deployment in the 1990s. The first Single Inline Memory Modules (SIMMs) had 30 pins and were eight bits wide, which meant that pairs had to be used for 286 and 386SX computers (16 bit machines), and four SIMMs were required for 386DX systems and up (32 bit architectures). 30-pin SIMM and SIPP modules were available at 256 kB to 4 MB each, and they were replaced by the 72-pin PS/2 SIMMs in the mid ’90s. The fact that at least two or four modules had to be used has nothing to do with parallelism; it is only because the system bus width had to be matched.

72-pin SIMMs were used for fast page mode DRAM (FPM), which was quickly replaced by Extended Data Out (EDO) memory in the late 1990s. Although 64 MB PS/2 SIMMs existed, they typically maxed out at 32 MB per module. EDO delivers better read performance when multiple data is read out of a page, where the row address doesn’t have to be changed. EDO reached a peak bandwidth of 266 MB/s.

EDO was replaced by synchronous DRAM (SDRAM) on 128-pin DIMMs (64-bit data bus at 3.3 V), where the clock is defined by the system bus or the memory controller. First generation PC66 memory was already twice as fast as EDO-DRAM, and following generations scaled nicely: PC100 and PC133 became popular. After that, double data rate (DDR) SDRAM was introduced, using 184-pin DDR-DIMMs. These reduced the voltage to 2.5 V and doubled performance by transferring data on the rising as well as falling edge of the clock signal, at up to 200 MHz base clock speeds (DDR400). DDR2 memory on 240-pin DIMMs as well as DDR3 is still based on this same technology, but offers a larger prefetch and much increased clock speeds of up to 400 MHz with DDR2 (DDR2-800) and projected 800 MHz for DDR3 (DDR3-1600).

All of these technologies worked on a single memory channel, which means they increased bandwidth over previous generations by widening the memory bus and by accelerating memory speed.


Talkback
uk_gangsta 18/10/2007 07:28
Hide
-0+
uk_gangsta

cock

americanbrian 19/10/2007 10:01
Hide
-0+
americanbrian

for the raid zero test you really should have tried transferring files over a 1 Gbit network. This is where it rocks. Or even better, have a home network client stream video over the network from the test setups while the user on the raid 0 or single drive rig does other stuff, like play games, or also watches a different video.

You seem to have missed the point as to the practical advantages of a enthusiast raid 0 rig. some common situations are really helped out by better storage performance.

leexgx 19/10/2007 05:54
Hide
-0+
leexgx

hmm

leexgx 19/10/2007 06:01
Hide
-0+
leexgx

i respond to raid 0, 2 and 4 disks setup

Most built in raid seem to be stuck at max bandwith of 133 (basicly UDMA 133 IDE speeds spec and thats Total for all 6 ports combined in some caes) so when you add 1-2 more disk speed does not increase much or at all

bobwya 19/10/2007 07:09
Hide
-0+
bobwya

Bulls**t article...

Visa a vie raid-0 on HD. Try chucking big data about - that won't be cached in RAM... I do this a lot and bless striping everytime (linear access flies)! Sharing over a network (especially gigabit) benefits as well of course...

Does the author actually understand how computers work? Using CPU bound applications to demonstrate that raiding drives is a waste of time... !! What have the THG authors been smoking these days?? Do you guys have any formal qualifications in Computer Science or anything??

Pointless article...

Now how about an article about the guy who has hit 6Ghz recently with the new 45nm Intel chips...

Or an article about striping 8 266x CF cards with SATA-CF(ide) converters and booting Windows of it?? That would be interesting...

What happened to THG of old with the videos of liquid nitrogen experiments...

A big yawn to the author...

Bob


Tattysnuc 20/10/2007 11:36
Hide
-0+
Tattysnuc

I hate to be siding with the negative comments, but I have to agree with BobWya.

Where are all the EXCITING/INTERESTING articles nowadays? I can't remember the last time I saw a NEW video that was worth watching all the way through.

3dGameman is doing a cracking job of doing some h/w reviews. Nothing destructive, but worth checking out. Tom's seem to have gone on a green crusade of late, and while the solar panel pc is an interesting testbed, and pioneering, the majority of us would not build such a thing.

Enthusiast is a strong word, but I really thing that Tom's have gone completely mainstream, and with the diversification of the page layouts across the world wide sites, I'm finding myslef spending much more time in the madshrimps, hexus and ntcompatible sites as they still have something new to offer.

The next evolution in PC performance, imo, is the solid state drive. A week or so ago you wet our whiskers with an article that touched on the concept and basics. How about a more comprehensive breakout detailing the parts.

And please stop this childish blockade against raptor drives. They may well have some ideosynchracies, but there are so many of us out here that recognise them for what they are, surely you would be better off if you were to still include them and show the problem. If anything is going to get WD to fix the problem it'll be someone showing problems, rather than a site ignoring their product. New readers to your site, or people who missed your article wont have a clue as to why you dont use it, unless they actually look for the problem.

Just my pennies worth....

X3Style 20/10/2007 06:54
Hide
-0+
X3Style

Sad but true THG is dieing a ferry slow and painfully boring death.

I think u guys need to recruit new talents with new ideas.

We all know u cant come up with good things forever.

Stop trying to be irrational.

As i noticed recently ur just making articles for the sake of making them. And seriously noone cares about articles showing nothing changes.

But everyone wants to see articles that show a difference.

I am an old fan of THG and i dont want to see u guys go down like this.

Best of Luck.

Note You are going to post a comment as anonymous.



Google Ads