AMD touts LRDIMM memory for x86 servers
Pushing up to 384GB per socket
It looks like Advanced Micro Devices is first to market with support for load reduced DIMM DDR3 main memory for x86 and quite possibly all kinds of servers, and is trotting out Inphi, the maker of the isolation memory buffer chip that is at the heart of this technology.
With LRDIMM memory, you take out the register on a DDR3 memory module and plunk in a buffer chip that allows the memory chips on the module to run at a higher clock speed and also allows for more memory chips to be put on each channel.
When the Opteron 4200s and 6200s were launched last November, AMD said that it would support unregistered, registered, and load reduced DDR3 memory modules with the chips. To add LRDIMM support, the main memory controllers on the Opterons had to be tweaked so they could talk this new chatter and support the higher clock speeds on the memory. What might not have been obvious at the time was that the extra memory that AMD was offering was only available through the use of LRDIMM memory.
As it stands, says John Fruehe, director of product marketing for server, embedded, amd FireStream products at AMD, if you use unregistered DIMMs, you top out at 64GB per socket, and registered DIMMs are at the 256GB per socket upper limit, twice the maximum memory as the prior generation of Opteron 6100 processors. By moving to LRDIMMs, you can boost that memory capacity to 384GB per socket without having to resort to special memory extender electronics that add costs to the server.
At the moment, AMD has put the Inphi iMB buffer chip through the testing paces and has certified it as compliant with its Opteron 6200s. While the Opteron 4200s can technically support LRDIMM memory, AMD doesn't expect customers to do that. But they might, given the low-power server buyers and builders that like the Opteron 4200 chips and that for a given amount of memory, LRDIMMs burn less juice.
Paul Washkewicz, vice president of marketing at Inphi, tells El Reg that if you compare a 1.5 volt RDIMM with 16GB of capacity, you can move to a 1.35 volt LRDIMM with 32GB of capacity and burn 20 per cent less juice. And Washkewicz says that Samsung, Elpida, Hynix, and Micron are getting close to price parity between RDIMM and LRDIMM memory at identical capacity, so you don't have to pay a big premium, either.
Intel is expected to support LRDIMM memory in its forthcoming "Sandy Bridge-EP" Xeon E5 processors, which launch this quarter. ®
All nice if you can find them
Well, let's have a look. AMD new fusion APUs, 64GB per socket, motherboards support 64GB per socket (Asus F-series), so let's go find some memory... Nope. Nada. Nil. Zilch.
All that talk about more memory is nice, but nobody is walking the walk.
For the time being the BIG DIMM capacities are mostly in the mythical range. Once you go past 8GB per DIMM the options are scarce and extortionately priced (with or without extender in the equation). Once you go past 16G per DIMM you are definitely in la-la land.
I'll be blunt - can I see those LRDIMMs please. Crucial catalogue number or at the very worst a vendor part number. Nope I cannot. So well, then let's move along then...
Inphi comments at Stifel Nicolaus conference 3
Inphi pushing out LRDIMMs to "Ivy Bridge" - perhaps because at 32GB and 64GB they have a better chance to compete with RDIMMs (i.e. the "high latency" issues become less important if RDIMMs are going to be 4-rank for 32GB etc.) - even though the 32GB LRDIMMs cannot do 3 DPC at 1333MHz. However all that discussion assumes there is no third party Netlist IP providing memory modules that outdo LRDIMMs, and also have none of the "high latency" issues of LRDIMMs.
at the 17:35 minute mark ..
But I think once the .. uh .. once the Romley systems get out and people are able to verify that you can HAVE 50% more virtual machines or you could have, you know, better throughput, why wouldn't they go with the LRDIMM.
They are going to run in that direction.
And .. uh .. we think that's all good .. uh .. we think as Ivy Bridge comes in you'll see more .. uh .. applications of LRDIMM ..
Here Inphi suggests that DDR4 with their decentralized architecture is "same architecture as LRDIMM" centralized buffer (628-pin buffer chip) - Netlist has been saying for some time that DDR4 intersects NLST IP - and articles by same author is also suggestive of that: "Netlist puffs HyperCloud DDR3 memory to 32GB - DDR4 spec copies homework"):
The good news is when we get to DDR4 people are more interested in gravitating towards the LRDIMM .. uh .. architecture, where people can choose to buy a separate register chip and as many buffers that the would like.
And .. uh .. it allows for a more ubiquitous .. uh .. implementation of of the same .. uh .. architecture as LRDIMM - you just do away with 2 independent products and once product can scale into .. into what anybody might need.
Inphi comments at Stifel Nicolaus conference 2
However, Inphi does convey an uncertainty about the demand for LRDIMM - perhaps correctly suggesting that OEMs have little interest in pushing memory that would obviate need to buy more servers (if adding memory will allow to create more virtual machines on same servers etc.):
at the 16:10 minute mark ..
Tore Svanberg - Stifel Nicolaus (Analyst):
And, you know, you showed a chart, I think it was maybe an IVC (?) chart looking at you know how 32GB and then eventually 64GB will ramp.
Umm .. but when you talk to your biggest customers, let's say, you know Micron and Samsung, I mean how how are they looking at that type of ramp.
Uh .. is it very similar .. uh .. and you know what types of penetrations are they talking about both this year and next year ?
John Edmunds - CFO:
Umm .. so .. it's difficult in the supply chain to get a lot of .. uh .. uh .. forecast.
Uh .. because essentially 32GB it's a new product altogether.
So in in general people don't know how demand there will be for that.
So we actually think it'll become .. uh .. uh .. it'll become demand .. it will be pull driven in effect, because you will have CIOs who are saying I'm going to order a batch of Romley systems, but I want them to be the LRDIMM configured, because I've tried that .. I can see that really benefits my application.
Uh .. and .. because .. for two reasons, the new product and because .. uh .. the the system and memory guys don't know how many customers (are) out there - they are going to call for that.
They are conservative right now on what their thinking will will actually be the demand or what will be the shipment for ..
at the 17:25 minute mark ..
As I showed you earlier on that one example, there is also less hardware to ship if you are shipping that kind of configuration.
So I'm not sure anybody's out there banging the bush sayin' "hey do this and buy less hardware" right ? It's a little bit of an anomaly in that sense.