Naturally, the first question is: “How do they do that?”
To get all the chips into the DIMM format Samsung uses TSV interconnects on the DRAMs. The module’s 36 DRAM packages each contain four 8Gb (1GB) chips, resulting in 144 DRAM chips squeezed into a standard DIMM format. Each package also includes a data buffer chip, making the stack very closely resemble either the High-Bandwidth Memory (HBM) or the Hybrid Memory Cube (HMC).
Since these 36 packages (or worse, 144 DRAM chips) would overload the processor’s address bus, the DIMM uses an RDIMM protocol – the address and control pins are buffered on the DIMM before they reach the DRAM chips, cutting the processor bus loading by an order of magnitude or more. RDIMMs are supported by certain server platforms.
Everspin and Northwest Logic have just announced full interoperability between Northwest Logic’s MRAM Controller Core and Everspin Technologies’ ST-MRAM (Spin-Torque Magnetic RAM) chips. This interoperability is hardware proven on a Xilinx Virtex-7 FPGA and is now available for designs needing low-latency, high memory throughput using MRAM technology.
Since The Memory Guy knew that Everspin’s EMD3D064M ST-MRAM was fully DDR3 compatible, I had to wonder why the part would require a special controller – couldn’t it simply be controlled by any DDR3 controller?
Everspin’s product marketing director, Joe O’Hare, took the time to Continue reading
Intel and Micron today announced that the new version of Intel’s Xeon Phi, a highly parallel coprocessor for research applications, will be built using a custom version of Micron’s Hybrid Memory Cube, or HMC.
This is only the second announced application for this new memory product – the first was a Fujitsu supercomputer back in November.
For those who, like me, were unfamiliar with the Xeon Phi, it’s a module that uses high core-count processors for problems that can be solved with high degrees of parallelism. My friend and processor guru Nathan Brookwood tells me Continue reading
Spansion recently introduced a NOR flash that the company boasts is the: “World’s fastest NOR flash memory”. Named HyperFlash, the chip taps into high-speed SPI interface, doubling its width and adding a differential clock to run at an I/O rates as high as 333MB/s.
In this post’s graphic (click to enlarge) Spansion compares the HyperFlash chip’s sustained read rate (right-hand column) to that of (from left to right) asynchronous parallel NOR, single-bit SPI, industry-standard DDR Quad SPI, and Spansion’s faster rendition of DDR Quad SPI, which Spansion tells us, until now, has been the fastest flash on the market. The company points out that HyperFlash is five times the speed of industry-standard Continue reading
Rambus and Micron announced on Tuesday that they have signed a patent cross license agreement. Micron receives rights to Rambus IC patents, including memories. Both Micron and Elpida products will be covered. The companies have thus settled all outstanding patent and antitrust claims in their 13-year court battle.
Micron will make royalty payments to Rambus of up to $10 million per quarter over the next seven years, totaling $280 million, after which Micron will receive a perpetual, paid-up license.
Rambus and Micron both have Continue reading
Today Micron Technology announced that it is sampling the Hybrid Memory Cube (HMC) a DRAM packaging technology that it has been working on with the HMC Consortium.
Micron has been pushing to rapidly advance the HMC’s development and seems to have reached this point in an impressively brief time, given the complexity of the technology. It has only been two years since the first public appearance of the HMC at the 2011 Intel Developer Forum.
Some pretty advanced technology was used to make this product. DRAM processes are not very good at Continue reading
Some recent news mentioned cMLC flash, which is short for “consumer MLC.” This term is used to differentiate between the cheapest available product, mainstream MLC, and products that are aimed at the computing segment, and thus carry higher price tags.
There are several of these higher-end products. Some have longer endurance, like eMLC and SLC flash. Some have faster interfaces, like ONFi and Toggle Mode. Then there are the combinations of these: a fast interface with enhanced reliability.
There are disadvantages to these. The consumer market Continue reading
The new product is said to deliver the capacity and performance of an SO-DIMM in a 16x16mm BGA. It is built using Invensas’ xFD technology.
I have seen examples of Invensas’ xFD and the first thought that struck me was: “Why didn’t I think of that?!?” It’s an elegantly simple approach to today’s connection conundrums. By staggering chips and mounting them face-down over holes for bonding wires the company connects DRAMs with far shorter interconnect lengths and less scrambling, leading to higher performance.
Although this technology is not yet covered in any of our current reports, we do have a report on cell phone packages: Flash Packaging: What Phone Makers Want and Why, that can be purchased for immediate download on the Objective Analysis website.
This product appears to be a variation on the Hybrid Memory Cube, or HMC concept detailed in a prior post.
Remember that the HMC stacks a number of DRAM chips atop a logic chip. The memories store data and communicate to the logic chip through thousands of through-silicon vias (TSVs) while the logic chip handles communications with the outside world. Continue reading
I got a phone call yesterday from Russell Fish of Venray Technology. He wanted to talk about how and why computer architecture is destined for a change.
I will disclose right up front that he and I were college classmates. Even so, I will do my best to give the unbiased viewpoint that my clients expect of me.
Russell is tormented by an affliction that troubles many of us in technology: We see the direction that technology is headed, then we consider what makes sense, and we can’t tolerate any conflicts between the two.
In Russell’s case, the problem is the memory/processor speed bottleneck.