USA: Memory manufacturers have put more effort into developing alternative solutions to DRAM and flash memory as the struggle to scale DRAM capacitors and flash memory storage cells nears the physical limitations using current process technologies. Consequently, alternative memory solutions featuring storage techniques other than a capacitor or floating gate are being considered.

Several different specialty memories targeting specific end uses are currently available. Though none is likely to challenge volume products anytime soon, all hold at least the possibility of providing the characteristics that could make a universal memory including fast read and write with low power; non-volatility with very long data retention and cycle life; and small cell size.

Magnetoresistive RAM (MRAM), Ferroelectric RAM (FRAM), and Phase-change RAM (PCRAM) are among the most talked about options to DRAM and flash memory. As noted in the figure, each technology has some decided advantages over current DRAM and flash memory options.  However, each also has some drawbacks from a cost or technology standpoint.

Resistance (resistive) RAM ReRAM/RRAM is another possible solution. Fujitsu, Intel, Samsung, Sharp, and Spansion are working on this technology. These companies believe ReRAM will be 100x faster than flash, yet scale much better than other advanced memories like phase-change RAM or MRAM.

An option not listed in the chart but showing considerable promise is carbon nanotube (CNT) technology.  This could be the answer for system memory near-term and complex logic farther out. Nanotube-based RAM (NRAM) cells are constructed using several CNTs suspended above a metal electrode. When a small voltage bias is applied to the tubes, they "sag' toward the electrode until making contact. At that point, the tubes are considered in the logic 1 state.

When the voltage bias is removed, they pull back away from the electrode and the logic state becomes 0 once again.  The benefits of NRAM include having the speed of SRAM, densities far exceeding DRAM, and lower power consumption than DRAM or flash. Also, it stands up well to harsh environments and scales well.

Additionally, the recent announcement of the "memristor"—the circuit element that improves in performance as you scale it down—to smaller sizes could be the technology breakthrough that memory suppliers hope for to create a universal memory.

IC Insights continues to believe that the memory market (particularly, DRAM) is evolutionary, not revolutionary.  Memory suppliers will push all they can out of existing technology/architectures before having to move on to an alternative such as nanotubes, PCRAM, or MRAM.  So far, no alternative memory offers such a compelling solution that makes it better than anything currently existing.