Researchers Testing Green ‘Memory On Demand’

Scientists at an American University are touting new memory technology that can switch off parts of a computer’s memory not being accessed by the system, in order to drastically lower power consumption.

The claim come from researchers at Carolina State University and was reported in EE Times and outlined in the journal IEEE Computer.

The universal memory technology is reportedly still undergoing testing, but claims to combine the advantages of two common forms of memory used today, namely the speed of DRAM with the non-volatility and density of flash memory.

The explanation about how it works is a bit on the technical side, so those with no techy inclination should look away now.

The Techy Bit

The technology uses a double floating-gate field-effect-transistor (FET) that is apparently as fast as DRAM, but will not need to be refreshed as often. Professor Paul Franzon at NC State told the EE Times that, despite this, the densities will be about the same as flash.

The double floating-gates use direct tunneling when storing charge to represent bits – instead of hot electron injection like flash – thus enabling operation at lower voltages. The first floating-gate in the stack is leaky, thus requiring refreshing about as often as DRAM (16 milliseconds). But by increasing the voltage its data value can be transferred to the second floating-gate, which acts more like a traditional flash memory, offering long-term nonvolatile storage.

In operation, those computers that are using the double floating-gate FETs for their main memory can operate normally until they become idle, at which time their data values can be transferred to the second gate in order to power down the memory chip. Then when the stored values need to be accessed again by the computer, the second gate quickly transfers their stored charge back to the first gate and normal operations can resume.

And Now In English

What this means in plain English is that the computer can still operate normally until it becomes idle.  Then the data values are transferred to the second gate in order to power down the memory chip.

When the computer needs the stored values, the second gate quickly transfers their stored charge back to the first gate and normal operations can resume.

According to Professor Franzon, this new memory technology should enable computers to power down memories not currently being accessed, so called ‘power-proportional computing’, as it allows  memory to be turned off during periods of low use.

This will result in drastically cutting the machine’s energy consumption, and when this technology is being utilised not just in normal desktop PC fleets, but also in large server farms and data centres, the potential energy savings become increasingly compelling.

“We believe our new memory device will enable power-proportional computing, by allowing memory to be turned off during periods of low use without affecting performance,” EE Times quoted Professor Franzon as saying.

Power Concerns

The team also believe it could also be used for PCs that are required to start immediately – so called ‘instant on’ computers.

The focus on making computers ever-more energy efficient is gathering steam, but this seems to be the first attempt at utilising memory technology as a potential power saving option. In September, Gartner warned that energy-related costs makes up 12 percent of data centre expenditure.

Of course one of the biggest power consuming culprits are those PCs which are left on overnight, and for those corporations looking to cut their power bills, the advice is to make sure all PCs are powered down after the working day ends.