Racetrack memory
Encyclopedia
Racetrack memory is an experimental non-volatile memory
device under development at IBM
's Almaden Research Center
by a team led by Stuart Parkin
. In early 2008, a 3-bit version was successfully demonstrated. If it is developed successfully, racetrack would offer storage density
higher than comparable solid-state memory devices like flash memory
and similar to conventional disk drives, and also have much higher read/write performance. It is one of a number of new technologies trying to become a universal memory
in the future.
-coherent electric current
to move magnetic domains along a nanoscopic permalloy
wire about 200 nm across and 100 nm thick. As current is passed through the wire, the domains pass by magnetic read/write heads positioned near the wire, which alter the domains to record patterns of bits. A racetrack memory device is made up of many such wires and read/write elements. In general operational concept, racetrack memory is similar to the earlier twistor memory
or bubble memory
of the 1960s and 1970s. Delay line memory
, such as mercury delay lines of the 1940s and 1950s are a still earlier form of similar technology, as used in the UNIVAC
and EDSAC
computers. Like bubble memory, racetrack memory uses electrical currents to "push" a magnetic pattern through a substrate. Dramatic improvements in magnetic detection capabilities, based on the development of spintronic magnetoresistive sensing materials and devices, allow the use of much smaller magnetic domains to provide far higher areal densities.
In production, it is expected that the wires can be scaled down to around 50 nm. There are two ways to arrange racetrack memory. The simplest is a series of flat wires arranged in a grid with read and write heads arranged nearby. A more widely studied arrangement uses U-shaped wires arranged vertically over a grid of read/write heads on an underlying substrate. This allows the wires to be much longer without increasing its 2D area, although the need to move individual domains further along the wires before they reach the read/write heads results in slower random access times. This does not present a real performance bottleneck; both arrangements offer about the same throughput. Thus the primary concern in terms of construction is practical; whether or not the 3D vertical arrangement is feasible to mass produce.
. The authors of the primary work also discuss ways to improve the access times with the use of a "reservoir," improving to about 9.5 ns. Aggregate throughput, with or without the reservoir, is on the order of 250-670 Mbit/s for racetrack memory, compared to 102400 Mbit/s for dual channel DDR2 DRAM, 1000 Mbit/s for high-performance hard drives, and much slower performance on the order of 30 to 100 Mbit/s for flash memory devices. The only current technology that offers a clear performance benefit over racetrack memory is SRAM
, on the order of 2 ns, but is much more expensive and far lower density.
Flash memory, in particular, is a highly asymmetrical device. Although read performance is fairly fast, especially compared to a hard drive, writing is much slower. Flash memory works by "trapping" electron
s in the chip surface, and requires a burst of high voltage to remove this charge and reset the cell. In order to do this, charge is accumulated in a device known as a charge pump
, which takes a relatively long time to charge up. In the case of NOR flash memory, which allows random bit-wise access like racetrack memory, read times are on the order of 70 ns, while write times are much slower, about 2,500 ns. To address this concern, NAND flash memory allows reading and writing only in large blocks, but this means that the time to access any random bit is greatly increased, to about 1,000 ns. In addition, the use of the burst of high voltage physically degrades the cell, so most flash devices allow on the order of 100,000 writes to any particular bit before their operation becomes unpredictable. Wear leveling
and other techniques can spread this out, but only if the underlying data can be re-arranged.
The key determinant of the cost of any memory device is the physical size of the storage medium. The reason for this is due to the way memory devices are fabricated. In the case of solid-state devices like flash memory or DRAM, a large "wafer" of silicon is processed into many individual devices, which are then cut apart and packaged. The cost of packaging is about $1 per device, so, as the density increases and the number of bits per devices increases with it, the cost per bit falls by an equal amount. In the case of hard drives, data is stored on a number of rotating platters, and the cost of the device is strongly related to the number of platters. Increasing the density allows the number of platters to be reduced for any given amount of storage.
In most cases, memory devices store one bit in any given location, so they are typically compared in terms of "cell size", a cell storing one bit. Cell size itself is given in units of F², where F is the design rule
, representing usually the metal line width. Flash and racetrack both store multiple bits per cell, but the comparison can still be made. For instance, modern hard drives appear to be rapidly reaching their current theoretical limits around 650 nm²/bit, which is defined primarily by our capability to read and write to tiny patches of the magnetic surface. DRAM has a cell size of about 6 F², SRAM is much worse at 120 F². NAND flash memory is currently the densest form of non-volatile memory in widespread use, with a cell size of about 4.5 F², but storing three bits per cell for an effective size of 1.5 F². NOR flash memory is slightly less dense, at an effective 4.75 F², accounting for 2-bit operation on a 9.5 F² cell size. In the vertical orientation (U-shaped) racetrack, about 10-20 bits are stored per cell, which itself can have a physical size of at least about 20 F². In addition, bits at different positions on the "track" would take different times (from ~10 ns to nearly a microsecond, or 10 ns/bit) to be accessed by the read/write sensor, because the "track" is moved at fixed rate (~100 m/s) past the read/write sensor.
Racetrack memory is one of a number of new technologies aiming to replace flash memory, and potentially offer a "universal" memory device applicable to a wide variety of roles. Other leading contenders include MRAM, PCRAM
and FeRAM. Most of these technologies offer densities similar to flash memory, in most cases worse, and their primary advantage is the lack of write endurance limits like those in flash memory. Field-MRAM offers excellent performance as high as 3 ns access time, but requires a large 25-40 F² cell size. It might see use as an SRAM replacement, but not as a mass storage device. The highest densities from any of these devices is offered by PCRAM, which has a cell size of about 5.8 F², similar to flash memory, as well as fairly good performance around 50 ns. Nevertheless, none of these can come close to competing with racetrack memory in overall terms, especially density. For example, 50 ns allows about five bits to be operated in a racetrack memory device, resulting in an effective cell size of 20/5=4 F², easily exceeding the performance-density product of PCM. On the other hand, without sacrificing bit density, the same 20 F² area can also fit 2.5 2-bit 8 F² alternative memory cells (such as RRAM or spin-torque transfer MRAM
), each of which individually operating much faster (~10 ns).
A difficulty for this technology arises from the need for high current density
(>108 A/cm²); a 30 nm x 100 nm cross-section would require >3 mA. The resulting power draw would be higher than, for example, spin-torque transfer memory or flash memory.
has traced this problem to microscopic imperfections in the crystal structure of the wires which led to the domains becoming "stuck" at these imperfections. Using an X-ray microscope
to directly image the boundaries between the domains, their research found that domain walls would be moved by pulses as short as a few nanoseconds when these imperfections were absent. This corresponds to a macroscopic performance of about 110 m/s.
The voltage required to drive the domains along the racetrack would be proportional to the length of the wire. The current density must be sufficiently high to push the domain walls (as in electromigration
). For example, a permalloy racetrack of resistivity
5×10-7 ohm-m, that is 1 cm long to cover an entire chip array, and uses a current density of 3×108 A/cm², would require a driving voltage of 15 kV along the racetrack.
Non-volatile memory
Non-volatile memory, nonvolatile memory, NVM or non-volatile storage, in the most basic sense, is computer memory that can retain the stored information even when not powered. Examples of non-volatile memory include read-only memory, flash memory, ferroelectric RAM, most types of magnetic computer...
device under development at IBM
IBM
International Business Machines Corporation or IBM is an American multinational technology and consulting corporation headquartered in Armonk, New York, United States. IBM manufactures and sells computer hardware and software, and it offers infrastructure, hosting and consulting services in areas...
's Almaden Research Center
Almaden Research Center
The IBM Almaden Research Center is in San Jose, California, and is one of IBM's nine worldwide research labs. Its scientists perform basic and applied research in computer science, services, storage systems, physical sciences, and materials science and technology. The center opened in 1986, and...
by a team led by Stuart Parkin
Stuart Parkin
Stuart Stephen Papworth Parkin, Ph.D. is an experimental physicist, IBM Fellow and manager of the magnetoelectronics group at the IBM Almaden Research Center in San Jose, California. He is also a consulting professor in the Department of Applied Physics at Stanford University and director of the...
. In early 2008, a 3-bit version was successfully demonstrated. If it is developed successfully, racetrack would offer storage density
Computer storage density
Memory storage density is a measure of the quantity of information bits that can be stored on a given length of track, area of surface, or in a given volume of a computer storage medium. Generally, higher density is more desirable, for it allows greater volumes of data to be stored in the same...
higher than comparable solid-state memory devices like flash memory
Flash memory
Flash memory is a non-volatile computer storage chip that can be electrically erased and reprogrammed. It was developed from EEPROM and must be erased in fairly large blocks before these can be rewritten with new data...
and similar to conventional disk drives, and also have much higher read/write performance. It is one of a number of new technologies trying to become a universal memory
Universal memory
Universal memory may mean:* any memory device combining cost benefits of DRAM, speed of SRAM, and non-volatility of flash memory** magnetoresistive random-access memory ** Bubble memory** Racetrack memory** ferroelectric random-access memory...
in the future.
Description
Racetrack memory uses a spinSpin (physics)
In quantum mechanics and particle physics, spin is a fundamental characteristic property of elementary particles, composite particles , and atomic nuclei.It is worth noting that the intrinsic property of subatomic particles called spin and discussed in this article, is related in some small ways,...
-coherent electric current
Electric current
Electric current is a flow of electric charge through a medium.This charge is typically carried by moving electrons in a conductor such as wire...
to move magnetic domains along a nanoscopic permalloy
Permalloy
Permalloy is a nickel-iron magnetic alloy, with about 20% iron and 80% nickel content. It is notable for its very high magnetic permeability, which makes it useful as a magnetic core material in electrical and electronic equipment, and also in magnetic shielding to block magnetic fields...
wire about 200 nm across and 100 nm thick. As current is passed through the wire, the domains pass by magnetic read/write heads positioned near the wire, which alter the domains to record patterns of bits. A racetrack memory device is made up of many such wires and read/write elements. In general operational concept, racetrack memory is similar to the earlier twistor memory
Twistor memory
Twistor is a form of computer memory, similar to core memory, formed by wrapping or closing magnetic tape around a current-carrying wire. Although the developers, Bell Labs, had high hopes for Twistor, it was used for only a brief time in the marketplace between about 1968 and the mid-1970s...
or bubble memory
Bubble memory
Bubble memory is a type of non-volatile computer memory that uses a thin film of a magnetic material to hold small magnetized areas, known as bubbles or domains, each storing one bit of data...
of the 1960s and 1970s. Delay line memory
Delay line memory
Delay line memory was a form of computer memory used on some of the earliest digital computers. Like many modern forms of electronic computer memory, delay line memory was a refreshable memory, but as opposed to modern random-access memory, delay line memory was serial-access...
, such as mercury delay lines of the 1940s and 1950s are a still earlier form of similar technology, as used in the UNIVAC
UNIVAC
UNIVAC is the name of a business unit and division of the Remington Rand company formed by the 1950 purchase of the Eckert-Mauchly Computer Corporation, founded four years earlier by ENIAC inventors J. Presper Eckert and John Mauchly, and the associated line of computers which continues to this day...
and EDSAC
EDSAC
Electronic Delay Storage Automatic Calculator was an early British computer. The machine, having been inspired by John von Neumann's seminal First Draft of a Report on the EDVAC, was constructed by Maurice Wilkes and his team at the University of Cambridge Mathematical Laboratory in England...
computers. Like bubble memory, racetrack memory uses electrical currents to "push" a magnetic pattern through a substrate. Dramatic improvements in magnetic detection capabilities, based on the development of spintronic magnetoresistive sensing materials and devices, allow the use of much smaller magnetic domains to provide far higher areal densities.
In production, it is expected that the wires can be scaled down to around 50 nm. There are two ways to arrange racetrack memory. The simplest is a series of flat wires arranged in a grid with read and write heads arranged nearby. A more widely studied arrangement uses U-shaped wires arranged vertically over a grid of read/write heads on an underlying substrate. This allows the wires to be much longer without increasing its 2D area, although the need to move individual domains further along the wires before they reach the read/write heads results in slower random access times. This does not present a real performance bottleneck; both arrangements offer about the same throughput. Thus the primary concern in terms of construction is practical; whether or not the 3D vertical arrangement is feasible to mass produce.
Comparison to other memory devices
Current projections suggest that racetrack memory will offer performance on the order of 20-32 ns to read or write a random bit. This compares to about 3,000,000 ns for a hard drive, or 6-40 ns for conventional DRAMDram
Dram or DRAM may refer to:As a unit of measure:* Dram , an imperial unit of mass and volume* Armenian dram, a monetary unit* Dirham, a unit of currency in several Arab nationsOther uses:...
. The authors of the primary work also discuss ways to improve the access times with the use of a "reservoir," improving to about 9.5 ns. Aggregate throughput, with or without the reservoir, is on the order of 250-670 Mbit/s for racetrack memory, compared to 102400 Mbit/s for dual channel DDR2 DRAM, 1000 Mbit/s for high-performance hard drives, and much slower performance on the order of 30 to 100 Mbit/s for flash memory devices. The only current technology that offers a clear performance benefit over racetrack memory is SRAM
Static random access memory
Static random-access memory is a type of semiconductor memory where the word static indicates that, unlike dynamic RAM , it does not need to be periodically refreshed, as SRAM uses bistable latching circuitry to store each bit...
, on the order of 2 ns, but is much more expensive and far lower density.
Flash memory, in particular, is a highly asymmetrical device. Although read performance is fairly fast, especially compared to a hard drive, writing is much slower. Flash memory works by "trapping" electron
Electron
The electron is a subatomic particle with a negative elementary electric charge. It has no known components or substructure; in other words, it is generally thought to be an elementary particle. An electron has a mass that is approximately 1/1836 that of the proton...
s in the chip surface, and requires a burst of high voltage to remove this charge and reset the cell. In order to do this, charge is accumulated in a device known as a charge pump
Charge pump
A charge pump is a kind of DC to DC converter that uses capacitors as energy storage elements to create either a higher or lower voltage power source. Charge pump circuits are capable of high efficiencies, sometimes as high as 90–95% while being electrically simple circuits.Charge pumps use some...
, which takes a relatively long time to charge up. In the case of NOR flash memory, which allows random bit-wise access like racetrack memory, read times are on the order of 70 ns, while write times are much slower, about 2,500 ns. To address this concern, NAND flash memory allows reading and writing only in large blocks, but this means that the time to access any random bit is greatly increased, to about 1,000 ns. In addition, the use of the burst of high voltage physically degrades the cell, so most flash devices allow on the order of 100,000 writes to any particular bit before their operation becomes unpredictable. Wear leveling
Wear leveling
Wear leveling is a technique for prolonging the service life of some kinds of erasable computer storage media, such as Flash memory used in solid-state drives and USB Flash drives...
and other techniques can spread this out, but only if the underlying data can be re-arranged.
The key determinant of the cost of any memory device is the physical size of the storage medium. The reason for this is due to the way memory devices are fabricated. In the case of solid-state devices like flash memory or DRAM, a large "wafer" of silicon is processed into many individual devices, which are then cut apart and packaged. The cost of packaging is about $1 per device, so, as the density increases and the number of bits per devices increases with it, the cost per bit falls by an equal amount. In the case of hard drives, data is stored on a number of rotating platters, and the cost of the device is strongly related to the number of platters. Increasing the density allows the number of platters to be reduced for any given amount of storage.
In most cases, memory devices store one bit in any given location, so they are typically compared in terms of "cell size", a cell storing one bit. Cell size itself is given in units of F², where F is the design rule
Design rule checking
Design Rule Checking or Check is the area of Electronic Design Automation that determines whether the physical layout of a particular chip layout satisfies a series of recommended parameters called Design Rules...
, representing usually the metal line width. Flash and racetrack both store multiple bits per cell, but the comparison can still be made. For instance, modern hard drives appear to be rapidly reaching their current theoretical limits around 650 nm²/bit, which is defined primarily by our capability to read and write to tiny patches of the magnetic surface. DRAM has a cell size of about 6 F², SRAM is much worse at 120 F². NAND flash memory is currently the densest form of non-volatile memory in widespread use, with a cell size of about 4.5 F², but storing three bits per cell for an effective size of 1.5 F². NOR flash memory is slightly less dense, at an effective 4.75 F², accounting for 2-bit operation on a 9.5 F² cell size. In the vertical orientation (U-shaped) racetrack, about 10-20 bits are stored per cell, which itself can have a physical size of at least about 20 F². In addition, bits at different positions on the "track" would take different times (from ~10 ns to nearly a microsecond, or 10 ns/bit) to be accessed by the read/write sensor, because the "track" is moved at fixed rate (~100 m/s) past the read/write sensor.
Racetrack memory is one of a number of new technologies aiming to replace flash memory, and potentially offer a "universal" memory device applicable to a wide variety of roles. Other leading contenders include MRAM, PCRAM
Phase-change memory
Phase-change memory is a type of non-volatile computer memory. PRAMs exploit the unique behavior of chalcogenide glass. Heat produced by the passage of an electric current switches this material between two states, crystalline and amorphous...
and FeRAM. Most of these technologies offer densities similar to flash memory, in most cases worse, and their primary advantage is the lack of write endurance limits like those in flash memory. Field-MRAM offers excellent performance as high as 3 ns access time, but requires a large 25-40 F² cell size. It might see use as an SRAM replacement, but not as a mass storage device. The highest densities from any of these devices is offered by PCRAM, which has a cell size of about 5.8 F², similar to flash memory, as well as fairly good performance around 50 ns. Nevertheless, none of these can come close to competing with racetrack memory in overall terms, especially density. For example, 50 ns allows about five bits to be operated in a racetrack memory device, resulting in an effective cell size of 20/5=4 F², easily exceeding the performance-density product of PCM. On the other hand, without sacrificing bit density, the same 20 F² area can also fit 2.5 2-bit 8 F² alternative memory cells (such as RRAM or spin-torque transfer MRAM
MRAM
Magnetoresistive Random-Access Memory is a non-volatile computer memory technology that has been under development since the 1990s. Continued increases in density of existing memory technologies – notably flash RAM and DRAM – kept it in a niche role in the market, but its proponents...
), each of which individually operating much faster (~10 ns).
A difficulty for this technology arises from the need for high current density
Current density
Current density is a measure of the density of flow of a conserved charge. Usually the charge is the electric charge, in which case the associated current density is the electric current per unit area of cross section, but the term current density can also be applied to other conserved...
(>108 A/cm²); a 30 nm x 100 nm cross-section would require >3 mA. The resulting power draw would be higher than, for example, spin-torque transfer memory or flash memory.
Development difficulties
One limitation of the early experimental devices was that the magnetic domains could be pushed only slowly through the wires, requiring current pulses on the orders of microseconds to move them successfully. This was unexpected, and led to performance equal roughly to that of hard drives, as much as 1000 times slower than predicted. Recent research at the University of HamburgUniversity of Hamburg
The University of Hamburg is a university in Hamburg, Germany. It was founded on 28 March 1919 by Wilhelm Stern and others. It grew out of the previous Allgemeines Vorlesungswesen and the Kolonialinstitut as well as the Akademisches Gymnasium. There are around 38,000 students as of the start of...
has traced this problem to microscopic imperfections in the crystal structure of the wires which led to the domains becoming "stuck" at these imperfections. Using an X-ray microscope
X-ray microscope
An X-ray microscope uses electromagnetic radiation in the soft X-ray band to produce images of very small objects.Unlike visible light, X-rays do not reflect or refract easily, and they are invisible to the human eye. Therefore the basic process of an X-ray microscope is to expose film or use a...
to directly image the boundaries between the domains, their research found that domain walls would be moved by pulses as short as a few nanoseconds when these imperfections were absent. This corresponds to a macroscopic performance of about 110 m/s.
The voltage required to drive the domains along the racetrack would be proportional to the length of the wire. The current density must be sufficiently high to push the domain walls (as in electromigration
Electromigration
Electromigration is the transport of material caused by the gradual movement of the ions in a conductor due to the momentum transfer between conducting electrons and diffusing metal atoms. The effect is important in applications where high direct current densities are used, such as in...
). For example, a permalloy racetrack of resistivity
Resistivity
Electrical resistivity is a measure of how strongly a material opposes the flow of electric current. A low resistivity indicates a material that readily allows the movement of electric charge. The SI unit of electrical resistivity is the ohm metre...
5×10-7 ohm-m, that is 1 cm long to cover an entire chip array, and uses a current density of 3×108 A/cm², would require a driving voltage of 15 kV along the racetrack.
See also
- Bubble memoryBubble memoryBubble memory is a type of non-volatile computer memory that uses a thin film of a magnetic material to hold small magnetized areas, known as bubbles or domains, each storing one bit of data...
- Giant magnetoresistance (GMR) effect
- Magnetoresistive Random Access Memory (MRAM)
- SpintronicsSpintronicsSpintronics , also known as magnetoelectronics, is an emerging technology that exploits both the intrinsic spin of the electron and its associated magnetic moment, in addition to its fundamental electronic charge, in solid-state devices.An additional effect occurs when a spin-polarized current is...
- Spin transistorSpin transistorThe magnetically-sensitive transistor , originally proposed in 1990 and currently still being developed, is an improved design on the common transistor invented in the 1940s...
External links
- Redefining the Architecture of Memory
- IBM Moves Closer to New Class of Memory (YouTubeYouTubeYouTube is a video-sharing website, created by three former PayPal employees in February 2005, on which users can upload, view and share videos....
video)