The Missing Memristor

11 May, 2010 § 3 Comments

Hewlett Packard recently announced that they have fabricated memristor chips and plan to bring them to market by 2013 [8]. This paper is an evaluation of the memristor and focuses on the potential impact of its introduction.


The memristor is an electrical circuit element that is similar to a resistor but has the potential to maintain state between turning power on and off. The memristor was first described by Leon Chua in 1971 as the fourth fundamental element, based on a relationship between charge and flux [1]. These memristors are about half the size of the transistors found in current flash storage technology such as iPods and portable USB drives, allowing capacity of these devices to double. Further, Hewlett Packard has claimed the ability to improve the data-write cycle limit 10-fold, from flash technology’s 100,000 to the memristor’s 1,000,000 data-write cycles [3].

A Computer Architecture Perspective

The memristor hopes to replace the memory hierarchy that is seen in almost all computers today. Because the memristor’s state is nonvolatile and is planned to reach about four nanometers in width, large amounts of memory can be stored directly on the processing chip. The memory hierarchy seen in today’s computers may have three or four levels of volatile cache, another level of volatile DRAM, and finally a level of nonvolatile memory. Putting the nonvolatile memory close to the logical unit on the chip has the potential to allow memory accesses of 1,000 times faster [1].

Memristor caches may be able to remember state and resume said state even as the power to the devices comes and goes. During current computer start up routines, the bootloader needs to populate the translation lookaside buffer (TLB) at each startup. With memristor technology, the TLB can remember state between power cycles, allowing a much quicker start up process. As the technology matures, this same thinking could be extended all the way to resuming running processes when the machine starts back up.

At its initial implementation, Hewlett Packard’s goal is place memristors between DRAM and disk technology, eventually spreading in both directions to replace disk and DRAM [4]. Due to the memristors speed and data integrity guarantees, this goal is likely to bring hard drive-like reliability with speeds faster than DRAM. Consumers are likely to see these speed increases when writing to their iPods and USB drives, since the higher write-cycle limits will allow the same sector to be written to instead of current flash implementations that try to write to the same sector less often.

Another use for memristors is for computing logic. Using material implication, a group of three memristors can be used to compute any boolean logic equation that is requested [2]. The motivation of using memristors for logic computation comes from the ability to run the computations on the nanoscale. The use of material implication complicates the boolean logic but is not enough to deter the gains in size reduction. The gains are enough to have one memristor replace 15 transistors [9].

Combining the use of memristors for storage and logic may also present a new application for the device. Hypothetically, a crossbar of memristors will have the ability to re-purpose groups of memristors for data-intensive or compute-intensive operations. Memristors are also very efficient power users [1]. Embedding memristors within sensor networks has the possibility to allow sensor networks to collect more data and have much longer lifetimes.

The power walls and memory walls can effectively be written off if the memristor proves its claims.

Impact for the General User

As previously mentioned, the memristor technology will allow much faster write-times compared to flash technology. General computer users will be able to sync up their portable devices in 1/1000th of the time that it takes them today. Not only will devices be able to sync faster, but they will be able to store more data than could have ever been concieved of before. HP expects to reach a storage density of about 20 gigabytes per square centimeter by 2013. As a comparison, the surface area of an Apple iPod Nano is about 64 cm2 which could translate to over 1.25 terabytes of data on the surface of the device alone.

General users can expect to see devices that use this extra storage space to continuously collect data. The vast amounts of data that can be collected are in line with Intel’s “Era of Tera” forecast [5]. Devices will be able to use the data for facial recognition of past acquaintances, weather forecasting through crowd-sourced data collection, and even more interesting applications that have yet to be thought of.

Critiques of the Memristor

To implement the memristor, titanium dioxide is used as the semiconductor instead of the traditional silicone. There is still much to be understood about titanium dioxide, as a team from the National Institute of Standards and Technology (NIST) said [6]. Silicone has had many years to mature as a semiconductor element in use with electrical circuits, whereas researchers are just beginning to understand how to use titanium dioxide in electrical circuits. “The fundamentals of why these metal oxides switch the way they do are not well understood,” said NIST researcher Curt Richter [3]. Further time and research may resolve the unknowns, but at this point there might not be enough known information about it to be sure of its promise.

Hewlett Packard and Intel have had a long history of working together to introduce new technology, such as their work with the EPIC Itanium architecture. Intel has been approached to work on the memristor technology with Hewlett Packard but has decided not to take Hewlett Packard up on their offer. Instead, Intel is looking to focus their energy on phase-change memory [4]. Intel has already shipped phase-change memory samples in 2008 and plans to start shipping mass quantities in 2010 [7], as opposed to Hewlett Packard’s expected release date of 2013 for the memristor.

In Conclusion

The memristor presents an amazing opportunity for change in the computer memory hierarchy, storage capacity, and nonvolatile state. Lab results for the research have shown that we are just at the cusp of all the capabilities of memristors. With that being said, there are still some unknowns about the technology that will challenge its adoption. The material used as the semiconductor is relatively new to electronic circuits, and research from competitors is appearing to be quicker to market than the memristor.

Much of the publications for the memristor are from it’s main researcher, R. Stanley Williams of Hewlett Packard. Many of the claims have not been validated by a third-party and could very easily be exaggerated. If given the opportunity to invest in memristor technology, I would advise taking a deeper look in to phase-change memory and it’s other competitors that look to be closer to reaching the market. Computing technology changes very quickly and there is a low probability that the computing environment three years from now is unchanged from today.


[1] Adee, S. The Mysterious Memristor. IEEE Spectrum. May 2008.

[2] Borghetti et al. ‘Memristive’ switches enable ‘stateful’ logic operations via material implication. Nature. Vol 464. Apr 8, 2010.

[3] Bourzac, K. Memristor Memory Readied for Production. Technology Review. MIT Press. Apr 2010.

[4] Foremski, T. Tha amazing memristor – beyond Moore’s law and beyond digital computing. ZDNet: IMHO. Apr 19, 2010.

[5] Garver, S. and Crepps, B. The New Era of Tera-scale Computing. Intel Software Network. Jan 15, 2009.

[6] Jones, W. A New Twist on Memristance. IEEE Spectrum. Jun 2009.

[7] Miller, M. Memristors: A Flash Competitor that Works Like Brain Synapses. PCMag: Forward Thinking. Apr 14, 2010.

[8] Null, C. Memristor technology gets real; commercial release planned for 2013. Yahoo! News: Today in Tech: The Working Guy. Apr 2010.

[9] Williams, R. S. How We Found the Missing Memristor. IEEE Spectrum. Dec 2008.

Tagged: , , ,

§ 3 Responses to The Missing Memristor

  • A.J. Orians says:

    I like the faster write-times compared to flash technology. I’ll be looking forward to this!

  • Brendan Grebur says:

    Brendan Grebur
    CSE 820
    Final Paper
    Evalution of Memristors

    Over thirty years ago, a professor at UC Berkeley uncovered a theoretical missing element in the area of circuits[1]. Such a triumphant discovery was met with little praise, as any ability to physically build such an element remained unseen. For a great time, the memristor lived as a series of convoluted mathematical formulas on paper. In 2006, researchers at HP devised the first physical models for constructing such a device. A paper[2] in April 2008 officially announced HP’s find. Since this time, memory designs have been proposed[3], along with FPGA implementations[4], but a revolution in computing has yet to occur. The memristor may be a monumental discovery in the field of electrical engineering, but the new circuit element comes with its own set of hurdles that will determine its impact on computer architecture. In light of currently available information, I believe memristor technology will have minimal impact on current architecture.

    The concept of a memory resistor is quite simple—resistance is determined by the applied voltage and period of time applied. This resistance persists until new voltage is applied. As alluded to by the name, such a device could easily store data in the non-volatile form of resistance. Since resistance is variable, data need not be in binary form. Analog data can accurately be stored providing an advantageous platform for signal analysis[6]. However, one of the most hopeful areas of application are in neural networks. Memristors inherently display the characteristics of neurons that communicate with one another in the brain.

    New computing features are now available through memristors. Non-volatile memory at one tenth the size of current DRAM technology[7] certainly sound agreeable to a computer architect. Logic gates have even been formed using sets of memristors[8].

    Amidst the few advantageous characteristics lies a sea of new problems to properly utilize the devices. Memristors used in memory experience the same nuances of today’s DRAM modules—reading destroys the data. Applying current across the memristor to determine its resistance inherently changes the stored value. A simple solution of applying a reverse voltage of the read out was proposed by Chua[3]. Of course, the solution has now introduced a delay between consecutive reads, but more importantly, extra energy is required to restore the read value.

    HP argues that converting current DRAM with memristors would save energy as systems would never need to boot up from disk[9]. The energy saved from boot, especially in highly available machines, typically gets lost in the noise of normal operation. Even for desktop users who boot machines a few times a day, the energy for 30 seconds of boot operations hardly seems worth mentioning, especially since most computers already have power saving modes. Matching this with the fact that operating systems constantly optimize their boot time, such a switch holds little advantage. A more interesting comparison would be in the way of operating energy. However, logically memristors would be more energy efficient since they are not bound by electrical leakage like DRAM capacitors. Although, more hardware is needed to actually read and write to memristors.

    One of the greatest attractions to memristors are their nanoscale size, on the order of ten square nanometers[7]. With DRAM modules topping out at 40 nanometer features, FLASH manufacturers are pushing the bounds and predicting a 20 nanometer future[11]. Even in light of this speculation, memristors hold a significant advantage. Unfortunately, there is a price for such a scale—titanium dioxide memristors are ten times slower than competing DRAMs[10]. Such a performance hit during an era when memory latency limits processor performance is unacceptable. Therefore, DRAM is not likely to be replaced by memristors anytime soon. Consequently, such a limitation precludes memristors in areas that could benefit tremendously from nanoscale size, such as on-board chip caches.

    Logic gates using memristors have already been proposed[8], which leads to the question of whether memristors could replace the current MOSFET technology. In spite of their restrictive speed, packing memristors on processors would have its advantages. Current memristors have a feature size of roughly 3 x 3 nanometers, in comparison to the minimum transistor size achieved in the laboratory at 20 nanometers. Replacement would allow distance between logic gates to be reduced and effectively speed up communications. Such a move would open up processor space for more on-board caches, but adding more cache could remove any energy advantages of the memristors. It is also unclear of whether compacting these new circuit elements will play a significant role in heat generation, a leading concern among transistors today. The ultimate hurdle for realization in this area is the lifetime of memristors. Since the technology[7] relies on the physical movement of ionized oxygen bubbles, sustained use could result in loss of these bubbles, similar to the wear on copper wires. When operating on the level of gigahertz, any sensitivity to this issue becomes quickly magnified.

    One of the major hurdles that memristor face is the market, specifically the dominating chip manufacturers. Intel, who has entered the FLASH drive market sometime ago, has made no announcement for partnering up to release memristor based drives, nor has any other heavyweight. Even with memristors capable of being produced on current semiconductor fabrication stations[7], such a delayed introduction to computing presents an almost insurmountable wall for the technology. Transistors are well established and represent an investment most chip manufacturers are unwilling to forsake. However, this does not preclude the memristor presence into other computing fields.

    Memristors naturally mimic the actions of nerve cells interacting. Individual neurons communicate with one another across synapses, using chemical or electrical signals firing at different rates. Based on the firing rate, a threshold may be passed, and a signal passed on. A firing rate easily translates into an applied current to a memristor, which over time would increase voltage and “fire” off a transistor. Perhaps mimicking a low order number of neurons might this approach be feasible, but some neurons can have thousands of connections. Allowing transistors to dynamically establish remote connections appears a major hurdle. Another impedance occurs after a threshold has been reached, the connection needs to be reset. In a memristor, the resistance would remain where it left off, increasing until a maximum resistance was reached. Such a problems require additional hardware to effect the reset, which severely adds to the overall space. Again, many hurdles face the realization of memristors in various fields.

    The fundamental problem facing current computer architecture remains the speed of light and instruction level parallelism. Applications cannot run faster, thus imploring a parallel solution. Introducing a new circuit element will not resolve the issue, at least not apparently. Memristors will inevitably make their way into both memory and storage devices, but will not change the current computer architecture. Some applications may arise in crossbars when nanoscale production of transistors have unacceptable yields, along with other research based areas. Memristors fail to provide the necessary features to effectively impact computer architecture.


    [1] Chua, L. O., “Memristor-the missing circuit element,” IEEE Transactions on Circuit Theory, vol. 18, no. 5, pp. 507-519, 1971

    [2] Strukov, D.B., Snider, G. S., Stewart, D. R. and Williams, R. S. “The missing memristor found,” Nature 453, pp. 80-83, 2008

    [3] Hyongsuk Kim; Sah, M.P.; Changju Yang; Chua, L.O.; , “Memristor-based multilevel memory,” Cellular Nanoscale Networks and Their Applications (CNNA), 2010 12th International Workshop on , vol., no., pp.1-6, 3-5 Feb. 2010

    [4] Robinett, W., Snider, G. S., Kuekes, P. J., and Williams, R. S. 2007. “Computing with a trillion crummy components”. Commun. ACM 50, 9 (Sep. 2007), 35-39.

    [5] “HP Labs Discovery Holds Potential to Fundamentally Change Computer System Design,” April 8, 2010. [Online]. Available: [Accessed: May 02, 2010].

    [6] Stork, M.; Hrusak, J.; Mayer, D.; , “Memristor based feedback systems,” Applied Electronics, 2009. AE 2009 , vol., no., pp.237- 240, 9-10 Sept. 2009

    [7] Williams, R.; , “How We Found The Missing Memristor,” Spectrum, IEEE , vol.45, no.12, pp.28-35, Dec. 2008

    [8] “Memristive’ switches enable ‘stateful’ logic operations via material implication,” Julien Borghetti, Gregory S. Snider, Philip J. Kuekes, J. Joshua Yang, Duncan R. Stewart & R. Stanley Williams, Nature 464, 873-876(8 April 2010)

    [9] “HP Labs Proves Existence of New Basic Element for Electronic Circuits,” April 30, 2008. [Online]. Available: [Accessed: May 02, 2010].

    [10] Markoff, John, “H.P. Reports Big Advance in Memory Chip Design,” NY Times, May 01, 2008. [Online]. Available: [Accessed: May 02, 2010].

    [11] Gruener, Wolfgang. “Samsung announces 40 nm Flash, predicts 20 nm devices”. TG Daily. September 11, 2006. [Online]. Available: [Accessed: May 02, 2010]

  • […] The Missing Memristor May 2010 2 comments 5 […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

What’s this?

You are currently reading The Missing Memristor at JAWS.


%d bloggers like this: