Computational Phase Change Memory?

Subject: General Tech | November 1, 2017 - 02:03 PM |
Tagged: PCM, IBM

A team of researchers at IBM Zurich have come up with a way to utilize PCM as a simple computational device which does not follow the traditional Von Neumann architecture.  Phase change memory works in a way somewhat analogous to optical storage, with changes to the physical state of the storage medium being used to represent a 1 or 0.  In this case it is a substance that switches from amorphous to crystalline and back again with the application of electrical current; the article at The Register describes this in more detail

This research envisions connecting to a sensor which can send an electrical pulse to PCM to change its state; the example given involves detecting rain and changing the memory to a 1 if rain is detected, a 0 if not. With the application of a algorithm to detect the state of the PCM you can read out rainfall patterns from storage without requiring a processor.  While the computational power of PCM will be quite simple, describing how this works is certainly not so follow the links to the research if your curiosity is piqued.

View Full Size

"But memory has no processor so some aspect of a memory device has to be used, an aspect that changes its nature depending upon the data contents of the memory device. Also the computation is going to be quite primitive"

Here is some more Tech News from around the web:

Tech Talk

 

Source: The Register

November 1, 2017 | 04:52 PM - Posted by MemoryAsSomeComputeThingy (not verified)

The Register article states:

"Each process is assigned to a single phase change memory cell. Whenever the process takes the value 1, a SET pulse is applied to the PCM device. The amplitude or the width of the SET pulse is chosen to be proportional to the instantaneous sum of all processes. By monitoring the conductance of the memory devices, we can determine the correlated group."

The article source that the Register article links to is in the science journal Nature Communications and goes into grater detail and also uses Maths to do so but there are more diagrams in the Nature Communications article/entry which is not paywalled currently.

"Temporal correlation detection using computational phase-change memory"

https://www.nature.com/articles/s41467-017-01481-9#Fig1

November 1, 2017 | 05:01 PM - Posted by MemoryAsSomeComputeThingy (not verified)

P.S. From the Nature Communications article/journal entry:

"In addition to performing logical operations, resistive memory devices, when arranged in a cross-bar configuration, can be used to perform matrix–vector multiplications in an analog manner. This exploits the multi-level storage capability as well as Ohm’s law and Kirchhoff’s law. Hardware accelerators based on this concept are now becoming an important subject of research11,12,13,14,15,16,17." [See link to Nature Communications journal article in post directly above]

How About Phase Change Memory doing shading calculations or Ray Tracing Interaction Calculations in the Phase Change Memory to accelerate that computationally intensive process.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.