Kara-Moon Forum
August 25, 2019, 02:30:50 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: You can go back to the main site here: Kara-Moon site
 
   Home   Help Search Login Register  
Pages: [1]
  Print  
Author Topic: Cleaning RAM  (Read 659 times)
folderol
Kara-Moon Master
****
Posts: 4796

Who? Me?


WWW
« on: April 01, 2018, 09:39:55 AM »

While most issues with computers slowing down are well known, and pretty much a solved problem. There is one that is only slowly becoming noticeable as systems get progressively faster.

This is the problem of memory slowing down as it ages. Every time a bit is flipped the underlying structures are stressed and an infinitesimal change takes place that makes it slightly more difficult for the next change. We are of course talking incredibly tiny amounts here so it's hardly surprising that it's not really been noticed up to now.

However, it is something that has interested our most {cough} mature {cough} developer, Mary (she of ion trap fame). Once Mary gets interested in something there is simply no stopping her, so we just left her too it, and concentrated on our latest software build problem that only appears on the latest 'felt cowboy hat' distribution.

Our Mary is a really methodical person, and she had this weird setup where she was rapidly switching RAM in various bit patterns and then every few hours, stopping, letting it rest then measuring the switching speed. This was fully under software control of course in an Automated Recursive Sequencing Environment.

Once she had several sticks showing measurable speed drops (takes several months) she put half of them to one side as a control, then tried out every idea she could think of to get the others to 'relax'.

The first was quite obviously thermal - the Applied Heat Neutralising Optimiser - but was inconclusive, and applied long enough it produced dead cells in the RAM.

Then it was mechanical - her Vibrating Axis Guided Uniform Exciter idea. Nope, simply shook things apart.

A sort of combination of both of these was to attack the problem with a Selective Acoustic Wavefront - quite ineffective unfortunately.

Of course, some electronics can be altered with light, so Mary tried using Phased Optical Resonance Notching, although that raised a few eyebrows in the workshop.


With no real breakthrough Mary moved on to trying software methods. Unfortunately I'm not permitted to explain how these work, so you'll have to try to work it out from their names.

The first was Branched Asynchronous Selective Insert Code.

Then there was the Binary Indexed Latency Extractor.

The last of these was Consolidated Ram Accretion Pruner.


Almost a year had passed now and Mary was showing visible signs of desperation (something quite shocking to witness). Almost in panic she tried combining hardware and software in just about every configuration.

This morning she suddenly got up and almost ran out of the workshop without a word. She'd left her last scribbled note out though - it read:

V.A.G.U.E Hardware OK, A.H.N.O  C.R.A.P software tests A.R.S.E.

We just hope she'll calm down by lunchtime.
Logged

If you have a poem, I have a tune, and we exchange these, we can both have a poem, a tune, and a song.
- Will
MarioD
Kara-Moon Master
****
Posts: 3759


« Reply #1 on: April 01, 2018, 03:15:33 PM »

 Grin Grin Grin Grin Grin Grin Grin Grin Grin Grin Grin
Logged

The older I get the better I was!
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.038 seconds with 19 queries.