Redundant Array of Independent Paperweights

01/11/01: For immediate release
Keywords: Redundant Paperweight, Gravity Failure

Electromagnetic Networks has always kept you ahead of the technology game by providing you with innovative products years ahead of all other manufacturers. In keeping with this line of advanced products, we present the Redundant Array of Independent Paperweights (RAIP).

Electromagnetic Networks provides an open, standards based RAIP system to protect your most valuable hard documents against localized gravity failures. This protection is achieved using a system of redundant paperweights to maintain gravitational force should as many as one less than the number of paperweight units (PUs) installed experience localized gravity failures.

To ensure that your paperweight system stays at peak performance, Electromagnetic Networks has taken our breakthrough system to the next level:

Indicator lights on the front of the unit ensure constant monitoring of the RAIP's operation with heads-up alert lights in the case of localized gravity failure(s).
Easy-to-replace hot-swappable paperweight units, conforming to industry standard form-factors, allow for rapid PU replacement without locking you in to one vendor.
Stylish looks and ergonomic design keep you from having to sacrifice appearance for functionality.

Our engineers have rigorously tested the design. Electromagnetic Networks is proud to share our test specifications, so that you may independently verify our claims.

Scalability Tests
1. Our engineers first tested the ability of a Paperweight Array to scale in protection from three (3) PUs (the standard configuration) all the way up to six (6) PUs.¹ Our test environment was a stack of twenty (20) sheets of A4 size letterhead, each labeled with an identifier to track stacking order and simulate data. As we added PUs to the system, the safety of the data remained uncompromised. Six (6) PUs was found to be the upper limit of scalability in this environment, as additional Paperweight Arrays did not fit on the stack of paper.
2. Next, we tested the ability of a three (3) PU system to protect up to a ream of A4 letterhead. Starting with a stack of five (5) sheets, we added letterhead five (5) pages at a time. We ran out of letterhead at one (1) ream, and ended the test.
Failover Tests
1. First, we tested the robustness of a three (3) PU system in a localized gravity failure situation.² We further increased the severity of the test by passing a draft from a standard desk fan over the protected stack of letterhead. Again, the letterhead was individually identified to guarantee detection of any data misordering. We first tested failure on each of the PUs individually, and found the integrity of the protected data was unimpacted. Next, we tested simultaneous failure on two (2) units at once, again cycling through the array. Again, data integrity was maintained. Finally, we tested simultaneous failure of 100% of the PUs, causing a massive data integrity failure, as expected.
2. Finally, we tested reliability of our PUs as they increased in array capacity. Although the three (3) PU model is only effective at rates of up to 66.6% failure, each additional PU in the system dramatically increased the number of Paperweights that could fail without affecting stability.

Note 1: Since we do not offer a five (5) PU array, we used a three (3) PU array with a PU removed, effectively simulating the proposed five (5) PU RAIP Array.
Note 2: Since no localized gravity failures were predicted for the test environment, we simulated localized gravity failure by removing PUs one at a time from the Array Storage Unit.

Failure Statistics for Each Model of RAIP Array

Number of PUs	Failure Tolerance
Three (3)	66.6%
Four (4)	75.0%
Six (6)	83.3%
Eight (8)	87.5%

As you can see, the protection of the array increases with the number of PUs, but efficacy rolls off exponentially. Although the eight (8) PU system is sufficiently superior to a four (4) or six (6) PU system, additional PUs do not provide a cost-effective solution. For circumstances requiring additional arrays, Electromagnetic Networks suggests the deployment of redundant arrays of RAIP arrays.

RAIP Arrays are available in both striped and non-striped formats. Please specify the desired type when ordering.

Comments on this Document
Add a comment

Redundant Paperweights
by CYBERHONKY v9.25 () on 03/23/2001 at 02:40 (#1)
Are they susceptable to such gravitational anomolies as glowing cow turds? GCT happens when the black angus variety happen upon the many nuclear waste fascilities in the mid-west and grazes upon the fields in close proximity. I am rather curious to know.

Sorry...
by Ham Monger () on 03/26/2001 at 12:04 (#2)
We did not design this product to that specification and thus have no data on it at this time. We will look into acquiring some GCT for testing.

WinME compatability
by Hawklord (hawklord2112@aol.com) on 09/04/2001 at 16:07 (#3)
hi guys i recently upgraded to winME and i'm having difficulty with my RAIP network - is there a patch available?

In development...
by Ham Monger () on 09/04/2001 at 17:58 (#4)
As it stands, there's no way to hook it up to a computer anyway, so naturally it won't work under WinME. However, our engineers are working on such an interface. Yet, you will still not be able to use it under WinME as we will only support Linux x86 and SPARC Solaris.

technology spin-off
by KGB () on 09/07/2001 at 20:47 (#5)
The other day while trying to work with a deffective monitor, I discovered and interesting effect: by piling up a senmail book (the bat book) and an old UPS on top of the monitor, it would stop flickering. However, I had to regularly re-stack them to prevent them from falling on my hands on the keyboard. I wonder if a scaled-up model of the RAIP array would do the trick in a better way?

another method
by xcpu () on 09/08/2001 at 13:29 (#6)
you can add elecrostatic holders incase of TOTAL gravity failure
also add Elicromagnetic PU's in case of Columb failure.

Re: technology spin-off
by Pasta Man () on 10/25/2002 at 02:32 (#7)
Dear KGB,
We have investigated the feasiblity of your proposed product
suggestion, and have found that such a development could
indeed benefit both the Laboratories and the general public.
Once we have successfully run this by what remains of Legal, we will prioritize development of this revision of the product. Good luck with your monitor!