Is The Future Of Storage QED?

[note: this was my April Fool's post-- but since it's not April 1 anymore ...]

These days, there's a lot of new and radical enabling technology to discuss in the storage domain. 

Automatic tiering technologies like FAST.  New approaches such as global federation that's built on distributed cache coherence.

And -- out there on the horizon -- is a technology that has the potential to completely redefine how we think about information storage.

It's QED

What Is QED?

QED stands for Quantum Entangled Data. 

To fully understand the potential of QED, we need to enter the bizarre world of quantum physics.

Much has been made of the potential for quantum physics to change the very nature of computing through quantum computing ; but a few are tracking the potential of a similar revolutionary effect in the storage domain.

Quantum entanglement refers to the unusual property of synchronizing the quantum state between two particles using a burst of energy.

Students of quantum theory know that two particles in an entangled state can record state: perhaps a "1" for being entangled, a "0" for unentangled.

Particles can now be entangled with multiple partners ; recent work indicates that there appears to be no theoretical limit of the number of entanglement pairs that can be formed with a given particle.

This gives rise to the potential of mind-boggling information densities as a result. 

A pair of particles can record one bit.  Three particles can record three bits.  Four particles can record 6 bits.  Five particles can record 10 bits, six can record 14 and so on.

This give rise to a near-exponential in storage capacity as the number of participating particles increase.   A few billion particles have the theoretical capability of recording all the information on the planet, for example -- and potentially occupy the space smaller than a grain of sand.

Quantum entanglement states take no energy to preserve.  Some energy is involved in encoding (entanglement) and reading state, but the energy requirements per petabyte are apparently many orders of magnitude below current technological approaches.

Wait, There's More ...

As people familiar with the topic will know, the relationship between two particles is completely independent of distance, and not subject to the typical speed-of-light constraints. 

It takes a while to accept that entanglement properties are complete distance and latency insensitive, but the potential is there -- the ability to read and write data across distances with no measurable latency.

And no network whatsoever.

What Does All This Mean?

First, imagine incredibly cheap and dense storage where capacity, space, power, cooling, etc. is simple a non-issue.  Indeed, the read-write mechanisms will probably take more resources than the storage medium itself.

Now imagine that these storage nodes supporting this model could be arbitrarily separated with no distance and latency constraints -- and without using any sort of network.

As many nodes as you'd like -- at any distance you'd care to imagine.

To say that this technology could completely redefine our understanding of information storage is a bit of an understatement -- we'd perceive storage as more of a universal fabric -- any node would be a seamless and instantaneous part of the overall storage complex.

Well, words fail me.

The Disruptive Potential

Clearly, a number of industries could be potentially disrupted by the promise of infinitely cheap and infinitely connected storage.  Certainly, any manufacturer of storage media is subject to an exponential crash in the cost of media. 

And, once you think about it, network technology providers are subject to the same sort of disruption.  The simple act of writing information to any storage element would make it instantaneously appear on all related nodes -- no bandwidth required!

And that's just for starters ...

How Real Is This?

Hard to say. 

Most of the fundamental work is being done by Dr. Lipra Sloof, a rather reclusive individual who tends to publish infrequently, but periodically offers an update to the storage research community on an annual basis in the springtime.

For those of us in the technology community, we look forward to these unique glimpses into the future research that has the potential to restate and redefine the art of the possible -- or the impossible -- as the case may be.

Happy April 1st, everyone ...

Comments

[Rob]

Hi Chuck. great post. Have you ever seen "Who Killed the Electric Car"?
A synopsis:
In 1990, the state of California passed the Zero Emissions Vehicle Mandate which instructed the automakers that if they cannot sell an automobile in California if it has a tailpipe.
GM responded in 1996 with the EV1. It was an electric car that could travel between 70 to 100 miles on a single charge. It was available as a lease only and GM put about 700 on the streets.
Problem: The oil companies woke up, wiped the drool off their chins and said “Hey! That thing doesn’t use my product!” They called George senior and complained.

The parts manufacturers complained because an electric motor uses and fraction on the parts that a combustion engine uses.

The Auto Dealers complained because an electric engine requires a simple greasing and tire rotation every few months. No Oils, no filters and really doesn’t break.

This paradigm shift in the auto industry and its impact to its tributary industries was caused such a disruption in the BAU cash flow that California rescinded its ZEV mandate and GM withdrew the product citing poor acceptance. The funny part about that is when EV1 owners pleaded to keep their cars and buy out the leases, GM declined, took the cars back and crushed them all.

Whether or not Quantum Computing and Storage becomes available to the consumer will have little to do with the benefits to the Joe on the street. It will be decided by “How much money gets diverted from whom”.
1. No requirement for bandwidth? Bye bye Brocade and Cisco et al.
2. It’s a blob of atoms. MTBF 6.02X10E23 years. There goes about 50% of the computer industry’s revenue.
3. No parts (depending on how we read and write to the thing).
4. All the information on the planet contained on a grain of sand accessible from anywhere? Tells me you need one. OK, two for DR. So unless accessing the data requires a Hadron collider, there may not be a lot of money in this.

Quantum computing is a game changer and when the game changes, winners can become losers very quick.

I know I’m being simplistic but you do see my point, I hope. Then again, I could be wrong.

"Everything that can be invented has been invented."
Charles H. Duell, Commissioner, U.S. patent office, 1899

"Computers in the future may weigh no more than 1.5 tons."
Thomas Watson, the chairman of IBM in 1943

"Who the heck needs a 10GB hard drive in their home?"
Me - 1989

[excITingIP.com]

Too good... All the Quantum physics "research" engineers should read this!! But perhaps, just perhaps, something similar might become true as well... :)

excITingIP.com

PS: The link on the April Fools day page - wikipedia was too funny...

[Stuart Savill]

very good - this made me chuckle! 8-)

[Haiyun Bao]

"Five particles can record 10 bits, six can record 14 and so on."

Six can record 15, right?