• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

very interesting read!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Hard drive design sidesteps traditional paradigm.

LONDON, UK -- (INTERNETWIRE) – 10/28/2002 Dataslide filed a new patent application on the 24th October 2002, in the US and the UK, for a computer hard-drive design.

The new drive combines a mix of current technologies in a fundamental design paradigm shift, aimed at sidestepping a number of key technical barriers in existing hard drive performance development.

Latency and seek times will be reduced to more closely match CPU cycle times, (10 to 50 ns vs. 2 to 5 ms). The initial design will provide data transfer rates greater than current, high performance disk drives, with further architectural potential for massively parallel data transfer rates and fundamental database access re-designs. As a result Wait I/O will be significantly reduced, making use of lost CPU cycles. Power consumption is projected to be less than 10% of current devices.

Initial manufacturing costs are projected to be similar to those of current commercial grade SCSI devices, with ongoing significant cost improvements with economies of scale. It is anticipated that the usual product cost cycle would apply.


The new drive has the potential to replace all hard drives in commercial servers within one to two product cycles.

The typical product cycle transition to second-generation devices for personal computers would be an obvious progression, finally moving to inexpensive third generation devices. In addition the nature of the design offers the potential to open up and acquire new, larger markets currently unavailable to traditional designs, for example, hand held devices requiring zero stored power.

Dataslide director, C F J Barnes, a consultant on capacity planning and performance analysis to a number of major corporations over 20 years, was constantly confronted with data flow bottlenecks and system imbalance. Seeking a method of removing or reducing some of these bottlenecks has been a driving motivation for him.

Dataslide has looked at all current developments and initiated a new construction approach, taking a number of existing technologies, that have been advancing in parallel, and combined them in a way that has not been fully considered before.

The new design will continue to use existing manufacturing equipment and very significantly reduce labour costs in making a step change in hard drive performance, and will not require re-capitalization of the industry.

Dataslide has secured international trademarks, and is currently anticipating that the optimum business model will be licensing and partnership in development.

Further information: www.dataslide.com

It was patented a while ago. God I want this.

A lot.

Like now.
 
I'm quite suspicious about these guys. I see plenty of claims about what the new technology can do, but no details about how they are going to do it. Since the technology is patented (or so they claim), there is no reason not to provide some details.

Edit: Something's not kosher here. They claim to have filed for a US patent on October 24, 2002, but there is no record of a patent application from "dataslide".

Edit 2: The only mention of these guys on the web is various copies of their press release, and their web site. In particular, they appear to have no prior record of any work on data storage systems.
 
Last edited:
I think you are still at risk of someone stealing your design even when patented when you dont have a prototype... or is it a patent pending?
 
If a patent application has been filed, they will be protected as soon as the patent is granted. The protection is retroactive to the date when the patent was filed -- if anyone has used techniques covered by the patent between the application date and the date the patent is granted, it's an easy-win case for patent infringement.

None of this has any bearing on why I can't find a record of a patent application with the US Patent Office.
 
The patent has been filed by an individual on the advice of Tax attorneys and will be assigned to either a UK 'Ltd' or 'PLC' both of which were set up some time ago for that precise purpose.

The technology has been protected in the ROW by a PCT.
 
Dataslide Ltd has been working for a number of years with a group of senior academics and international corporations on a fundamentally new design of storage technology and now has working mechanical proof of concept prototypes.

First generation product will be directed to the hard drive market, with second and third generations being directed to flash memory.

This technology has a number of implications for the storage industry and the general management of data.

Because of this Charles F J Barnes, the Chairman of Dataslide Ltd is contacting the managers of a number of potentially affected industries with a view to opening dialogue with interested parties to manage the impact of such changes in the most effective; and least damaging, way forward.


Characteristics and implications of technology.

The technology uses piezo-electric actuators to give constant micro-oscillations in a single plane, instead of rotation; and an array of standard read and write detectors using LCD fabrication technology, on material with a ‘zero’ coefficient of expansion, with direct addressing to data in 512 byte page segments, on standard magnetic media in full contact, with taC (diamond) surface coatings.

Standard hard drive form factor and protocols will be used to ensure maximum interoperability with current hardware and software standards.

Capacity of the first generation product will be similar to common corporate SAN
areal densities.

Price

The price of the first generation product is expected to be similar to, or possibly competitive with, corporate SAN SCSI type hard drives at a Net of Sales of 3:1, and also to have an attractive price/performance ratio for gamers.

The estimated price for second and third generation product is expected to be much lower.

Speed

The current prototype has the mechanical equivalent of 72,000 rpm, with the potential using currently available materials and technologies, to reach the mechanical equivalent of 12,000,000 rpm.

A fundamental outcome of the architecture, means that any increases in areal density will concomitantly increase data transfer rates by the same factor.


Access

The total surface of the magnetic media is potentially accessed twice during each cycle giving direct addressing of data as well as providing the opportunity for massive parallelism.

The separation of read and write functions means that cache requirements are very considerably less than current hard drives.

Bandwidth

The potential bandwidth of the architecture is of the order of kilobit addressing, physical size is limited only by the development of LCD technology, and massive parallelism could provide opportunities for new supercomputer designs.

Energy requirements

The energy requirements of the product are 3.5% of current hard drives in use and an estimated 0.06% for lifetime cost of ownership.

Leverage of current assets

More CPU cycles, particularly Wait I/O will be available, so that purchase of more CPU power may be delayed for at least one upgrade cycle.

Also because of the reduced energy requirements, any current environmental installation is likely to be more than adequate for a substantial period of time, or may even be reduced.

Database design

Because the data recording surface is accessed in two dimensions, with formal logic mapping and relational calculus structures, the use of fully de-normalised file structures, and data and indices being directly parameterised, the design of historical data can provide performance benefits.

Architectural changes

Hard drive access speeds are the last major bottleneck in common current computer architectures, with up to six orders of magnitude difference between the CPU speeds and the data access response times, and regardless of caching and predictive data management of all kinds, eventually the data has to be obtained from a storage media which operates at millisecond response times, to CPUs operating at nanosecond cycle times.

Because of the change in dimensionality, information structures, surfaces, matrices, manifolds and networks could be directly mapped, and it may be effective to provide SQL primitives within firmware, using closer logical to physical mapping, with the potential for computer architecture overall to move nearer to Alan Turing’s Universal Machine principles.

De-fragmentation

De-fragmentation is not likely to be needed for performance, but has a possible role in data mapping for optimum access.

Management software

Because the product has a faster access time and a directly addressed access methodology and because the technology is potentially applicable through second and third generation product it will be targeted to compete in all the storage technology markets from tapes to hard drives and even flash memory, the requirements for storage hierarchy management software will therefore be considerably different.

Future vision.

The potential for computer systems which have storage media from hard drive to on-board cache from the same architecture and technology provides opportunities for modes with significant simplification in protocols, and therefore performance benefits, without necessarily losing the general operating systems ability to use other technologies.

Because the energy required to access the total media surface is so little, it is theoretically possible for a number of common reference and historic data retrieval functions to be available using little or no stored energy, using piezo power as well as motive force.

This would make it possible to reach a very considerably increased customer base in geographical and economic areas of the world where resources are severely limited, but where particularly aspirational demand for education and access to computing power is very high.
 
Whoa.... holy thread resurrection batman......

EDIT: M33P I hate you ;) Serves me right for having several tabs open and taking time to reply :rolleyes:

EDIT EDIT: At least I can spell :p
 
David said:
Whoa.... holy thread resurrection batman......

EDIT: M33P I hate you ;) Serves me right for having several tabs open and taking time to reply :rolleyes:

EDIT EDIT: At least I can spell :p

haha

Beat by a ninja and stole my ninja at the same time :p

We now return you to your original thread about - uh what exactly? How does this thing work?
 
Back