[ Pobierz całość w formacie PDF ]
.024 MHz.Whenthe IBM PC XT launched in 1981, itwas already about eight times fasterthan the Apollo computer.The next generation ofsmartphone we will be using in thenext two to three years will haveFigure 9.2 The Apollo Guidance1 GHz processor chips.That isComputer, circa 1970roughly one million times faster than(Credit: ComputerHistory.org)the Apollo Guidance Computer.Theoretically, Moore s Law will run out of steam somewhere in the nottoo distant future.There are a number of possible reasons for this.Firstly,the ability of a microprocessor silicon-etched  track or circuit to carry anelectrical charge has a theoretical limit.At some point when these circuitsget physically too small and can no longer carry a charge or the electricalcharge  bleeds , then we ll have a design limitation problem.Secondly, as successive generations of chip technology are developed,manufacturing costs increase.In fact recently, Gordon Moore said that eachnew generation of chips requires a doubling in cost of the manufacturingfacility as tolerances become tighter.At some point, it will theoreticallybecome too costly to develop the manufacturing plants that producethese chips.Lastly, the power requirements of chips are also increasing.More power= more heat = bigger batteries, etc.At some point, it becomes increasinglydifficult to power these chips while putting them on smaller platforms.Eventoday, some laptops are literally getting too hot to put on your  lap.Deep Impact  Technology and Disruptive Innovation289So when will this all happen? Gordon Moore himself predicted theend of Moore s Law around 2022.Other predictions range from 2015 to2060.But industry analysts have  called the end of Moore s Law morethan a few times before.Additional transistor densities may be achievedin other ways too, such as via 3D semiconductor structures, or moreexotic approaches such as carbon nanotubes, silicon nanowires, molecularcrossbars and spintronics.Even if Moore s Law does end, thisdoes not mean that technology then The usable limit forstagnates and hits some permanent semiconductor processperformance ceiling.If transistor size technology will be reachedbecomes a constant, then the burden when chip processof computer development will simply geometries shrink to beshift to other parts of the classical smaller than 20 nanometerscomputing ecosystem.There are (nm) to 18nm nodes.Atplenty of opportunities for improving those nodes, the industrycomputing performance in other areas, will start getting to thesuch as design of the device or system point where semiconductoritself, software and firmware efficiency, manufacturing tools arenew board and network architectures too expensive to depreciatethat do more work with less silicon.with volume production,The fact is that systems such as i.e., their costs will be soWindows and much of the software we high that the value of theirhave today could be far more efficient, lifetime productivity canbut they remain bloated with legacy never justify it.Len Jelinek,code that has been kept over successive Chief Analyst, iSuppli, June 2009generations of development.How this translates is that forthe next 10 15 years (a very long time in business life-cycles), if you arein banking, you will remain under constant pressure to integrate better,faster technologies and techniques to stay relevant to customers and to staycompetitive.If you aren t constantly updating your technologies, platformand multichannel capabilities within the bank, you WILL fall behind.Thisis just a fact of life get used to it.290 BANK 2.0After Moore s LawWhen we look further into the future, there are really only two promisingsolutions that will replace the silicon paradigm that underlies the flawlessperformance of Moore s Law to date.Those two solutions are quantumcomputing and DNA (or biological) computing.Quantum computing essentially utilises the quantum state of qubit(the equivalent of a normal bit/bite in computing terms, but at the quantumlevel).Like a traditional bit, a qubit has an on and off state, but whereasa bit can only be 1 or 0, a qubit can also produce a superposition of bothstates.Thus, depending on configurations, implementation, the principlesof entanglement and superposition (quantum mechanical phenomena),a quantum computer will likely operate on an underlying bit structurethat contains at least eight different three-bit strings.But because of thenature of quantum mechanics, it can simulate the calculations of almostany combination of results simultaneously.This means a completely different type of programming wouldbe required, but it results in massive computing power.Programmes,calculations or simulations that would take weeks, months or even yearsto complete on today s platforms could be executed in real time almostinstantly.Chips the size of a grain of rice would be more powerful thantoday s supercomputers, and use almost no power at all.The other promising replacement for silicon technology is DNAcomputing which uses DNA, biochemistry and molecular biology.Itwas first demonstrated as a concept by Leonard Adleman of the SoCal(University of Southern California) in 1994.Adleman demonstrateda proof-of-concept use of DNA as a form of computation which solvedthe seven-point Hamiltonian path problem.He used an oligonucleotide,which is just a really fancy name for a polymer.But if you ve ever watchedan episode of CSI where they take a piece of evidence with a suspect sDNA and put it in a solution to identify to whom it belongs, then youare watching one typical use of oligonucleotides, as they are often used toamplify DNA in what is called a polymerase chain reaction.Ok, ok, enough of the technobabble & well almost.Deep Impact  Technology and Disruptive Innovation291Materials ScienceThe A380 Airbus and the Boeing 787 aircraft are unique as commercialaircraft go because 20 years ago we simply could not have built these aircraftwith the materials available at the time.We re talking in the 90s notexactly the Dark Ages [ Pobierz całość w formacie PDF ]