Slashdot Mirror

That's kind of amazing seeing as the cost of Iter alone is 20 billion +
https://www.sciencemag.org/new...

and the NIF was nearly 4 Billion
https://lasers.llnl.gov/about/...

Pretty sure there's been more than 2 projects.

I have no reason to doubt the author's research on the specific regions that produce barley.

They aren't using research on specific regions, the paper says they are using CMIP5 for future predictions, which is very non-specific (by the way that link is the closest I could find for what they were trying to link to, since the link in the paper itself is actually broken).

They may know a lot about the regions producing barley but as far as I could tell didn't have a very strong explanation for predictions they were giving as to changes in that region.

Am I the only one who tried to read the paper? It doesn't seem like a greta paper and the links they use for further evidence are not at all specific.

Yes, and if it's about parallelism, then at least the /. summary is wrong. Sure "single program performance only grew 3% last year", if you correct "program" to "thread", but we get more cores every year now. And multi-core is so commonplace, even smartphones routinely have them.

I still suspect that we cannot fully make up with parallelism, what has been eroding away from Moore's Law. At least not in every case. However it's not crazy to be thinking about bog standard PCs with dozens or hundreds of cores in the foreseeable future.

There is no way that software engineering has fully mined parallelism, not even close. I'll bet there are hundreds of top-ranked software programs, used by millions, that have no parallel support. None at all! And simplistic parallel designs put artificial limits on how many threads they will support, yet I'll bet that the real world of extant parallel software is dominated by simplistic parallel designs.

Lots of people here on /. are quick to quote Amdahl's Law as a barrier to parallel design opportunities. Yes it's a constraint, but that's far down the road as a major concern. I'll bet optimized supercomputer programs run into Amdahl's law fairly regularly. Your browser though? Your spreadsheet? Your database? I very much doubt that.

https://hpc.llnl.gov/tutorials/introduction-parallel-computing/limits-and-costs-parallel-programming

We overproduce electricity by as much as 66% if you look at an energy flow diagram https://flowcharts.llnl.gov/co... you see that 25 quads were rejected out of 38 quads of electricity generated. Yet transmission loss is on the order of 5%. The rest of the rejected energy is overproduced.

In Washington they turn off windfarms and Microsoft wastes energy to meet its wholesale quota. We should be paid to use electricity.

Uh, no. You don't understand the chart you linked to.
Rejected energy is NOT overproduction of electricity generated.
Rejected energy, in the chart you referred to, is mostly energy lost due to thermodynamic cycle efficiencies and heat loss due to design limitations. Think Carnot cycle, Rankine cycle, etc.
This has nothing to do with the periods of overproduction in wind farms (or solar).

We overproduce electricity by as much as 66% if you look at an energy flow diagram https://flowcharts.llnl.gov/co... you see that 25 quads were rejected out of 38 quads of electricity generated. Yet transmission loss is on the order of 5%. The rest of the rejected energy is overproduced.

In Washington they turn off windfarms and Microsoft wastes energy to meet its wholesale quota. We should be paid to use electricity.

The situation is much more complicated than that:

https://www.llnl.gov/news/atmo...

Even if salinity were simply decreasing, such a geographically concentrated correction would still be devastating. If you were able to cool a small area of the earth to near absolute zero, it would also be locally devastating.

Scott Adams has some interesting things to say [dilbert.com] about the subject.

No, what he's saying is absolutely stupid. Climate models are based on physics models, not just a bunch of made-up offsets.

Here's a well known model forecast, with real temperatures plotted inside it: https://pbs.twimg.com/media/DK...

Here you can find all the details about these models:
https://cmip.llnl.gov/index.ht...

Yes this is a small milestone. The rate-of-growth for this realm is on an exponential curve (see: https://www.forbes.com/sites/r... ). Perhaps the 10% that would impress you for Kansas will take some time, but globally generation is increasing quickly. The EIA projects the next 2 years to be steady at 10% (see: https://www.eia.gov/outlooks/s... )

Percentage of load is a different stat entirely. There, it may be better to look at the conversion channels of energy (fuel vs electricity) via something like https://flowcharts.llnl.gov/ which show large opportunities for renewables. Load is increasing each year, so this skews all percentage claims.

That wood flooring was 3D printed?!

It could have been, since 3D printing wood is a thing. Or it might not be wood at all.

How did they 3D extrude the wiring and meet code?

Maybe like this? I don't know what "code" Russia has, but a machine could be much more precise than a human.

Concrete is a good insulator for russian winters, right? Amazing! How good was the R-value? How was the rebar extruded?

Maybe they 3D printed some foam insulation? And why would they use rebar in a one storey construction that small?

By models:

At the center of the Sun, fusion power is estimated by models to be about 276.5 watts/m3.[1] Despite its intense temperature, the peak power generating density of the core overall is similar to an active compost heap, and is lower than the power density produced by the metabolism of an adult human. The Sun is much hotter than a compost heap due to the Sun's enormous volume.[2]

Ref [1] here.
Ref [2] this page, which is not very good; and a simple model here.

All the rocket fuel used annually is absolutely dwarfed by, e.g., automotive fuel.

Check out this flowchart of where roughly 97.4% of US energy comes from/goes to. Note the massive waste (rejected energy) in transportation AND electric generation. However a move to electric vehicles over the next few decades will still reduce wastage because of lower amounts of energy consumed overall. Plus the wastage in energy generation and transmission can also be greatly reduced.

But we have to admit this is a problem. For example shifting from coal to natural gas, which is happening because gas is a lot cheaper, could save us a lot of energy wastage and carbon emissions -- if we're careful to regulate methane emissions from natural gas production and distribution. Methane, the main component of natural gas, is a very potent greenhouse gas. Thus scrapping environmental regulations would actually hurt coal production because the gas industry would be able to externalize its costs and sell at an artificially low price.

There's every reason to believe we can make a big dent in the roughly 60% of energy that is simply wasted, and if on top of that we develop more carbon neutral sources like nuclear, wind, and biofuels we can over the course of a twenty years or so really reduce our carbon footprint while improving our energy independence. We're almost energy independent now, and if things continue on the path we've been on for the past six years or so we'll be a net exporter of energy in the next two or three years.

Conservation and clean energy will allow us to grow our economy, become independent of foreign oil supplies from unstable regions, and create jobs. But it will take changing the status quo, which is why people who benefit from the status quo don't want us to acknowledge the problem.

As for rocketry, it may not be a big deal in aggregate, but single orbital launch still puts out a lot of CO2 -- about the same as an average car driven for almost 50 years. However I think that could be reduced too, by introducing more biofuels as well as developing alternative launch technologies.

That's right, melting ice caps can alter ocean salinity, which could alter ocean currents, leading to further climate changes:

https://www.llnl.gov/news/atmo...

Natural climate changes could affect salinity in the same way. It doesn't pose much of a risk to marine life compared to other effects of global warming.

When I talk about models I'm mostly talking about the Couple Model Intercomparison Project (CMIP). Different models have different strengths and weaknesses and I'm not sure it's possible to pick any one model that is the best.

One thing that is possible is to cherry pick individual model runs from an ensemble of runs that happen by coincidence to match real world natural variability (especially of ENSO). when you do that the model output matches the observations quite well.

Here is an interesting article "Comparing CMIP5 & observations". It shows that observations are within the the ranges of the CMIP5 output although mostly at the lower end of the range but 2015 has pushed observations into the middle of the range.

One thing that has come up lately that may be an issue with comparing models and observations is that models project the 2 meter air temperature over the globe while observations use the 2 meter air temperature over land but use the sea surface temperature for the ocean area. It sounds like there will be a paper out sometime this year on this issue. It should be interesting.

How about this: Suppose we apply a hypothesis test to each model, one at a time (which is at least semi-justifiable) and reject all of the models altogether whose envelope of predictions deviates from reality by some reasonable threshold. Then we can look at the predictions of the survivors. That would, of course, eliminate maybe 2/3 of the models from the "ensemble" right there, and the remaining models would have a much lower ECS. Where ECS itself has been in "free fall" for the last few years because the planet simply hasn't warmed as fast as the model ensemble average was predicting and it was becoming an embarrassment.

Different models have different strengths and weaknesses. The Coupled Model Intercomparison Project provides data for the modelers to explore those strengths and weaknesses and improve their models. You understand that even for a single model the the final output is a result of a number of model runs combined together, don't you? Another interesting test of models is to segregate out the individual model runs that by coincidence happened to match the real world natural variations (such as ENSO) or to rerun the model forcing the real world natural variations on them. When you do that you find they match the real world observations quite well. It feels to me that when you're looking at ECS in relation to climate models you're using too short a period to make your judgement on. There's a reason that 30 years is the classical period for climate (as defined by the World Meteorological Organization).

Sheesh! It's like I have to lead you like a toddler. When I said 9.3 I meant the whole thing. If you had any interest in actually finding out you would have found section 9.3.3 which has graphs that the ones in the IOP paper I linked to are derived from. But in the IOP paper they only used the projections up to 2015 whereas the IPCC report goes out to 2100 so it's a little difficult to compare them. Even if I lead you directly to the original source data
for the IPCC at the CMIP project the numbers would just overwhelm you since I doubt you have the scientific chops to understand them. It's difficult enough for me too but with enough work I could get there though it would probably take me several weeks.

how do natural gas generators beat that, and can whatever tricks those are be applied back to coal/gas/nuclear?

When it comes to heat engines you have something called the carnot efficency which is an upper bound on the possible efficency. The carnot efficiency is 1-(thot/tcold) where thot and tcold are absoloute temperatures. While real efficiencies are obviously lower than carnot it is still generally true that widening the gap between thot and tcold improves efficiency.

For a given type of heat engine the hot end is limited by material considerations wihle the cold end is limited by size considerations and/or ambient temperatures. However different types of heat engine work in different temperature ranges. By combining gas turbines which work at very high temperatures with steam turbines which operate at lower temperatures the temperature range can be widened and the efficiency improved.

However gas turbines are internal combustion which limits the fuel types. IIRC someone did try running one off powdered coal but the ash destroyed the turbine blades. I guess you could in theory build and external combustion gas turbine but you would need a very high termpature working fluid to carry heat to the turbine which I suspect would create practical engineering problems and drag down your overall efficiency. People have also run them off coal gassification systems but losses in gassification drag down the overall efficiency. So that leaves oil and gas.

Oil isn't used much* for electricity generation due to being substantially more expensive** than coal or gas. I expect the oil efficiency figures are dragged down by use in peak load plants, supplies for isolated communities and similar situations which are too small and/or too rarely used to justify CCGT.

* https://flowcharts.llnl.gov/co...
**http://www.eia.gov/tools/faqs/faq.cfm?id=107&t=3

Interestingly, this is how scientists recently showed that adult humans still generate brain cells.

Here is the actual patent:
http://pdfpiw.uspto.gov/.piw?P...

It's almost gibberish. It's full of sentences like (and I'm quoting)
"Alternatively, when propellant 18c of FIG 4 is utilized in the embodiment of FIG 1, the laser system 22 of Fig. 1 may comprise one or more free-electron lasers for providing pulsed laser beams to vaporize, using pulsed laser beams, pellets each comprising the propellant 18c of Fig 4."

Fig 1 is basically the drawing from the Business Insider article with the parts numbered. Fig 4 is a circle.

Or, it suggests we can use "light-emitting diode (LED) driven Alexandrite lasers" instead of free-electron lasers.
Or maybe a flash lamp driven ruby laser. No kidding.

And then the patent says that the fast neutrons from the Deuterium-Tritium fusion will cause the U-238 to fission and explode.
Again, quoting from the patent:
"The secondary explosion recompresses more of the Deuterium and Tritium, causing more fusion energy to be released beyond the 'breakeven' level vaporizing the remaining pellet materials of the propellant 18c of FIG 4 and increasing the overall thrust and exhaust velocity. Use of this embodiment reduces exhaust molecular weight, and increases exhaust velocity and specific impulse."

I did not mistype that.

I'm wondering how large it will be.
AFAIK, this is what a laser fusion device looks like, except that this one isn't ready for prime-time.
https://lasers.llnl.gov/media/...
Nor this one:
http://www.washington.edu/news...
http://www.washington.edu/news...

I would go with the free-electron laser because this is clearly an attempt to make the largest possible engine for the least thrust.

Also, looking at the diagrams in the article, I don't see anything that suggests they've addressed the problem that hitting the pellet with a laser on one side simply causes the pellet be vaporized and driven away without fusion (somewhat like squeezing a watermelon seed). How can they grant patents from devices that cannot work as designed?

If you go to the CMIP web site it has a list of the models used. Something tells me you're not into even a minimum amount of research and want everything handed to you on a silver platter.

Here's what they said about it in the press release:

This NASA dataset integrates actual measurements from around the world with data from climate simulations created by the international Fifth Coupled Model Intercomparison Project. These climate simulations used the best physical models of the climate system available to provide forecasts of what the global climate might look like under two different greenhouse gas emissions scenarios: a “business as usual” scenario based on current trends and an “extreme case” with a significant increase in emissions.

The web site for the Fifth CMIP is here.

If you dig around a bit you will find links to some but not all of the models used in the project. If you dig even deeper chances are you could get code for many of the other models from their original sources if you're nice to them.

18 years is not long enough. Also, it only works when you start in 1998 because it was a very hot year. But 2014 was even hotter, so it is wrong to say there has been no warming during those 18 years.

It wasn't started in 1997, it was started at the present and worked back as far as you could go without showing additional warming.
1. The basic premise of Global Warming is adding CO2 increases Earth surface air temperatures,
  2. Ben Santer said it takes 30 years to show an anthropogenic climatic signal,
each year that passes without additional warming casts increasing doubt on the GCMs future predictions; it will be difficult for the temperatures to increase fast enough to get bach on track for the predictions.

The Nasa climate scientists who claimed 2014 set a new record for global warmth last night admitted they were only 38 per cent sure this was true. Nasa climate scientists: We said 2014 was the warmest year on record... but we're only 38% sure we were right

Some people would consider saying something that they were only 38% sure of to be lying.

In the first place 18 years is not long enough to invalidate a climate model. In the second place observations are still within the ranges projected by models. Most of the graphs you see on climate model output are from the Coupled Model Intercomparison Project (CMIP) that takes the results hundreds of individual runs from about 30 different coupled* models and produces a weighted average and spread** of them. That smooths out the effects of natural variability so it's expected that observations will be under model projections part of the time and over them part of the time.

*Coupled means they have an atmospheric model and an ocean model coupled together.

**By spread I mean how big the range is when individual model runs differ from each other.

Yes, there are other factors at play and they are the subject of much study. To the extent possible they are included in climate models. A number of those factors such as ENSO, other oceanic cycles and volcanic activity are not predictable ahead of time so models have to simulate realistic scenarios for them. A recent paper "Well-estimated global surface warming in climate projections selected for ENSO phase" by Risbey, et. al. found that when you selected the model runs that were largely in phase with with the real world ENSO they matched observations pretty well.

Considering that CO2 increase has been constant, but temperatures have not significantly risen in the last 18 years.

Yet the oceans where more than 90% of the energy goes have continued to warm and major ice sheets have continued to melt. It doesn't take much of a shift in the amount of heat absorbed by the oceans to significantly affect the atmosphere.

The engine room of "Into Darkness" was the finest ignition facility in the American armada!
https://www.llnl.gov/news/nati...

You're right, and that made a HUGE difference in the look of the film. But I'm pretty sure that some of the shots were still in the brewery.

The engine room of "Into Darkness" was the finest ignition facility in the American armada! https://www.llnl.gov/news/nati...

Disappointing that the Star Trek tie-in was mentioned but the link was omitted...

National Ignition Facility provides backdrop for "Star Trek: Into Darkness"

As far as I am aware the highest radiation dose anyone has received was Cecil Kelley, whom was exposed to a criticality accident at a plutonium processing plant. When the tank stirrer turned on, the geometry of the plutonium solution became critical, exposing him to ~12,000 rem. He died 36 hours later.

See Page 16 for a description of the accident here: http://ncsp.llnl.gov/basic_ref/la-13638.pdf

Or the wiki here: http://en.wikipedia.org/wiki/Cecil_Kelley_criticality_accident

If you want to quote Dr. Santer, then link to his press release.
https://www.llnl.gov/news/news...

By linking instead to wattsupwiththat, you're just self-labellling yourself as a denialist moron.

don't be surprised when we reach that timeline with no warming and then take the good Doctor at his word - there is a pause in global warming, and we only need 17 years to make that determination.

Wait? Why do we have to wait? We always have 17 years of latest data. You mean wait until you can cherry pick a period starting with 1998 again? Moron.

I recall seeeing a youtube video where a scientist set a pile of carbon nano-tube fibers on a surface and flashed them with a typical photo-strobe, and after a few seconds sparks were visiable in the fibers and a few more the pile burst into flames! So appearently the plan is to take one of the highest conducting non-superconducting materials we know of, attach it to the Earth at the equater where almost daily rain and thunder storms occure due to the moisture and static generated by the Trade winds and have machines shinny up and down the worlds biggest lightning rod that is likely to burst into flames if lightning even gets near it; man I would love to see that!

There are several datasets that show a long term warming trend.

Can you point me to one of those datasets, that shows warming over the last 17 years?

17 years is a clear attempt to cherry pick data as outlined in this Forbes article.

Nope. It's not cherry picking. Dr. Ben Santer, Lawrence Livermore and UEA's CRU climate scientist explicitly stated that you need 17 years to identify a trend. That's a world-respected climatologist who's been unapologetic about his support for the AGW model. Well, we've surpassed his 17 year statement. Can we now identify a trend?

After all, Phil Jones, Richard Lindzen, and Pat Michaels (all noted climatologists from the pro-AGW side of things) have identified the trend starting about 17 years ago. So the trend does exist, and per a respected climatologist, it's plenty long to identify as an independent trend - separate the signal from the noise.

PS: in accordance with standard AGW-supporter techniques, you can only disagree with the 17 year claim if you, in fact, are a PhD climatologist. They are apparently the only ones who can speak authoritatively on issues related to climate. Otherwise you need to have credentialed climatologists who state that the trend does not exist, or that the 17 year period is not suitable for identifying climate trends.

• * The per-clock throughput of CPUs has been on an exponential increase curve, per last decades, ie. last 20-30 years matter a lot
• * The proliferation of silicon compute devices has been on the increase for decades, ie last 10-20 years matter a lot (from the very small to the very big)
• * The duty cycle of most silicon devices is not 100%; there are two notable exceptions: infrastructure (embedded) systems and supercomputers
• * Parallel computation implies fast interconnects for quick message passing; for some initial info, ref. https://computing.llnl.gov/tutorials/parallel_comp/
• * Despite regular interconnect technology upgrades, MPI has been the agreed standard for message passing since the early 90s

ref. http://www.top500.org/ ->Statistics for more details; so, here we talk about big machines, of high duty cycle, of using mostly a uniform API for synchronization.

Given Weather, Climate & Computational Fluid Dynamics codes (and some more), the temporal density of such calls is pretty good.

Concluding, MPI should be the most common *API* being called nowadays per unit of time;
there is still room for challenging this though: MPI Send/Recv calls have a few variants plus,
MPI stack implementors may have fragmented the codebase, to declare a clear winner...

HPC scale: https://computing.llnl.gov/linux/downloads.html

There's an interesting case study on Chapel, Charm++, Liszt, and Loci in:

Exploring Traditional and Emerging Parallel Programming Models Using a Proxy Application
presentation
paper

..size of Rhode Island monstrosity that runs off unicorn horns and yeti farts.

That sounds like it would be worth doing on its own merit. :-)

Fusion research has made painfully slow progress and no amount of techno-masturbation, wish fullfillment, Iron Man fantasy makes it otherwise.

It sure has, but that's not entirely down to fusion is hard, we all know it hasn't received the funding it should, yadda.

What LM Skunkworks is doing does not appear to be anything revolutionary. Their designs are a completely believable evolutionary progression of the current thinking. Seriously, how hard is it to imagine one of these clever scientists having a lightbulb appear over his or her head, turning to a colleague and asking "I wonder how a cylinder might work?" Spheromaks already exist so it isn't much of a logical leap (not detracting in any way from their great work.)

Of course to the faithful, it's always just around the corner and there are an endless stream of excuses why it wasn't, when looking retrospectively. It's like talking to a doomsday evangelist. "Fusion is Near! Repent ye lost souls of carbon!"

There's no need for faith in Science, only faith in the Scientific Method. Even that is just bad poetry; there is already plenty of obvious evidence of the Scientific Method's efficacy.

No one seems to realize the NIF is just to study fusion for weapons research. It has no hope of ever leading to an energy source.

People don't realize this because it is NOT true.

NIF does not have a single mission. A big part is weapons research, but that's not all they do.

Below is a quote from their web site:

Achieving nuclear fusion in the laboratory is at the heart of the directorate's three complementary missions:

* Helping ensure the nation's security without nuclear weapons testing (see National Security).

* Blazing the path to a carbon-free energy future (see Energy for the Future).

* Achieving breakthroughs in a wide variety of scientific disciplines, including astrophysics, materials science, the use of lasers in medicine, radioactive and hazardous waste treatment, particle physics, and X-ray and neutron science (see Understanding the Universe).

He should call it radiation pressure, not light pressure, because of the different wavelength of x-rays. But the problem is the same, physically he is correct.

You need to educate yourself: https://lasers.llnl.gov/programs/nic/icf/how_icf_works.php

The X-rays ablate the hohlraum walls. Those walls, which are not any form of radiation but rather ionized solid material, implode on the fuel and compress it. You could replace the lasers with some other source to ablate the hohlraum.

Thus, all compression is from the imploding hohlraum, not radiation and the statement "Doubt it. Light pressure is what compresses and heats the fuel." is incorrect.

It seems that in this case, indeed, Helium would be the byproduct. More specifically, Helium-4 according to the list of important fusion reactions on Wikipedia.

But as you can see from this list, there are several fusion reactions theoretically available for terrestrial use - most produce Helium, but there are also ones that produce isotopes of Beryllium, Tritium, Lithium and even an aneutronic one that produces Carbon.

Nonetheless, the high energy yield of fusion reactions means that, although we'd get immense amounts of energy from them, the amount of Helium created as a byproduct would be negligible, so it would be unlikely to solve our helium shortage problems. Much more likely is that availability of cheap, safe and clean energy from fusion would make it feasible to establish permanent mining colonies on the moon to extract helium from its soil and deliver it to Earth.

The story makes it sound like there's no support for Fortran in MPI but there totally is: https://computing.llnl.gov/tutorials/mpi/ I recognize that some kind of abstraction layer in a support library on top of MPI might be useful, but let's call it what it is.

I write massively parallel scientific simulation software for a living (the kind that runs on the biggest machines in the world)... and trying to come up with a way to display GBs or TBs of information from some of our largest simulations can be _tough_.

We use several open source packages ( Mostly http://www.paraview.org/paraview/resources/software.php and https://wci.llnl.gov/codes/visit/ ), but most of our best visualizations are actual done using a commercial package ( http://www.ceisoftware.com/ )

For some examples check out the YouTube video here: http://www.youtube.com/watch?v=V-2VfET8SNw

(That's me talking in the video). Those aren't necessarily our best visualizations - just some that happen to be on YouTube...

We find that the reactions to these simulations are mixed. They are certainly eye-pleasing... but sometimes if you go too far in making it look good it can actually turn scientists off. They will start to think that it looks "too good to be true" (I literally had a senior scientist in a room of 200 stand up at the end of one of my presentations and proclaim that "This is too good to be true!"). Because of this we try to do do just enough visualization that you can see all of the features of the simulation and understand what's happening without going overboard.

You have to realize that a lot of scientists still remember the days when they created line plots _by hand_ for publications! I suspect that as young scientists come up through the ranks this feeling that "slick graphics = not real" will go away.

At least, I hope....

Sorry, forgot to also post link to the OpenMP tutorial:
https://computing.llnl.gov/tutorials/openMP/

There is a high barrier to entry for piddling around with graphic cards. Fortunately, most home computers are already parallel (2-8 cores). I do extensive parallel programming, and I do most of the testing (for small problems, anyhow) on my desktop or laptop, which each have 8 cores.

There is simply a set of "parallel" function calls which can be built directly into your code. You then just need to compile your code with the proper libraries, usually either mpich or OpenMP. I believe both are available in the ubuntu repository. Pick the one that is most promising for your problem. They are fundamentally two distinct approaches to parallelism, each useful at times. Lawrence Livermore has a great tutorial site, including loads of example fortran and C codes of both openMP and mpich. THey are ready to compile and run on your home computer. Happy computing!

Tutorial: https://computing.llnl.gov/tutorials/mpi/
Example codes: https://computing.llnl.gov/tutorials/mpi/exercise.html#Exercise1

There is a high barrier to entry for piddling around with graphic cards. Fortunately, most home computers are already parallel (2-8 cores). I do extensive parallel programming, and I do most of the testing (for small problems, anyhow) on my desktop or laptop, which each have 8 cores.

There is simply a set of "parallel" function calls which can be built directly into your code. You then just need to compile your code with the proper libraries, usually either mpich or OpenMP. I believe both are available in the ubuntu repository. Pick the one that is most promising for your problem. They are fundamentally two distinct approaches to parallelism, each useful at times. Lawrence Livermore has a great tutorial site, including loads of example fortran and C codes of both openMP and mpich. THey are ready to compile and run on your home computer. Happy computing!

Tutorial: https://computing.llnl.gov/tutorials/mpi/
Example codes: https://computing.llnl.gov/tutorials/mpi/exercise.html#Exercise1

Looking at the energy flow diagrams from LLNL https://flowcharts.llnl.gov/energy.html#2011 they only have the state-by-state breakdown for 2008 https://flowcharts.llnl.gov/content/energy/energy_archive/energy_flow_2008/2008StateEnergy.pdf. The total electricity generation from renewable (solar,hydro,wind,geothermal) comes to 6.6+237.8+53.1+270.8 = 568.3 BTU. As a fraction of the electricity generation of 1907.9 BTU this comes to 29.8%. Seems a little closer than half-way there already. But since we are only talking about increasing solar and wind, they would need an extra 67.7 BTU to meet this target, so if they doubled wind, and tripled solar then they would reach this target, so your comment of "aren't half-way there yet", makes sense if you look at it this way.

Looking at the energy flow diagrams from LLNL https://flowcharts.llnl.gov/energy.html#2011 they only have the state-by-state breakdown for 2008 https://flowcharts.llnl.gov/content/energy/energy_archive/energy_flow_2008/2008StateEnergy.pdf. The total electricity generation from renewable (solar,hydro,wind,geothermal) comes to 6.6+237.8+53.1+270.8 = 568.3 BTU. As a fraction of the electricity generation of 1907.9 BTU this comes to 29.8%. Seems a little closer than half-way there already. But since we are only talking about increasing solar and wind, they would need an extra 67.7 BTU to meet this target, so if they doubled wind, and tripled solar then they would reach this target, so your comment of "aren't half-way there yet", makes sense if you look at it this way.

. These views include disbelief in climate change and skepticism science in general

Well lets see what the scientists have to say,

LIVERMORE, Calif. -- In order to separate human-caused global warming from the "noise" of purely natural climate fluctuations, temperature records must be at least 17 years long, according to climate scientists. ...
The research team is made up of Santer and Livermore colleagues Charles Doutriaux, Peter Caldwell, Peter Gleckler, Detelina Ivanova, and Karl Taylor, and includes collaborators from Remote Sensing Systems, the National Center for Atmospheric Research, the University of Colorado, the Canadian Centre for Climate Modeling and Analysis, the National Oceanic and Atmospheric Administration, the U.K. Meteorology Office Hadley Centre, and Lawrence Berkeley National Laboratory. Separating signal and noise in climate warming

so if your saying you follow the science then Dr. Ben Santer’s 17-year test: "if there is no warming for 17 years, the models are wrong." applies and there has been no statistically significant warming for 17 years and 4 months, so unless your one of the "anti-scientist" republicans, there is no climate change due to human activity.

"The simulations rule out (at the 95% level) zero trends for intervals of 15 yr or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy with the expected present-day warming rate."

  "In order to separate human-caused global warming from the "noise" of purely natural climate fluctuations, temperature records must be at least 17 years long, according to climate scientists."

The IPCC’s Fifth Assessment Report suggests that the Earth will warm rapidly in the 21st century. However, this is not being borne out by observations.

No-one disputes that the earth's atmosphere is warming - this has been going on for some time now. What is disputed is the contribution that human activity makes to the degree/acceleration/rapidity of warming. The original models had man's contribution to an increase in warming as minimal at best. Then the IPCC re-jigged the models to take into account the theory that CO2 (and other emissions) would cause a climate forcing, i.e. the effect of the increasing CO2 levels would not be linear but would drive GW at a much higher rate than what would be expected naturally. These models have all predicted rapidly increasing global temperatures with no pausing. In order to account for small variations in the annual results, the IPCC et al initially said you needed 10 years of no warming to invalidate the models. Then as 10 years got close, that became 15 years.* Then 17 years. That has now come to pass. Even the most conservative of models do not match the observed results, therefore it's time to revisit the modelling.

*"The simulations rule out (at the 95% level) zero trends for intervals of 15 yr or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy with the expected present-day warming rate. From: http://www1.ncdc.noaa.gov/pub/data/cmb/bams-sotc/climate-assessment-2008-lo-rez.pdf
It's a very large PDF.

For the purpose of this post, I'll accept that they can convert protons to neutrons as described, although I'm very dubious about this.

Here is a table of nickel isotopes.
Here is the first source I found for neutron cross sections of nickel isotopes (pdf). (See figure 12, look at the left hand side of each 'destruction channels for ??Ni' plot for what low energy (thermal) neutrons will do.)

Cross sections are in barns, and are approximate as I'm eyeballing them off a logarithmic scale.
58Ni [stable, 68% abundant] (0.006 barn) -> 59Ni [-> 59Co, 76000 yr half life]
59Ni [unstable but long lived] (0.02b) -> 59Co [stable] or (0.005b) ->56 Fe [stable] or (0.004b) -> 60Ni [stable]
60Ni [stable, 26%] (0.006b) -> 61Ni [stable]
61Ni [stable, 1%] (0.002b) -> 62Ni [stable]
62Ni [stable, 4%] (0.006b) -> 63Ni [->63Cu, 100yr]
63Ni [unstable] (0.001b)-> 64Ni [stable]
64Ni [stable, 1%] (0.004b) -> 65Ni [->65Cu, 2.5 hr]

None of the cross sections are hugely larger than the others, so all these reactions will occur with reasonable frequency. So irradiating nickel with thermal neutrons, you are going to produce radioactive 59Co (76000yr), 63Ni (100yr) and 65Ni (2.5hr). The 65Ni isn't a problem - turn off the reactor, wait a couple of days, and it will all be gone. The 59Co is only a bit of a problem - with such a long half life, it isn't very radioactive. The 63Ni however is nasty. Like 137Cs (30yr) from the Fukashima reactors, the half life is short enough to be quite radioactive but long enough that you can't just wait it out. Finally, the nickel won't be 100% pure, so you have to worry about what neutron irradiation will do to the impurities.

The 65Ni means when you turn off your reactor, it will continue to produce residual heat for hours.

The article gives the impression that weak nuclear reactions aren't dangerous, but this is not so. If it were, nuclear reactor waste wouldn't be dangerous.

This reactor will be producing ionizing radiation when running (mostly gamma rays, some beta rays mostly from 65Ni decay, and a tiny amount of alpha particles from 59Co(n,a)56Fe.) This will require some pretty heavy shielding to stop it. (A good sized water bath should work, every 7cm of water halves the radiation and you want hot water anyhow. But concrete is less prone to leak away.) You'd also need to worry about stray neutrons, although I expect that can be fixed with a thin layer of something that has very high thermal neutron cross section and no dangerous daughter products.

In short, I don't think I want this in my basement.

The Department of Energy is the one with the extra cycles.

Well a quick google and we get this

https://asc.llnl.gov/computing_resources/sequoia/configuration.html

16 cores per CPU/chip. (or according to wikipeda 18, but one is used for the OS and one is saved as a spare).

Note also that each dimension of the torus does not have to be the same, so the constraint is

No of cores = 16*A*B*C*D*E

That is assuming each node on the torus has 16 cores (potentially could be a multiple of 16).

Anyway according to
http://en.wikipedia.org/wiki/Blue_Gene

The system is 96 racks
Each rack is 32 node cards, each node card has 32 modules, each module has 16 cores.=1,572864 total cores.

I would guess there are two dimensions between the racks (12 x 8) then the remaining dimensions are inside the racks, potentially 12x8x16x2x32 but probably split a bit more than that so the longest dimension is not 32, 12x8x16x4x16 maybe, where there are two seperate links inside the node cards splitting them into two groups of 16 modules.

There is a couple of different ways they could wire up the 3 dimensions in the cabinets so that part it is hard to be sure.

https://computing.llnl.gov/tutorials/bgq/images/5Dtorus.jpg
.
is the relevant picture showing a "Midplane, 512 nodes, 4x4x4x4x2 Torus". So a five-dimensional torus of size 4 x 4 x 4 x 4 x 2 is divisible by three.