Re: Judges names
> Which publication used “Sue, Grabbit & Runne. Solicitors and Commissioners for Oaths.”?
Private Eye, when not discussing Ugandan Affairs
146 posts • joined 26 Jun 2007
Macho flops/£ and achieved flops/£ are very different things, especially once software development and maintenance costs are factored in, and also electricity bills. But the honest truth is for this project the vast majority of users didn't need or want GPGPUs to do their science, and in many cases GPU versions of the applications just don't exist, so early on a decision was made to be purely CPU based - the machine is there to do science, and if it can't do the science the user base needs it doesn't matter how many macho flops it can do. As a result I can't answer your question as we never did that calculation.
Sigh ... To all those who don't understand the archaic temperature units which are the primary measure in the article can I remind you of
so they can be converted into something more comprehensible. For starters 120F=2.9Hn
Just to add some numbers over the period 2014-2016 UK researchers and Innovators received 15.2% of the Horizon 2020 funding, for a total of just under €3.8bn funding. This was the second highest in Europe, just behind Germany (16.7%): https://ec.europa.eu/unitedkingdom/news/horizon-2020-three-years-uk-tops-league-participants_en
"To be sure, 6 minutes instead of 16 years is quite the improvement, but only some cases ?
I do hope that they know which cases, because it would be a shame if they applied it wrong yet still used the answer that their statistical analysis machine gave them."
Well speaking as somebody who develops DFT codes on HPC machines I would use the following technical term to describe their comparison: Bollocks. You can solve systems with a few thousand atoms relatively easily on a few 10s or 100s of cores on a modern HPC machine, maybe 1200 hours is fair, but 16 years, WTF are they on?
Oh, and of course out of a quantum mechanical calculation you get the wavefunction which allows deep insight into the properties of the system under investigation. Out of curve fitting you get the fitted curve. Don't get me wrong, this can be extremely useful in certain circumstances. But what is being presented here is complete over-sell - "News: Drawing a line through some points is quicker and easier than solving the fundamental laws of nature, film at 11!!!"
"We want people to be able to take advantage of the wide choice of communication services available and shop around with confidence, so that they can get the best deals for their needs."
On the other hand we could have a situation where companies don't have the right to rip the customer off the moment they put a foot wrong. We could have a situation where the default is that companies can not rip off consumers at will. Maybe governments could understand that while representing me one very important responsibility is to spare me working through a plethora of compare the market type sites and rather let me HAVE A LIFE. But no. As long as their hedge fund managers maximise their profit that is all that matters. I am just a working droid to maximise somebody else's profits. Fuck 'em.
Great, we can build these machines on some kind of timescale. But without the applications that can exploit them what is the point? Where's the funding for the software development, especially if novel architectures require a large overhaul of existing million line codes?
"The BLOODHOUND Project is a global Engineering Adventure, using a 1,000mph World Land Speed Record attempt to inspire the next generation to enjoy, explore and get involved in science, technology, engineering and mathematics."
Go and see one of their presentations. They are bloody brilliant.
"What is disappointing is that Uber's immediate response is throw rattles out of the pram and publicly state that they will go to the courts"
Disappointing maybe, but surprising, no. This seems to be their basic corporate policy - if crossed threaten litigation. One wonders if they learnt from a certain well known cult.
Obscure!? Sigh ... kids these days
That said I'm mildly confused by this. There's no reason boron won't be found in Mar's surface, and it's no big surprise that any boron on Mars will be in the form of borates given the highly oxidised state of the surface - on Earth (almost?) all boron containing minerals are borates. So given that why the big fuss about life? For me the most amazing thing is that the truly amazing Curiosity rover can detect a rare element like boron - I wonder what the concentration is?
There have been legitimate uses for comparing a variable to itself - in particular under IEEE754 all comparisons that have a NaN as one of the operands return false, so checking for equality of a variable with itself was a way of detecting whether the variable held a NaN. Unfortunately optimising compilers used to think "what is this idiocy" and throw it away, hence many languages contain a isNaN function or similar.
"You may want to try throwing a set of GPUs at a particular computational fluid dynamics problem to see if that architecture can handle the workload in a more effective way. If it doesn’t deliver the gains you expected, then you haven’t sunk capital into a hardware investment"
Yada, yada, yada. I take the point, effectively renting unusual hardware, but GPUs are a poor example, in academia (if not Industry) they have moved into the mainstream and any self respecting central University service will offer nodes including them. So how about FPGAs or Xeon Phi's? You offer them, great ... how much ?!?
The cloud being of use to academic HPC currently is a myth. High throughput computing, yes there is a place. But at the moment not HPC.
Don't quite get why this is particularly a Fortran problem - it's simply a result of floating point maths. In fact to expect bit wise identical results for every compiler/library/etc. combination in complex codes like those discussed here displays some ignorance about the nature of the beast, and that's before we even think about talking about parallelism. And in practice for the performance these guys need your choice of language is Fortran, C or C++, nothing else will cut it however much quiche you eat, and for all 3 the floating point issues are similar.
They're not talking about the actual mass, they're talking about the "effective mass". Wikipedia has an article on it
but I admit I haven't read it to see how good it is
Give the explosion of caffeine options since I were a lad I'm surprised he didn't do this via the standard Unix build automation tool. Then he could do things like
if he so wanted
Yes there is a difference. But either will be a disaster. May wants to cement the surveillance state. Leadsom has no interest in the "common man", she merely wants to legalise the (in my opinion) tax scams she has run throughout her career - see Private Eye passim.
"There’s a twist in the competition on this task: for the first time ever, students have to write their own algorithm to solve the graph problem. They aren’t allowed to use the reference implementation – they have to provide their own approach. Teams will be judged on the quality of their code as well as their solution to the problem. It’ll be interesting to see what happens on this application."
Good. Call me a grumpy old sod (you're a grumpy old sod Ian), but while good fun and a demonstration of problem solving ability I struggle to see the real relevance of these competitions; you simply don't cobble together your own clusters any more, or at least you shouldn't be doing. But often the software is a different matter - where's the recognition of the people who write the stuff that can actually exploit this hardware? While I've seen the names of the teams competing and we're promised videos introducing them there's not even a link to the applications web pages, let alone naming the teams or institutions that develop them.
I must get out more ...
Biting the hand that feeds IT © 1998–2020