That is Baizuo logic..
HN user
0x07c0
"In other words, high-performance code should ideally run so fast that any further improvement in the code would be pointless.
And then define pointless, to make a quantified decision on optimization you should use a performance model. Common model is the roofline model. https://en.wikipedia.org/wiki/Roofline_model
My comment was to the absurdity of having a trail for a 98 year old man. As I get from your emotional involvement in this case, you come from the region? As you then probably know, in one country the SS (local) are among many considered heroes, in the neighbour country criminals. Elements of the local SS units was after the war taken over by the CIA(Gehlen Organization). And was waging guerilla war against the soviets. There successor units are presently fighting in the eastern Ukraine and are US allies (Right sector and others)..
You're not serious. What a despicable claim. A very strong argument based on facts ?
Between 500.000 -2.000.000 German civilians, from the eastern part of Germany parish in the period 1945-1950.
https://en.wikipedia.org/wiki/Flight_and_expulsion_of_German...
Previous Polish governments have triad to take actions against some of the perpetrators:
http://www.telegraph.co.uk/news/worldnews/europe/poland/1376...
While I personally is against dragging old people in front of courts 70 years after the alleged crimes (polish or German), at list the Polish do not discriminate.
Most European country’s at least had that. Populist sentiments are of course against this.
And the polish killed their fare share of German civilians.. It's a reason that statute of limitations is considered a part of a civilized justice system. Sometimes letting the past be the past and move on is a good thing.
I did not watch you link, but many modern chemical weapons are binary compounds. Meaning the two compounds has to be mixed to get the final weapon. This makes leaking etc, not as big problem as leaking of actual chemical weapons...
This also my be relevant:
pre-commitment 2: Ecuador eae5c9b064ed649ba468f0800abf8b56ae5cfe355b93b1ce90a1b92a48a9ab7
pre-commitment 3: UK FCO f33a6de5c627e3270ed3e02f62cd0c857467a780cf6123d2172d80d02a072f74
claim from intelAnon that Ecuador had caved in to pressure from the Clintons:
https://pbs.twimg.com/media/Cu7pNidVIAAUweO.jpg:large
The hashes then being a message from Assange, or it my be Assange being paranoid.. Or something else entirely.
"Several states have reported attempted scans of their computer systems".. I don't find it likely that the Russians have mastered port scanning technology, this bears all the hallmarks of the hacker 4chan!
Also, this is getting really ridicules..
That population would be below the level necessary to maintain genetic diversity. At least in NZ, the Kiwis have the decency to say that they want a species eliminated outright. The Norwegian wolf population is a common population with the Swedish population. The total is something like 400 wolfs. (For the record I support a robust wolf population.)
they are not very noob friendly, "shift reduce conflict found in state 111" etc :)
Not looked at OPs code, but if you build this stuff how a compiler book tells you to... The front end of a interpreter and a compiler is the same. You build a AST, if you interpret you well interpret that. If you have a compiler, the AST is turned into the target language, in most cases asm. The front end is the scanner/Lexer, the parser and most likely semantic analyser. So to answer you question :), if you make a non optimizing compiler, turning the AST into asm is trivial. The hard part of making compilers is the optimization. Most of the front end you do not need to write actually! There are compiler-compilers for that.
https://en.wikipedia.org/wiki/Abstract_syntax_tree
https://en.wikipedia.org/wiki/Compiler#Front_end
https://en.wikipedia.org/wiki/Lexical_analysis
https://en.wikipedia.org/wiki/Parsing
https://en.wikipedia.org/wiki/Semantic_analysis_%28compilers...
This is heresy against the prophets K&R!
This being Iceland everyone knows that the world was created by the Gods ,emphasis on the s, not God, so stupid question. But on a more serious note the Norse creation myth is actually interesting:
http://norse-mythology.org/tales/norse-creation-myth/
Shorter version:
http://www.pitt.edu/~dash/creation.html
Executive summary:
We live inside the dead ice giants Ymir's eyebrows...
Interesting link. Radiation hormesis seems to be a real thing.
You may be right, but some people start believing there own delusions.. This can make them much more believable. I was thinking something like: https://en.wikipedia.org/wiki/Grandiose_delusions
That would be the rational thing to do, but you assumes Wright is sane. Witch he probably is not. Also has someone look into the claim that he has the 17. fastest Super Computer? This things draw a loot of power (4,499.87 kW he claims) and produces a loot of heat. You cant hide it in your mothers basement.., local authorities will know about it. So will the power company! So there should be a paper trail there? (also SGI denied selling him the system) Wright's alleged system, was 15. now 17. fastest in the world: http://www.top500.org/system/178468
The Tesla K40 has peak double performance of ~1.4 TFLOPS. It has 64 DP cores, the warp scheduler can schedule four warps per smx per cycle. It can therefore have two warps executing double instructions at the same time. But the number is not very interesting, the memory bandwidth on the other hand is, a GK110 has 288GB/s, take you code, get it's arithmetic intensity and you have a upper bound for your performance, assuming you are memory bound of course.
https://www.nvidia.com/content/PDF/kepler/NVIDIA-Kepler-GK11...
https://www.nvidia.com/content/tesla/pdf/nvidia-tesla-k40-20...
This is the same as the police helping ducklings over the road, or the fire department saving cats down from trees...
"made unintelligible by a feature, product, or service owned, controlled, created, or provided by the [company]."
So no more Hash functions I guess, better store those passwords in clear text. Also encryption is the key to any form of distributed system where you have misbehaving actors.., you know like the internet..
PCIe is the huge bottleneck... The two Xeon's in the box..
It's a wast to put Xeon's on this things if they use the PCIe, you end up in a loot of cases only using them to drive the GPU's.
When I talk about balanced (which is a huge influence in my architectural and system level designs)...
The DP performance on Tesla's is ridiculous, think it is a marketing ploy. People talk of buying gaming cards.., as you are almost always memory bound..
I'm working on my own HPC targeted chip..
Looks nice, you are throwing out all HW bloat and doing everything in software? Are you planing to have some form of OS running on this chips?
Thor Heyerdahl had this theory, Columbus did not go to Norway but his family is of the Norwegian Nobel Bonde family. They lost a power struggle, and had to go abroad, to Genova . This as I recall is based on a book written by a family member of Columbus, think the brother(if he had one). Stating that the name of the family is not from Italy, but it a Latinifaction of the original language from where the family is from. Bonde means farmer (In Norwegian), Columbus apparently means something of the same. He should then know about Vinland (name of America given by Norse explorer) from his family roots in Norway. I also think part of the theory also was him being a period (some years)aboard a Danish or Norwegian ship, this would have given him hint of a land on the other side of the ocean. Also there are Danish tax records stating that the fishing banks outside New Foundland was taxed by the Danish crown, from before Columbus. There are also tails of Norse-Indian settlements encountered by early settlers.. And I think some one found a Indian skeleton in Denmark or Norway some years back.
Only article I found.. (In Norwegian)
http://tux1.aftenposten.no/kul_und/kultur/d105453.htm
Some more sources here:
https://en.wikipedia.org/wiki/Norse_colonization_of_the_Amer...
Tesla's has more Double Float cores compared to gaming cards.
Was thinking the same, but look at the memory bandwidth of this thing, 720GB/sec.* That is the number you should look at, and that is a sweet number... Also the NVLink tech looks nice for multi-GPU/heterogeneous computing (I guess the latency is the most important, but no idea how that is ). Do's some one know if the Xeons are connected to the GPUs with NVLink or are they on PCIE ? (I know the new POWER chips have NVlink but haven’t read that Intel supports it.)
*http://www.anandtech.com/show/10222/nvidia-announces-tesla-p...
Thanks! This are really useful. (I'm actually now making a small dsl for distributing work on accelerators.)
I personally really like the idea from the the Halide language, having one language for algo, another for how the computation is done. If something like that could be made general purpose it would be very useful.
should be done in a macro.. Encourage c programmers to use macros is like encouraging alcoholics to drink :) But I guess you didn't think about pre processor macros.
I wish that to be true, but that is not what I'm seeing. (I'm doing HPC.) . It's not about native c performance vs some other language. Its about the the low level stuff you can do in C. You use avx (and the compiler don't help(they are supposed to, but don't do it very well), you have to use intrinsics or asm), then memory stuff, cash blocking, alignment, non temporal stuff. Same for CUDA, compiler don't get that much performance. You have to think about all low level stuff, usually memory, like alignment, use shared memory or not, cash line size etc.. . And then you are using multiple GPUs.. No help from compiler, you have to do all by your self. Had been nice with compiler doing it, and there are some compiler that helps. But you don't get max performance, and with some effort the performance you get by handcoding all this stuff is much greater then what compilers can give you. And that advantage is increasing.
+1
Before the difference between highly optimized code and ok code was maybe 2-3x speedup. Roughly one Moore's law doubling. With heterogeneous computing that is more like 20-30x or more. And Moore's law is dead! This will change a loot in the IT world (More servers are the solution, developer time is more expensive then computer time, etc..). Learn C, down on the metal programming is back, the future is heterogeneous parallel computing.
That said though, one advantage if you are diligent is that you probably by necessity learn techniques of checking your work (formally and informally) earlier than typical students, which is a good thing.
Or you can post them on math stackexchange and get feedback for free from bored? mathematicians :)
That is what we have fast math for :)
restrict is probably a better example! https://en.wikipedia.org/wiki/Restrict