a conventional CPU-based supercomputer doesn't have the necessary I/O bandwidth to do the work.
I work in HPC and the trend is towards heterogeneous architectures ( CPU+accelerators). Moore's law, power requirements and economics are dictating that trend. It's definitely a stretch to claim that you get better I/O bandwidth with GPUs. Even with PCI Gen 3, the effective bandwidth you get per CPU core is greater than that of an 'equivalent' GPU core.
Like many out there, I'm surviving the recent GNOME "upgrades" by running fallback mode which mimics GNOME 2.x. That's the only means to maintain sanity and a semblance of productivity.
Going at this rate, keep an eye for a GNOME branded one-button mouse, because right-click is for pussies.
I agree completely.
I just hope that Doctor Who would be more about the Timelord and less about the humans (Amy and Rory can go jump off a cliff for all I care). Too much drama about the two is nauseating.
http://sitebar.org/ is there but it doesn't have support for tags last time I checked. I'm in the same boat as you are, guess we just need to develop those capabilities now.
I use Lustre on the petaflop supercomputer at Oak Ridge and mostly satisfied with the I/O performance even for 64k+ cores. Hence my worry about its future direction. Oracle doesn't have any interest in HPC and hence not sure how they will treat Lustre. I would be very happy to admit if my worries are misplaced.
The login nodes run Linux but the compute nodes run CNL (Compute Node Linux), a lightweight OS designed to reduce system overhead. On an earlier version of this machine (XT3), the compute nodes used to run Catamount, a different lightweight OS.
Short answer: Quoting just one example from Astrophysics from DOE Report
Core collapse supernovae simulations with the spatial resolution required to properly model critical aspects of the explosion
dynamics (e.g., the evolution of the stellar core magnetic fields and their role in generating the supernova) will require
much higher resolution than today's terascale codes. These codes, in turn, will require exascale computing, particularly if a
number of simulations are to be performed across the range of stellar progenitors and input physics. One such simulation is
expected to take ~8 weeks, assuming 20% efficiency on an exaflops machine.
Conclusion of Report:
The broad computational science community has a golden opportunity to accelerate the availability of usable exascale
systems. To take full advantage of this opportunity to deliver exascale computing by 2017 will require an integrated program
of investments in hardware and software research and development, (R&D). Also required will be a tight coupling to a
selected set of science communities and the associated applied mathematics R&D. In some cases, such as astrophysics
and climate, the communities are well on the way to exploiting petascale systems. In other cases, such as socioeconomics
and multiscale biology, there is great opportunity for acceleration. Computational science and engineering opportunities in
energy are wide and deep and have an enormous potential impact on advancing energy technology and fundamental science.
If acceleration is to be achieved -- and there is every reason to both desire it and believe that it can be accomplished -- then
every minute will count, and even modest investments early in the cycle (e.g., 2008 and 2009) could have dramatic benefit
and will reduce uncertainties moving ahead.
Important applications(not an exhaustive list by any means) as gleaned from that Town Hall Meetings Report:
Energy Energy research offers significant opportunities to exploit computing at the exascale, in order to advance our understanding
of basic processes in areas such as combustion, which would naturally lead to a design capability for improving the efficient
use of liquid fuels, whether from fossil sources or renewable sources. First-principles computational design and optimization
of catalysts will become possible at the exascale, as will de novo design of biologically mediated pathways for energy
conversion.
Access to exascale systems and the appropriate applications codes could have a dramatic impact on nuclear fission reactor
design and optimization and would help accelerate understanding of key plasma physics phenomena in fusion science
critical to getting the most from the U.S. investment in ITER.
Exascale systems should also enable a major paradigm shift in the use of large-scale optimization techniques to search for
near-optimal solutions to engineering problems. Many energy and industrial problems are amenable to such an approach, in
which many petascale instances of the problem are run simultaneously under the control of a global optimization procedure
that can focus the search on parameters that produce an optimal outcome.
Environment
Three broad areas relating to the environment were discussed: climate modeling; integrated energy, economics, and
environmental modeling; and multiscale biological modeling from molecules to ecosystems.
Climate modeling. As the most mature of the three environmental application areas, climate modeling is expected to make
good use of exascale systems. The impact of these systems will be threefold.
Google also successfully rebuffed the U.S. Justice Department's demand Can anyone be sure that they haven't complied with a National Security Letter(NSL) demanding them to hand over user data? And even if they did comply, we wouldn't know about it because of the terms of a NSL.
So all this talk about Google standing up to protect user data from the US Administration is as true and verifiable as their motto itself ("Don't be evil").
It may be the best news available to you on TV but not 'The Best'. The PBS News Hour always provides two sides to a 'factual story' too, be it 'Global Warming' or 'Latest NIE Report Findings'. It features the uber moron, David Brooks.
'BBC News' is the closest thing to unbiased news that you get on TV in the US. And if you are really lucky, you can watch Amy Goodman(Democracy Now) on Cable/Satellite/Public television at some places.
End up being blamed for everything? Let me put that in perspective. There was no proper planning for partition and transition of power and the British haste culminated in chaos, massacres and the ensuing conflict.
Quoting from the reflections of one Christopher Beaumont, who played a central role in the partition of India in 1947: (http://news.bbc.co.uk/2/hi/south_asia/6926464.stm)
"The viceroy, Mountbatten, must take the blame - though not the sole blame - for the massacres in the Punjab in which between 500,000 to a million men, women and children perished," he writes.
"The handover of power was done too quickly."
Look here for more info (http://en.wikipedia.org/wiki/Partition_of_India#Perspectives )
From what I recall about Peloton(that's what the presenter called it), they wish to have a 14.8 TF/s scalable unit with 4x Infiniband interconnect. This scalable unit itself is more than half the power of Thunder(ranked 7 in Top 500) http://top500.org/lists/plists.php?Y=2005&M=06 They plan to have 16 such scalable units.
For those who are interested in the specs:
Peloton is 16 SU with 236.5 TeraFLOP/s, 215 TiB memory, 5.0 PB global disk system with 6,720 SMPs and 48+24 = 72 IBA 4x DDR sw. Power is 4.05 MW.
The Seaborg system is not the fastest one around. I personally have worked on more powerful systems than that like Tungsten at NCSA.(currently ranked 10th in the top 500)
I've been using several supercomputers for my research project. Most of them are very busy.
Eg. On the IBM P690(Cheetah) at Oakridge National labs,you have to wait for a week to get your 512 processor job scheduled. This is an extremely busy system.
On the other hand,you have systems like the Itaniun cluster at NCSA(National Center for Supercomputing Applications) which schedules your jobs a lot quicker. Actually you can check out the usage of this cluster online at http://tg-monitor.ncsa.teragrid.org/ (don't slashdot it, it is quite useful to a lot of researchers:-) )
Slashdot Mirror
Imagine a Beowulf cluster of moons. Well Jupiter has 66 moons . That's a start.
a conventional CPU-based supercomputer doesn't have the necessary I/O bandwidth to do the work.
I work in HPC and the trend is towards heterogeneous architectures ( CPU+accelerators). Moore's law, power requirements and economics are dictating that trend. It's definitely a stretch to claim that you get better I/O bandwidth with GPUs. Even with PCI Gen 3, the effective bandwidth you get per CPU core is greater than that of an 'equivalent' GPU core.
Please take a look at how things work in Denmark: Video
Wow, did not see that coming.
Like many out there, I'm surviving the recent GNOME "upgrades" by running fallback mode which mimics GNOME 2.x. That's the only means to maintain sanity and a semblance of productivity. Going at this rate, keep an eye for a GNOME branded one-button mouse, because right-click is for pussies.
"take out 3rd world dictators"
Can we please remove this out from our aspirations/goals? I think we already did enough harm to the world pursuing this goal.
http://www.imdb.com/title/tt0083530/ A great documentary indeed!
I agree completely. I just hope that Doctor Who would be more about the Timelord and less about the humans (Amy and Rory can go jump off a cliff for all I care). Too much drama about the two is nauseating.
Couldn't agree more. I'm ecstatic they killed this crappy show. Mission fucking accomplished!
http://sitebar.org/ is there but it doesn't have support for tags last time I checked. I'm in the same boat as you are, guess we just need to develop those capabilities now.
I use Lustre on the petaflop supercomputer at Oak Ridge and mostly satisfied with the I/O performance even for 64k+ cores. Hence my worry about its future direction. Oracle doesn't have any interest in HPC and hence not sure how they will treat Lustre. I would be very happy to admit if my worries are misplaced.
This doesn't bode well for some good hitherto lesser known products from Sun. Personally I'm a bit worried about Lustre.
The login nodes run Linux but the compute nodes run CNL (Compute Node Linux), a lightweight OS designed to reduce system overhead. On an earlier version of this machine (XT3), the compute nodes used to run Catamount, a different lightweight OS.
Nicely done! You summed it up very elegantly.
Short answer: Quoting just one example from Astrophysics from DOE Report
Core collapse supernovae simulations with the spatial resolution required to properly model critical aspects of the explosion dynamics (e.g., the evolution of the stellar core magnetic fields and their role in generating the supernova) will require much higher resolution than today's terascale codes. These codes, in turn, will require exascale computing, particularly if a number of simulations are to be performed across the range of stellar progenitors and input physics. One such simulation is expected to take ~8 weeks, assuming 20% efficiency on an exaflops machine.
Long answer:
I would refer anyone interested to the "Modeling and Simulation at the Exascale for Energy and the Environment Town Hall Meetings Report"
Conclusion of Report:
The broad computational science community has a golden opportunity to accelerate the availability of usable exascale systems. To take full advantage of this opportunity to deliver exascale computing by 2017 will require an integrated program of investments in hardware and software research and development, (R&D). Also required will be a tight coupling to a selected set of science communities and the associated applied mathematics R&D. In some cases, such as astrophysics and climate, the communities are well on the way to exploiting petascale systems. In other cases, such as socioeconomics and multiscale biology, there is great opportunity for acceleration. Computational science and engineering opportunities in energy are wide and deep and have an enormous potential impact on advancing energy technology and fundamental science. If acceleration is to be achieved -- and there is every reason to both desire it and believe that it can be accomplished -- then every minute will count, and even modest investments early in the cycle (e.g., 2008 and 2009) could have dramatic benefit and will reduce uncertainties moving ahead.
Important applications(not an exhaustive list by any means) as gleaned from that Town Hall Meetings Report:
Energy
Energy research offers significant opportunities to exploit computing at the exascale, in order to advance our understanding of basic processes in areas such as combustion, which would naturally lead to a design capability for improving the efficient use of liquid fuels, whether from fossil sources or renewable sources. First-principles computational design and optimization of catalysts will become possible at the exascale, as will de novo design of biologically mediated pathways for energy conversion.
Access to exascale systems and the appropriate applications codes could have a dramatic impact on nuclear fission reactor design and optimization and would help accelerate understanding of key plasma physics phenomena in fusion science critical to getting the most from the U.S. investment in ITER.
Exascale systems should also enable a major paradigm shift in the use of large-scale optimization techniques to search for near-optimal solutions to engineering problems. Many energy and industrial problems are amenable to such an approach, in which many petascale instances of the problem are run simultaneously under the control of a global optimization procedure that can focus the search on parameters that produce an optimal outcome.
Environment
Three broad areas relating to the environment were discussed: climate modeling; integrated energy, economics, and environmental modeling; and multiscale biological modeling from molecules to ecosystems.
Climate modeling.
As the most mature of the three environmental application areas, climate modeling is expected to make good use of exascale systems. The impact of these systems will be threefold.
So all this talk about Google standing up to protect user data from the US Administration is as true and verifiable as their motto itself ("Don't be evil").
It may be the best news available to you on TV but not 'The Best'. The PBS News Hour always provides two sides to a 'factual story' too, be it 'Global Warming' or 'Latest NIE Report Findings'. It features the uber moron, David Brooks. 'BBC News' is the closest thing to unbiased news that you get on TV in the US. And if you are really lucky, you can watch Amy Goodman(Democracy Now) on Cable/Satellite/Public television at some places.
End up being blamed for everything? Let me put that in perspective. There was no proper planning for partition and transition of power and the British haste culminated in chaos, massacres and the ensuing conflict.
Quoting from the reflections of one Christopher Beaumont, who played a central role in the partition of India in 1947: (http://news.bbc.co.uk/2/hi/south_asia/6926464.stm)
"The viceroy, Mountbatten, must take the blame - though not the sole blame - for the massacres in the Punjab in which between 500,000 to a million men, women and children perished," he writes.
"The handover of power was done too quickly."
Look here for more info (http://en.wikipedia.org/wiki/Partition_of_India#Perspectives )
I did the same thing right away after reading Rob's post! The post definitely resonated with my *soul* ;)
Flip flopper. Just kidding :)
P.S. Decade = 10 years (mod me as informative please,please,please)
From what I recall about Peloton(that's what the presenter called it), they wish to have a 14.8 TF/s scalable unit with 4x Infiniband interconnect. This scalable unit itself is more than half the power of Thunder(ranked 7 in Top 500) http://top500.org/lists/plists.php?Y=2005&M=06 They plan to have 16 such scalable units.
For those who are interested in the specs: Peloton is 16 SU with 236.5 TeraFLOP/s, 215 TiB memory, 5.0 PB global disk system with 6,720 SMPs and 48+24 = 72 IBA 4x DDR sw. Power is 4.05 MW.
The Seaborg system is not the fastest one around. I personally have worked on more powerful systems than that like Tungsten at NCSA.(currently ranked 10th in the top 500)
I've been using several supercomputers for my research project. Most of them are very busy. Eg. On the IBM P690(Cheetah) at Oakridge National labs,you have to wait for a week to get your 512 processor job scheduled. This is an extremely busy system. On the other hand,you have systems like the Itaniun cluster at NCSA(National Center for Supercomputing Applications) which schedules your jobs a lot quicker. Actually you can check out the usage of this cluster online at http://tg-monitor.ncsa.teragrid.org/ (don't slashdot it, it is quite useful to a lot of researchers :-) )
I'm from NC State. Here is the link to the original article at North Carolina State University.(including some pictures) http://www.ncsu.edu/news/press_releases/05_06/133. htm
This is the link to the actual news release. (I study at NC State ;-))
http://www.ncsu.edu/news/press_releases/05_03/075. htm