As described by robthebloke - It's a component in some KERS systems but not all.
The basics of KERS are known in the general auto industry as regenerative braking. It's a fairly common thing and is one of the largest benefits of hybrid vehicles. It's why hybrid vehicles are often matched in highway mileage by some traditional vehicles, but they crush traditional vehicles in city mileage (primarily because they don't take that mileage hit from stop-and-go driving, which wastes a lot of energy heating the brakes in traditional vehicles.
Most hybrid systems take the approach of using the energy storage system to permit a lower-power engine to be used in a vehicle without affecting drivability in most situations. This is why hybrid has such bad connotations among gearheads.
The hybrid systems in F1 cars and that Porsche are constructed with a different goal - keep the high-power engine, but augment it with energy storage to improve your lap times on a twisty/turny track that has lots of braking and acceleration. They often call it KERS to avoid the negative connotations of hybrid vehicles with gearheads, even though in reality, it's the exact same approach except with some alterations in the design parameters and goals.
Simple question. The OP asked WHY you feel that is a solution for large-cluster HPC.
It looks like so far your only reason is "i liek it!" - I personally have no opinion or experience with HPC clusters, but so far nearly all of those who do are recommending something that is either RHEL or RHEL-based (Rocks or Scientific Linux), if only because it allows you to leverage commonality with the big cluster operators with installations in the Top500.
Disclaimer: I'm an Ubuntu user, and I greatly enjoy it, but I have not seen many examples of actual scientific clusters running it.
It's toughest to manage in the first couple of hours after reactor shutdown. Had they kept the cooling systems going just a bit longer, there would likely have been significantly less damage.
"Proposals for 'passive' cooling systems involve putting a big tank of water over the plant. If the plant shuts down you let gravity feed the cooling system. If a major incident happens, such as an earthquake or tsunami, it is likely to damage the tank and let all of the water out. What good is a passive system that is subject to the same problems as the plant itself? " 1) So far there is no evidence that the plant (especially safety-critical items) suffered significant quake damage - One of the service pits cracked, but that's not supposed to be safety-critical. (Obviously, since they DID play a part, future plant designs are going to take that into account.) 2) Nothing within the reactor or turbine buildings was damaged by the tsunami - only unprotected (but unfortunately safety-critical) items outside of the reactor and turbine buildings. (This is why an ABWR probably would have survived without problems - they have additional backup generation inside the turbine building) 3) Newer plant designs have even more quake-hardening than the Fukushima reactors
End result - An AP1000 or ESBWR almost surely would have weathered this disaster without core damage. Both eliminate the need for backup generators.
Many of the status reports from early on indicated a partial meltdown. (It was described as "fuel damage" - but that's meltdown).
So how is this news? We already knew the fuel rods had suffered from partial melting/damage. It's almost a given when you see status reports indicating fuel with only partial water coverage.
Not even that many, unless you count psychological and political effects. TMI was an unmitigated devastating disaster if (and only if) you count psychological and political aftermath.
TMI happened more than 30 years ago, in the first decade or two of nuclear power generation. No one outside of the plant boundary was exposed to more radiation than eating a banana a day for a year, and plant design and operating procedure changes were made in response to the accident.
Fukushima was one of the oldest reactors on the planet and it took a disaster that killed 25,000+ people to cause any problems. In the end Fukushima will likely be a drop in the bucket compared to the direct effects of the disaster. That said, we should still strive to make sure other plants can weather such disasters successfully - which we already have. ESBWRs or AP1000s would have shrugged off that tsunami hit without problems, as both designs eliminate the need for backup generators for decay heat removal.
There have been a total of three major accidents in nuclear power generation (weapons production accidents like Kyshtm are irrelevant now in the civilized world, as only rogue states like NK are actively producing weapons).
The first (TMI) did not expose a single human outside of the plant boundary to any more radiation than eating a banana a day for a year. The car analogy made a while ago was a good one - is it right to declare all cars unsafe when a vehicle flips, multiple crumple zones get crumpled, and the airbags deploy, and the driver walks away without scratch?
Chernobyl is irrelevant when talking about non-Soviet nuclear power - only the Soviets built graphite-moderated water-cooled reactors (which are fundamentally dangerous - Google "positive void coefficient" and read up on the Chernobyl timeline please. Chernobyl wasn't an accident, it was an act of criminal negligence. It is clear that you refused to educate yourself even slightly about the history of nuclear power and nuclear safety. While we're on the analogy path - Chernobyl is the nuclear industry's equivalent of taking an old school bus, cutting the brake lines, removing the shocks and swaybars, filling it with kids, drinking a fifth of vodka, and driving down a windy mountain road in a blizzard.
Fukushima, then, is the first time in non-Soviet nuclear power generation that any member of the public has been at risk of exposure. It is a testament to nuclear safety that one of the oldest reactors on the planet (Unit 1 was originally scheduled for decommissioning prior to the earthquake) performed this well in the face of a disaster that killed 25,000+ people outright. Read up on nuclear safety over the past four decades - engineers haven't been sitting on their asses banging rocks together, they have been constantly asking "can we do better", and improving safety. Car analogy time - judging modern nuclear designs based on the performance of Fukushima is like judging modern highway safety based on the safety performance of technology widely deployed in cars in the 1970s. The failure modes seen at Fukushima have been addressed in modern designs like the AP1000 and ESBWR. (Even the ABWR probably would have not had any issues due to the gas turbine they added within the generator building - as far as I can tell, none of the generator buildings suffered significant direct damage from the tsunami.)
Some (actually most) of Uematsu's orchestral pieces are simply stunning.
The problem - Uematsu is taking lesser and lesser roles as time goes on. He basically had no hand in the FF13 soundtrack, and IT SHOWS. Elevator music in one area? (the Whateveritwascalled Massif) - You've got to be kidding me!
And of those three: 1979: No actual measurable public radiation exposure. Some animals did have measurably elevated levels of radioactive substances, but if you drank that milk for a year you'd receive 1/75 the dose you would from eating a banana daily 1986: Not an accident but a dangerous experiment gone wrong on a fundamentally unstable reactor design with no containment provisions whatsoever. Try to build an RBMK near me and I'll fight it tooth and nail. 2011: Required a disaster that outright killed 25,000+ people in order to trigger problems
The problem is that any attempt to build new modernized nuclear plants results in massive opposition.
And result is that the next most viable solution (service life extensions to old plants) is chosen.
From best to worst in terms of currently viable baseload generation (wind and solar are not currently viable for baseload, at best they're good for 10-15% penetration, the country with the highest wind/solar penetration in the world is the Netherlands at around 20% and that would not be viable if not for neighbors with lower wind power penetration levels. I'm fairly certain when their wind plants quit they buy electricity from France, which generates 70-80% of their electricity from nuclear power.) Modernized nuclear Old-school nuclear Hydroelectric (our resources here are tapped out, and hydro has killed far more people than nuclear in history) Coal (fundamentally dirty) Natural gas (clean-burning, but the process for extracting it from the ground has led to massive groundwater contaminations in many areas where it is being extracted)
As others have said, that site supposedly includes Chernobyl in its estimate, which puts nuclear in an even better light, considering that Chernobyl was not in any way representative of non-Soviet nuclear power in terms of plant design and operation.
Chernobyl's plant had two critical design differences from ANY plant in the United States: 1) Graphite-moderated water-cooled reactor which meant it had a highly positive void coefficient. This is fundamentally dangerous and unstable, which is why such plants have never been built in the US and as I understand it, the NRC will never approve such a plant. 2) No containment structure whatsoever. This meant a steam explosion of the reactor core exposed superheated graphite to outside air.
It also had a critical operational difference: 1) The reactor operators overrode multiple automatic shutdowns in order to continue a dangerous experiment. (The shift supervisor was a good Party man.)
Chernobyl was the nuclear industry's equivalent of taking a school bus, severing the brake lines, removing the shocks, filling it full of kids, drinking a fifth of vodka, and then getting in the driver's wheel to go down a windy mountain road in the middle of a snowstorm.
It's clearly worse than coal, a typical coal plant releases more radiation into the air due to trace amounts of uranium in their coal in one year than the entire lifetime of Three Mile Island. Some coal plant fly ash has such high uranium content that the Chinese are starting to mine it for nuclear plant fuel. It's clearly worse than gas - hydrofracturing operations in the past 5-10 years have sickened more people in the United States than the entire history of nuclear power in this country. It's clearly worse than hydroelectric - see Banqiao Dam, which alone killed 4-5 times as many people as Chernobyl immediately, and significantly more long-term due to famine and disease. Numerous other dam failures have matched Chernobyl's estimated long-term death count (in terms of number of cancer cases) The only "better" options in terms of safety are solar and wind - but the question is, once you take into account the energy storage requirements needed to achieve more than 10-20% penetration for solar/wind, will those massive battery banks full of toxic chemicals necessarily be safer?
You're wrong there - had the backup generators been at the top of the hill or possibly merely installed with snorkels, it would have been fine.
Had the reactors been ABWRs with a backup gas turbine inside the big concrete turbine building in addition to the diesel generators, it probably would have been fine. None of the buildings seem to have sustained any significant damage from the tsunami.
Had the reactors been ESBWRs (close to but not yet approved by the NRC), it would have been fine. ESBWRs don't need backup generators for decay heat removal. They don't need ANYTHING for the first 72 hours after a SCRAM, and the only thing they need beyond that is a fire truck to refill the ICCS pools. Probably once they're refilled you have longer since decay heat generation is constantly reducing.
Not true. Modern designs achieve all three. See for example GE's ESBWR design - cheaper, significantly safer, and more efficient than ABWRs, which were safer and more efficient (not sure about cheaper) than first-gen BWRs.
Unfortunately, Fukushima's units are first-gen BWRs, and in fact were some of the oldest operating reactors in the world.
Actually, while it was not a contributor to the root cause of the reactor failure (It can't be considered an "accident" - it was more of an act of willfill criminal negligence where the shift supervisor was a good Party man and insisted that a dangerous experiment go forward even though the operators reporting to him were recommending a shutdown), cost cutting WAS a significant contributor to the final severity of the Chernobyl. The Soviets decided that containment structures were too expensive and thus Chernobyl had none.
Some safety features just can't be retrofit into a plant, they must be a fundamental part of a plant's design.
The anti-nuclear lobby fights construction of new plants tooth and nail without proposing any viable alternatives. End result is the next most viable alternative (service life extensions and retrofitting what you can to old plants) is what we get.
Modern? How can you call one of the oldest reactors in the world, which was originally scheduled for end-of-life decommissioning prior to the earthquake, modern?
Calling Fukushima Unit 1 (or even any of the other reactors at the site, which were newer but still very old) "modern" just eliminates any credibility you have and shows your complete and total ignorance regarding nuclear safety and the improvements in nuclear safety made in the past 40 years.
Browns Ferry is also NOT a modern plant - its reactors are about as old as those at Fukushima, but at least they're not in a tsunami risk zone, and as I understand it US-based reactors have all been retrofitted with hydrogen control systems that would have prevented the hydrogen explosions that made Fukushima so complex. Also, while it got a "red" incident based on failure of a significant control valve, there are backup cooling loops. (Note that the valve in question was in the decay heat removal system coolant loop. Said system functioned as needed a week or so ago when all three Browns Ferry reactors SCRAMed due to a nearby tornado.)
Yup. One of the barriers to postprocessing is local clock inaccuracy - so having a local atomic clock would be great for survey-grade GPS units.
And as you stated - if receiver clock offset is 0, then you don't need to solve for it, and can get 3D position with 3 sats instead of 4. The actual effect of an inaccurate clock on the error is harder to determine - I have a feeling that with a reasonable quality local crystal oscillator (good enough not to cause cycle slips in the measured carrier phase, etc.) it's insignificant compared to ionospheric error and RF noise in the pseudoranges, along with multipath. The new L2C civilian signal will help some of these issues.
A highly accurate local clock might also make dead reckoning in a blockage situation (urban canyons, tunnels, etc) and signal reacquisition after blockage goes away faster.
Slashdot Mirror
As described by robthebloke - It's a component in some KERS systems but not all.
The basics of KERS are known in the general auto industry as regenerative braking. It's a fairly common thing and is one of the largest benefits of hybrid vehicles. It's why hybrid vehicles are often matched in highway mileage by some traditional vehicles, but they crush traditional vehicles in city mileage (primarily because they don't take that mileage hit from stop-and-go driving, which wastes a lot of energy heating the brakes in traditional vehicles.
Most hybrid systems take the approach of using the energy storage system to permit a lower-power engine to be used in a vehicle without affecting drivability in most situations. This is why hybrid has such bad connotations among gearheads.
The hybrid systems in F1 cars and that Porsche are constructed with a different goal - keep the high-power engine, but augment it with energy storage to improve your lap times on a twisty/turny track that has lots of braking and acceleration. They often call it KERS to avoid the negative connotations of hybrid vehicles with gearheads, even though in reality, it's the exact same approach except with some alterations in the design parameters and goals.
Simple question. The OP asked WHY you feel that is a solution for large-cluster HPC.
It looks like so far your only reason is "i liek it!" - I personally have no opinion or experience with HPC clusters, but so far nearly all of those who do are recommending something that is either RHEL or RHEL-based (Rocks or Scientific Linux), if only because it allows you to leverage commonality with the big cluster operators with installations in the Top500.
Disclaimer: I'm an Ubuntu user, and I greatly enjoy it, but I have not seen many examples of actual scientific clusters running it.
ESBWR is close - the core stays in the same place, but there are heatpipes going to large cooling pools at the roof of the building.
Worst-case, you need a fire truck at the 72 hour mark.
Google "decay heat".
It's toughest to manage in the first couple of hours after reactor shutdown. Had they kept the cooling systems going just a bit longer, there would likely have been significantly less damage.
"Proposals for 'passive' cooling systems involve putting a big tank of water over the plant. If the plant shuts down you let gravity feed the cooling system. If a major incident happens, such as an earthquake or tsunami, it is likely to damage the tank and let all of the water out. What good is a passive system that is subject to the same problems as the plant itself? "
1) So far there is no evidence that the plant (especially safety-critical items) suffered significant quake damage - One of the service pits cracked, but that's not supposed to be safety-critical. (Obviously, since they DID play a part, future plant designs are going to take that into account.)
2) Nothing within the reactor or turbine buildings was damaged by the tsunami - only unprotected (but unfortunately safety-critical) items outside of the reactor and turbine buildings. (This is why an ABWR probably would have survived without problems - they have additional backup generation inside the turbine building)
3) Newer plant designs have even more quake-hardening than the Fukushima reactors
End result - An AP1000 or ESBWR almost surely would have weathered this disaster without core damage. Both eliminate the need for backup generators.
Many of the status reports from early on indicated a partial meltdown. (It was described as "fuel damage" - but that's meltdown).
So how is this news? We already knew the fuel rods had suffered from partial melting/damage. It's almost a given when you see status reports indicating fuel with only partial water coverage.
Yes, they're nearly all making small tweaks for their specific hardware.
Not even that many, unless you count psychological and political effects. TMI was an unmitigated devastating disaster if (and only if) you count psychological and political aftermath.
TMI happened more than 30 years ago, in the first decade or two of nuclear power generation. No one outside of the plant boundary was exposed to more radiation than eating a banana a day for a year, and plant design and operating procedure changes were made in response to the accident.
Fukushima was one of the oldest reactors on the planet and it took a disaster that killed 25,000+ people to cause any problems. In the end Fukushima will likely be a drop in the bucket compared to the direct effects of the disaster. That said, we should still strive to make sure other plants can weather such disasters successfully - which we already have. ESBWRs or AP1000s would have shrugged off that tsunami hit without problems, as both designs eliminate the need for backup generators for decay heat removal.
There have been a total of three major accidents in nuclear power generation (weapons production accidents like Kyshtm are irrelevant now in the civilized world, as only rogue states like NK are actively producing weapons).
The first (TMI) did not expose a single human outside of the plant boundary to any more radiation than eating a banana a day for a year. The car analogy made a while ago was a good one - is it right to declare all cars unsafe when a vehicle flips, multiple crumple zones get crumpled, and the airbags deploy, and the driver walks away without scratch?
Chernobyl is irrelevant when talking about non-Soviet nuclear power - only the Soviets built graphite-moderated water-cooled reactors (which are fundamentally dangerous - Google "positive void coefficient" and read up on the Chernobyl timeline please. Chernobyl wasn't an accident, it was an act of criminal negligence. It is clear that you refused to educate yourself even slightly about the history of nuclear power and nuclear safety. While we're on the analogy path - Chernobyl is the nuclear industry's equivalent of taking an old school bus, cutting the brake lines, removing the shocks and swaybars, filling it with kids, drinking a fifth of vodka, and driving down a windy mountain road in a blizzard.
Fukushima, then, is the first time in non-Soviet nuclear power generation that any member of the public has been at risk of exposure. It is a testament to nuclear safety that one of the oldest reactors on the planet (Unit 1 was originally scheduled for decommissioning prior to the earthquake) performed this well in the face of a disaster that killed 25,000+ people outright. Read up on nuclear safety over the past four decades - engineers haven't been sitting on their asses banging rocks together, they have been constantly asking "can we do better", and improving safety. Car analogy time - judging modern nuclear designs based on the performance of Fukushima is like judging modern highway safety based on the safety performance of technology widely deployed in cars in the 1970s. The failure modes seen at Fukushima have been addressed in modern designs like the AP1000 and ESBWR. (Even the ABWR probably would have not had any issues due to the gas turbine they added within the generator building - as far as I can tell, none of the generator buildings suffered significant direct damage from the tsunami.)
Some (actually most) of Uematsu's orchestral pieces are simply stunning.
The problem - Uematsu is taking lesser and lesser roles as time goes on. He basically had no hand in the FF13 soundtrack, and IT SHOWS. Elevator music in one area? (the Whateveritwascalled Massif) - You've got to be kidding me!
I felt that FF12 was pretty good - you might want to give that a try.
FF13 was a massive disappointment after 12.
And of those three:
1979: No actual measurable public radiation exposure. Some animals did have measurably elevated levels of radioactive substances, but if you drank that milk for a year you'd receive 1/75 the dose you would from eating a banana daily
1986: Not an accident but a dangerous experiment gone wrong on a fundamentally unstable reactor design with no containment provisions whatsoever. Try to build an RBMK near me and I'll fight it tooth and nail.
2011: Required a disaster that outright killed 25,000+ people in order to trigger problems
As I understand it, most ISPs don't have (and for various reasons) don't want common carrier status as it already stands.
It may be a generic unintentional routing failure. Proxies cause a different route to be used.
It's not like TPB is perfect - it was down for me for 2-3 days straight last a week or two ago (ISP is Time Warner RoadRunner) but has since returned
Exactly because it is licensed under the Apache Software License v2 which allows this sort of thing.
The problem is that any attempt to build new modernized nuclear plants results in massive opposition.
And result is that the next most viable solution (service life extensions to old plants) is chosen.
From best to worst in terms of currently viable baseload generation (wind and solar are not currently viable for baseload, at best they're good for 10-15% penetration, the country with the highest wind/solar penetration in the world is the Netherlands at around 20% and that would not be viable if not for neighbors with lower wind power penetration levels. I'm fairly certain when their wind plants quit they buy electricity from France, which generates 70-80% of their electricity from nuclear power.)
Modernized nuclear
Old-school nuclear
Hydroelectric (our resources here are tapped out, and hydro has killed far more people than nuclear in history)
Coal (fundamentally dirty)
Natural gas (clean-burning, but the process for extracting it from the ground has led to massive groundwater contaminations in many areas where it is being extracted)
As others have said, that site supposedly includes Chernobyl in its estimate, which puts nuclear in an even better light, considering that Chernobyl was not in any way representative of non-Soviet nuclear power in terms of plant design and operation.
Chernobyl's plant had two critical design differences from ANY plant in the United States:
1) Graphite-moderated water-cooled reactor which meant it had a highly positive void coefficient. This is fundamentally dangerous and unstable, which is why such plants have never been built in the US and as I understand it, the NRC will never approve such a plant.
2) No containment structure whatsoever. This meant a steam explosion of the reactor core exposed superheated graphite to outside air.
It also had a critical operational difference:
1) The reactor operators overrode multiple automatic shutdowns in order to continue a dangerous experiment. (The shift supervisor was a good Party man.)
Chernobyl was the nuclear industry's equivalent of taking a school bus, severing the brake lines, removing the shocks, filling it full of kids, drinking a fifth of vodka, and then getting in the driver's wheel to go down a windy mountain road in the middle of a snowstorm.
It's clearly not the second worst we have.
It's clearly worse than coal, a typical coal plant releases more radiation into the air due to trace amounts of uranium in their coal in one year than the entire lifetime of Three Mile Island. Some coal plant fly ash has such high uranium content that the Chinese are starting to mine it for nuclear plant fuel.
It's clearly worse than gas - hydrofracturing operations in the past 5-10 years have sickened more people in the United States than the entire history of nuclear power in this country.
It's clearly worse than hydroelectric - see Banqiao Dam, which alone killed 4-5 times as many people as Chernobyl immediately, and significantly more long-term due to famine and disease. Numerous other dam failures have matched Chernobyl's estimated long-term death count (in terms of number of cancer cases)
The only "better" options in terms of safety are solar and wind - but the question is, once you take into account the energy storage requirements needed to achieve more than 10-20% penetration for solar/wind, will those massive battery banks full of toxic chemicals necessarily be safer?
You're wrong there - had the backup generators been at the top of the hill or possibly merely installed with snorkels, it would have been fine.
Had the reactors been ABWRs with a backup gas turbine inside the big concrete turbine building in addition to the diesel generators, it probably would have been fine. None of the buildings seem to have sustained any significant damage from the tsunami.
Had the reactors been ESBWRs (close to but not yet approved by the NRC), it would have been fine. ESBWRs don't need backup generators for decay heat removal. They don't need ANYTHING for the first 72 hours after a SCRAM, and the only thing they need beyond that is a fire truck to refill the ICCS pools. Probably once they're refilled you have longer since decay heat generation is constantly reducing.
Not true. Modern designs achieve all three. See for example GE's ESBWR design - cheaper, significantly safer, and more efficient than ABWRs, which were safer and more efficient (not sure about cheaper) than first-gen BWRs.
Unfortunately, Fukushima's units are first-gen BWRs, and in fact were some of the oldest operating reactors in the world.
Actually, while it was not a contributor to the root cause of the reactor failure (It can't be considered an "accident" - it was more of an act of willfill criminal negligence where the shift supervisor was a good Party man and insisted that a dangerous experiment go forward even though the operators reporting to him were recommending a shutdown), cost cutting WAS a significant contributor to the final severity of the Chernobyl. The Soviets decided that containment structures were too expensive and thus Chernobyl had none.
Some safety features just can't be retrofit into a plant, they must be a fundamental part of a plant's design.
The anti-nuclear lobby fights construction of new plants tooth and nail without proposing any viable alternatives. End result is the next most viable alternative (service life extensions and retrofitting what you can to old plants) is what we get.
Because they fought the construction of new reactors tooth and nail without proposing any viable alternatives.
End result: Next most viable alternative (service life extension of old reactors) gets chosen.
Modern? How can you call one of the oldest reactors in the world, which was originally scheduled for end-of-life decommissioning prior to the earthquake, modern?
Calling Fukushima Unit 1 (or even any of the other reactors at the site, which were newer but still very old) "modern" just eliminates any credibility you have and shows your complete and total ignorance regarding nuclear safety and the improvements in nuclear safety made in the past 40 years.
Browns Ferry is also NOT a modern plant - its reactors are about as old as those at Fukushima, but at least they're not in a tsunami risk zone, and as I understand it US-based reactors have all been retrofitted with hydrogen control systems that would have prevented the hydrogen explosions that made Fukushima so complex. Also, while it got a "red" incident based on failure of a significant control valve, there are backup cooling loops. (Note that the valve in question was in the decay heat removal system coolant loop. Said system functioned as needed a week or so ago when all three Browns Ferry reactors SCRAMed due to a nearby tornado.)
Yup. One of the barriers to postprocessing is local clock inaccuracy - so having a local atomic clock would be great for survey-grade GPS units.
And as you stated - if receiver clock offset is 0, then you don't need to solve for it, and can get 3D position with 3 sats instead of 4. The actual effect of an inaccurate clock on the error is harder to determine - I have a feeling that with a reasonable quality local crystal oscillator (good enough not to cause cycle slips in the measured carrier phase, etc.) it's insignificant compared to ionospheric error and RF noise in the pseudoranges, along with multipath. The new L2C civilian signal will help some of these issues.
A highly accurate local clock might also make dead reckoning in a blockage situation (urban canyons, tunnels, etc) and signal reacquisition after blockage goes away faster.