Oh, right, they can't tell you're trying to open an https connection instead of an http connection because they're hijacking the DNS query, not the browser query. That's why it's called *broken*.
And where do they put the opt-out button on ssh connections? Unlike email, where I'm usually not emailing to a www.* address, I fairly often want to ssh to a web server (admittedly, that's usually inside my own network, but not always), and they shouldn't be fraking with it - and they can't tell whether they are or not.
If you've watched technology-related law for a few years, you'd see lots of laws or bureaucratic regulations proposed by specific interested parties trying to get an advantage over their competition, but you'd also see an appalling number of rules or laws that were written simply because they seemed like a good idea at the time, and the details were borrowed from other laws or rules (which were also probably not well-written and don't apply directly to the current case, but share some buzzwords.)
In this case, I think somebody probably complained that phishers were imitating legitimate investment sites and scamming people (a legitimate problem), and the Congresscritter had his staff grab some spare legal code that seemed to be in the right space, and no, he not only didn't really understand the technology, and no, he didn't understand the *legal* environment surrounding that field of regulation that's evolved over the last couple of decades, but hey, there was a problem and he was Fixing It.
Colleges can sometimes be difficult to deal with, and a first glance at Google Maps suggests that there may also be other city-related facilities that are keeping you from getting a straight route to the CO. I don't know quite where you live, but I tried looking up several addresses near the university on the uverse.att.com page, and didn't find service; it may be that I was looking at commercial streets and not residential ones, or they may not have service set up yet (e.g. it may be some other kind of box?) Or it may be that uverse is available there but the call-center person you got didn't realize that a customer asking for "dsl" is also interested in "u-verse internet" and not just "the older DSL product" or whatever. Good luck!
It's unfortunate that SORBS deserved the bad rap they got, but they were quite effective at false-positive mis-detection of legitimate emails as spam, unlike lists such as Spamhaus, and quite unwilling to remove addresses that had been tagged as spam or as part of variously large blocks of ISPs that had had some spammers use them.
There are legitimate uses for a rabid-overkill list, such as directing mail from those IP addresses to a check-more-carefully server if you've got a multi-tier multi-server spam-blocking environment anyway, or perhaps adding a bit of weight to your Spamassassin weights, but they weren't a list you could depend on if you wanted to actually receive all your non-spam mail.
Remember that cable distance is often much different than walking distance - did you have to cross a railroad track or larger road? Sometimes there are limitations on where you can cross, and sometimes it's just a cable-layout thing. I'm about 7500 feet driving distance from my CO, but about 11-12000 cable feet, and I get 3 Mbps pretty reliably. YMMV, as they say.
If they're putting an AT&T U-verse(sm) box in front of your building, once the service is available, you should be able to get much faster DSL performance over that. I don't know if they're currently making it cost-effective to order just data without also getting TV service or not. The bandwidth available depends on your distance from the ugly box, typically about 25Mbps of which they burn ~15 for TV, and the Wikipedia article says they'll package Internet at up to 18 Mbps.
It was a better story when SARS and the Bird Flu were going to Kill Us ALL! ! and when the Bush Administration was telling us that there were Terrorists Under Our Beds, and when there were businesses and government agencies that weren't set up to handle large fractions of their users telecommuting. But enough of those businesses were out there that there's an entrenched consulting market for Pro-Active Pandemic Preparation, and it's flu season and there aren't enough flu shots, so We're All Gonna Die! yet again.
You'll keep seeing that story until the Obama Administration either gets off its ass and stops being the Leftover-Bush Administration, or until it redeploys those people into some other politically useful job, like moving them from the Department of Homeland Security to the Department of Health and Human Services were they can tell us that If We Don't Get National HealthCare, We're All Gonna Die.
Yeah, yeah, you can bitch about the price, and bitch about carriers that think bandwidth-capping is a cool idea they should adopt, and if you've only got regular DSL instead of cable modem or fiber or VHDSL, then you can bitch about how you're only getting a megabit or two and can't watch HDTV in real time over it, but that's not where the problems are here. The problems are the connections businesses (and government agencies) have at their ends, which affect whether they can handle having 99% of their employees telecommuting. If you've got a culture that thinks F2F meetings are the only way to get things done, and wants 25Mbps Telepresence(tm) at everybody's house, then maybe there's a network problem, or at least wants Skype Video and $29 webcams, but basically it's a question of whether the business has enough bandwidth at their headquarters/datacenters to handle their workers, and whether the business-oriented ISPs have resources deployed to handle them. (For the most part, they do, and if they don't, then they'll upgrade if they start seeing actual orders from pandemic-panicked businesses.)
The problem they're trying to address (well, other than organizational self-preservation by looking useful, and helping out their friends and accomplices in the drama-creation business) isn't that lots of sick people will be staying home. It's that if the Pandemic gets bad enough, schools and businesses will tell healthy people to stay home and work from home to avoid getting sick, and maybe even the government will tell people to stay home. They're concerned about the load that those people will put on the network.
Obpolitics: Of course it's "propaganda, disinfo, lies, and bullshit". That's what the DHS was created to *do*. And just because the Bush Administration set it up to do a better job of scaring the children than the Clinton Administration or anybody since the Cuban Missile Crisis and Joe McCarthy had done, and just because Obama was elected to "Not Be Bush", that doesn't mean that the Obama Administration is ready to turn around and become pro-freedom or anti-scary any time soon.
ObTechnoPolitics: And ever since the SARS and Bird Flu scares didn't turn into pandemics, there's been a whole industry of computing/telecom/business consultants trying to get people to buy pandemic-planning consulting services, and they'd be halfway out of a job if the DHS stopped scaring people to drum up business. And if the EFF/Cypherpunks/Banks/HiTech people hadn't spent the 90s arguing that encryption was critical for business as well as civil liberties, you wouldn't be using VPNs to commute to work, and those consultants would still be selling you Analog Modem Pools and ISDN (bwah-hah-ha!)
Having said that, though, there are some serious problems with organizations that aren't prepared to have lots of their employees telecommuting. In some cases (including government agencies as well as financial and other businesses) their employees don't have work laptops, or don't have their home computers set up to run VPNs, or don't have big enough pipes at the office to handle 99% of the people working from home instead of 5-10%. Those organizations need to be convinced to upgrade their infrastructure. And there are probably some lame ISPs that still try to block VPNs except for people who pay extra for "business" broadband packages, though there's a lot less of that these days, and telling them to Just Stop That is a good thing. (And of course there are businesses whose employees actually do Real Physical Stuff all day instead of working on computers, but even most of those people use computers to keep track of the Stuff, or have managers or clerks who do that for them.)
From a technology standpoint, most ISPs these days can support IPv4 prioritization using either TOS or DSCP, though most ISPs don't support it across peering links, and some of them charge different amounts for it. There never was a really clear business plan for it. The Network Neutrality debate didn't help any of that - it was focused on some bonehead telecom execs wanting to charge more money for Google and YouTube (even before they merged), but from my perspective I'd really like it if my ISP prioritized my downlink to prefer UDP and VPN traffic over web and YouTube, and especially if they prioritized everything-else over BitTorrent, so I could be talking on VOIP while browsing the web and downloading totally-legal music. But even without prioritization, if you've got enough bandwidth on the home-user end to run YouTube, there's enough bandwidth for people to do real work (at least if they can get their kids to stop watching YouTube), and the business ends of the connections really do load-balance fairly well.
Yeah, sure, the current Mayans probably don't use the old hieroglyphics. But even so, having a grammar and vocabulary for the language so you know how things fit together helps immensely, compared to having to infer them from the hieroglyphs. Still, that Spanish guy shouldn't have burned the books.
A name like Sparc wasn't a typo by somebody trying to spell Spark. It was a deliberate marketing decision to create a name that sounded cool and wasn't spelled like the generic word because that probably couldn't be trademarked without conflicting with lots of other existing electrical and electronic trademarks. SparkFun is also a name that includes sparkiness in it, and is differentiated from the generic English word in different ways.
Besides, doesn't somebody already have a business method patent on suing everybody who's got a name even vaguely similar to yours when you don't have a legitimate case? SPARC Inc could get themselves in a lot of trouble for violating it.
And if anything, the Girl Genius folks could now sue SI for trying to get into the Spark business when they're clearly not that bright... On the other hand, the SparkFun people make lots of cool little gadgets and generally support Mad Science, so they're just fine.
Disclaimer: The Sun-2 in my attic is not an Ultrasparc-2. Nor is it a Sparcstation-2. Some of us have been in the Sun business for a long time.
I'm not saying that just for disinformation, though I do think that Weapons of Mass Destruction are Bad Things - I've worked there on data network consulting applications. When I said not to confuse tool-building with pork-barrel, that was a response to somebody else who said that spending $32M on a cloud-computer was pork-barrel.
While there are a few people at DOE who do graphics or computing just because it's cool, or because almost nobody else who does high-security computing research does it on supercomputers, they mostly view computing as a tool to do science and/or engineering. They're trying to model nuclear explosions, and it takes a lot of computing horsepower to do that.
And the "Imagine What You Could Do With A Cray" poster on their walls was cool, but when there's only one broadband cable system to get data across campus, back when 10Mbps Ethernet was still fast, and a few thousand people imagining what they can do with that Cray, networking problems get really hairy.
But yeah, apparently somebody else had made the mushroom cloud joke a bit earlier than I had. Didn't say anything useful about their computing environment, though.
Yes, the author does rudely smack the original poster in the face, but as Captain Jack Sparrow said, he "may have deserved that."
The original poster didn't give us enough info. Aggregation from multiple ISPs is possible, but it's a lot dodgier performance-wise than aggregating multiple connections from the same ISP. On the other hand, your choices of possible solutions depend a lot on your problem - if you want to make a single fat TCP session go faster, for instance a big file transfer, that's a lot harder than load-balancing a bunch of smaller sessions. And of course most cheap consumer solutions are very asymmetric, so the upstream will be your limitation, and don't give you good Layer 2 feedback and probably aren't running TCP ECN either.
There are problems which really need high memory bandwidth and don't fit on smaller-than-super computers, so a time slice on a supercomputer can be worth far more than full-time access to dedicated fast conventional computers. But those problems become less and less common as regular computers get bigger and faster - your laptop probably has a graphics processor that's faster than a Cray-2 by now...
SETI works on what gets described in the trade as "embarrassingly parallel" problems - supercomputers deal with stuff that's harder to get good parallel speedup without throwing fancy hardware at it. DOE problems are often somewhere in between, and unfortunately the boinc/seti/screensaver approach to ad-hoc supercomputing isn't always good for applications like LINPAK, so it's hard to compare the real computing power. However, if you ignore that (:-), most of the top computers in the Top500 list are doing nuclear or military work, and some are for weather bureaus or Earth simulation, but about half of the last decade, the SETI people are volunteering 2-10x as much CPU just to look for little green men as the largest military supercomputers were providing. The supercomputer guys are back on top, and I don't know that we'll catch up with them again, but on the other hand some of them are now doing genetics and other Good Things instead of Evil.
This is the DOE. They mainly work on nukes, and if they're experimenting with cloud computing, it's not because they're trying to do commercial or academic work into cloud computing, it's because they're trying to find powerful and cost-effective ways to get the computing horsepower they need to work on more nukes, just as when they work with supercomputers, it's not just because they're "Imagining what they can do with a Cray", it's because they're trying to solve problems that are really that big.
So, like, back off man, they're scientists!
Or don't back off,. because they're developing weapons as well as civilian power applications, but don't confuse tool-building with pork barrel.
Back when 1GB was enough to hold Wikipedia, there was application for iPods with screens that could run Wikipedia. This sounds like basically an evolution of that, and of course today the iPods/iPhones have a lot more storage (and Wikipedia's a bit larger - not sure how many GB you need for 3M articles in English and.5-1M in each of another dozen languages? probably around 20GB before compression, and good compressibility?)
There are three reasons you run raid, other than because it's cookl
Goes faster
Protects you from losing all your data if a whole drive fails
Protects you from losing bits of your data if bits of a drive are bad
SSDs are about speed, but don't have high capacity unless you've got an infinite budget. That's not the problem here. The problems are that
if a whole drive fails and you need to rebuild it, it's going to take a long time which you may not be able to afford, and
If a whole drive fails and you need to rebuild it, you've no longer got parity information, so if there are bad bytes on your disks or bad bits in disk transmission, you can't recover those bits.
SSDs don't really help you on time, because you can't afford to run your whole shop on SSDs if it's big enough that RAID's not reliable. The recovery problem is helped somewhat by disk block error-correcting codes: you can detect if a block has corruption and rewrite/remap it when you're first storing it, so the ratio of bad blocks on disk isn't as bad as you'd expect from raw error rates, and you may need to continually recheck your raid block and rebuild bad blocks proactively rather than finding them after failures.
The big problem with RAID on bigger disks is that you're splitting data across N+1 disks, so if one disk fails catastrophically and needs to be rebuilt, you need to have good data on the remaining N disks, and as the disks become larger, the chances that all that data is good are no longer near-100%. So you're likely to get some bad data even with RAID, though you're not going to have as much bad data as you'd have without RAID, and you're still avoiding the problem that if you disk fails, you lose everything; you're just not guaranteed to lose nothing.
I pulled into a parking lot in Milpitas to make a phone call and use my computer. I didn't need to be online for the call, just look at stuff, but I was pleased to see that there was a wireless signal, they've got tons of free access points all over Milpitas, and the signal was pretty good., It wasn't foolproof - they have a login-timeout browser window thingy, and connecting to my company VPN meant killed its connection so it cut me off after about 5 minutes, but that was enough to download any new email, and I could log back in without the VPN and see the web and my home email just fine.
You've accused all the physicists in the world of crackpottery, and while perhaps you spent some time writing your essay, it's not "a well-formed argument"; it's closer to "not even clear enough to be wrong". It does in fact speak for itself, which is why I recommended the "-1 Crackpot" rating for your earlier posting, and I'm at least as identified around here as you are, having been online since ~1981...
You appear to be confused about what an "ad hominem" attack is. If you remember your junior high school Latin or logic classes, it's arguing "to the man" instead of "to the topic". If I had said "Oh, that's just Louis Savain, so anything he says is bogus", or "He's a New Ager, so anything he says is bogus", that would be "ad hominem".
That's not what's happening here - I went and read your web page, looked at a bunch of your arguments, even got as far as quoting your own words in my posting. I'm not saying people should discount your fine technical work simply because you're a crackpot - I'm saying that your arguments display a hopelessly cracked misunderstanding of physics, and that they shouldn't waste their time reading it except for entertainment purposes, but that you appear to actually believe them so you don't deserve a "-1 Troll" down-rating. (And as far as your posting's first two sentences go, I'm also highly skeptical about whether space-based power makes sense or will work, but the rest of your article and your website makes it clear that you think so for hopelessly wrong reasons.) If you want to claim that I'm insulting you because of your work, and that that's rude, well perhaps it's a fair cop (:-), but it's just the opposite of insulting your work because of you, which is what an ad hominem argument is, and given that you start many of your paragraphs describing how you think normal physicists think the work works as "The Crackpottery", well, you're not in much position to argue - you not only get Newtonian physics wrong, you also get quantum wrong and don't even get Yin and Yang right.
And yes, there are many Bill Stewarts out there - feel free to Google me; I'm the one who's not the drummer and wasn't killed in Nicaragua. And I've been out of town for a couple of days so I didn't see your reply until today.
Slashdot Mirror
Oh, right, they can't tell you're trying to open an https connection instead of an http connection because they're hijacking the DNS query, not the browser query. That's why it's called *broken*.
And where do they put the opt-out button on ssh connections? Unlike email, where I'm usually not emailing to a www.* address, I fairly often want to ssh to a web server (admittedly, that's usually inside my own network, but not always), and they shouldn't be fraking with it - and they can't tell whether they are or not.
If you've watched technology-related law for a few years, you'd see lots of laws or bureaucratic regulations proposed by specific interested parties trying to get an advantage over their competition, but you'd also see an appalling number of rules or laws that were written simply because they seemed like a good idea at the time, and the details were borrowed from other laws or rules (which were also probably not well-written and don't apply directly to the current case, but share some buzzwords.)
In this case, I think somebody probably complained that phishers were imitating legitimate investment sites and scamming people (a legitimate problem), and the Congresscritter had his staff grab some spare legal code that seemed to be in the right space, and no, he not only didn't really understand the technology, and no, he didn't understand the *legal* environment surrounding that field of regulation that's evolved over the last couple of decades, but hey, there was a problem and he was Fixing It.
Colleges can sometimes be difficult to deal with, and a first glance at Google Maps suggests that there may also be other city-related facilities that are keeping you from getting a straight route to the CO. I don't know quite where you live, but I tried looking up several addresses near the university on the uverse.att.com page, and didn't find service; it may be that I was looking at commercial streets and not residential ones, or they may not have service set up yet (e.g. it may be some other kind of box?) Or it may be that uverse is available there but the call-center person you got didn't realize that a customer asking for "dsl" is also interested in "u-verse internet" and not just "the older DSL product" or whatever. Good luck!
Spam? Flamers? Email?
It's unfortunate that SORBS deserved the bad rap they got, but they were quite effective at false-positive mis-detection of legitimate emails as spam, unlike lists such as Spamhaus, and quite unwilling to remove addresses that had been tagged as spam or as part of variously large blocks of ISPs that had had some spammers use them.
There are legitimate uses for a rabid-overkill list, such as directing mail from those IP addresses to a check-more-carefully server if you've got a multi-tier multi-server spam-blocking environment anyway, or perhaps adding a bit of weight to your Spamassassin weights, but they weren't a list you could depend on if you wanted to actually receive all your non-spam mail.
Remember that cable distance is often much different than walking distance - did you have to cross a railroad track or larger road? Sometimes there are limitations on where you can cross, and sometimes it's just a cable-layout thing. I'm about 7500 feet driving distance from my CO, but about 11-12000 cable feet, and I get 3 Mbps pretty reliably. YMMV, as they say.
If they're putting an AT&T U-verse(sm) box in front of your building, once the service is available, you should be able to get much faster DSL performance over that. I don't know if they're currently making it cost-effective to order just data without also getting TV service or not. The bandwidth available depends on your distance from the ugly box, typically about 25Mbps of which they burn ~15 for TV, and the Wikipedia article says they'll package Internet at up to 18 Mbps.
It was a better story when SARS and the Bird Flu were going to Kill Us ALL! ! and when the Bush Administration was telling us that there were Terrorists Under Our Beds, and when there were businesses and government agencies that weren't set up to handle large fractions of their users telecommuting. But enough of those businesses were out there that there's an entrenched consulting market for Pro-Active Pandemic Preparation, and it's flu season and there aren't enough flu shots, so We're All Gonna Die! yet again.
You'll keep seeing that story until the Obama Administration either gets off its ass and stops being the Leftover-Bush Administration, or until it redeploys those people into some other politically useful job, like moving them from the Department of Homeland Security to the Department of Health and Human Services were they can tell us that If We Don't Get National HealthCare, We're All Gonna Die.
Yeah, yeah, you can bitch about the price, and bitch about carriers that think bandwidth-capping is a cool idea they should adopt, and if you've only got regular DSL instead of cable modem or fiber or VHDSL, then you can bitch about how you're only getting a megabit or two and can't watch HDTV in real time over it, but that's not where the problems are here. The problems are the connections businesses (and government agencies) have at their ends, which affect whether they can handle having 99% of their employees telecommuting. If you've got a culture that thinks F2F meetings are the only way to get things done, and wants 25Mbps Telepresence(tm) at everybody's house, then maybe there's a network problem, or at least wants Skype Video and $29 webcams, but basically it's a question of whether the business has enough bandwidth at their headquarters/datacenters to handle their workers, and whether the business-oriented ISPs have resources deployed to handle them. (For the most part, they do, and if they don't, then they'll upgrade if they start seeing actual orders from pandemic-panicked businesses.)
The problem they're trying to address (well, other than organizational self-preservation by looking useful, and helping out their friends and accomplices in the drama-creation business) isn't that lots of sick people will be staying home. It's that if the Pandemic gets bad enough, schools and businesses will tell healthy people to stay home and work from home to avoid getting sick, and maybe even the government will tell people to stay home. They're concerned about the load that those people will put on the network.
Obpolitics: Of course it's "propaganda, disinfo, lies, and bullshit". That's what the DHS was created to *do*. And just because the Bush Administration set it up to do a better job of scaring the children than the Clinton Administration or anybody since the Cuban Missile Crisis and Joe McCarthy had done, and just because Obama was elected to "Not Be Bush", that doesn't mean that the Obama Administration is ready to turn around and become pro-freedom or anti-scary any time soon.
ObTechnoPolitics: And ever since the SARS and Bird Flu scares didn't turn into pandemics, there's been a whole industry of computing/telecom/business consultants trying to get people to buy pandemic-planning consulting services, and they'd be halfway out of a job if the DHS stopped scaring people to drum up business. And if the EFF/Cypherpunks/Banks/HiTech people hadn't spent the 90s arguing that encryption was critical for business as well as civil liberties, you wouldn't be using VPNs to commute to work, and those consultants would still be selling you Analog Modem Pools and ISDN (bwah-hah-ha!)
Having said that, though, there are some serious problems with organizations that aren't prepared to have lots of their employees telecommuting. In some cases (including government agencies as well as financial and other businesses) their employees don't have work laptops, or don't have their home computers set up to run VPNs, or don't have big enough pipes at the office to handle 99% of the people working from home instead of 5-10%. Those organizations need to be convinced to upgrade their infrastructure. And there are probably some lame ISPs that still try to block VPNs except for people who pay extra for "business" broadband packages, though there's a lot less of that these days, and telling them to Just Stop That is a good thing. (And of course there are businesses whose employees actually do Real Physical Stuff all day instead of working on computers, but even most of those people use computers to keep track of the Stuff, or have managers or clerks who do that for them.)
From a technology standpoint, most ISPs these days can support IPv4 prioritization using either TOS or DSCP, though most ISPs don't support it across peering links, and some of them charge different amounts for it. There never was a really clear business plan for it. The Network Neutrality debate didn't help any of that - it was focused on some bonehead telecom execs wanting to charge more money for Google and YouTube (even before they merged), but from my perspective I'd really like it if my ISP prioritized my downlink to prefer UDP and VPN traffic over web and YouTube, and especially if they prioritized everything-else over BitTorrent, so I could be talking on VOIP while browsing the web and downloading totally-legal music. But even without prioritization, if you've got enough bandwidth on the home-user end to run YouTube, there's enough bandwidth for people to do real work (at least if they can get their kids to stop watching YouTube), and the business ends of the connections really do load-balance fairly well.
Yeah, sure, the current Mayans probably don't use the old hieroglyphics. But even so, having a grammar and vocabulary for the language so you know how things fit together helps immensely, compared to having to infer them from the hieroglyphs. Still, that Spanish guy shouldn't have burned the books.
A name like Sparc wasn't a typo by somebody trying to spell Spark. It was a deliberate marketing decision to create a name that sounded cool and wasn't spelled like the generic word because that probably couldn't be trademarked without conflicting with lots of other existing electrical and electronic trademarks. SparkFun is also a name that includes sparkiness in it, and is differentiated from the generic English word in different ways.
Besides, doesn't somebody already have a business method patent on suing everybody who's got a name even vaguely similar to yours when you don't have a legitimate case? SPARC Inc could get themselves in a lot of trouble for violating it.
And if anything, the Girl Genius folks could now sue SI for trying to get into the Spark business when they're clearly not that bright... On the other hand, the SparkFun people make lots of cool little gadgets and generally support Mad Science, so they're just fine.
Disclaimer: The Sun-2 in my attic is not an Ultrasparc-2. Nor is it a Sparcstation-2. Some of us have been in the Sun business for a long time.
Hey, give Red Dwarf some props as well!
I'm not saying that just for disinformation, though I do think that Weapons of Mass Destruction are Bad Things - I've worked there on data network consulting applications. When I said not to confuse tool-building with pork-barrel, that was a response to somebody else who said that spending $32M on a cloud-computer was pork-barrel.
While there are a few people at DOE who do graphics or computing just because it's cool, or because almost nobody else who does high-security computing research does it on supercomputers, they mostly view computing as a tool to do science and/or engineering. They're trying to model nuclear explosions, and it takes a lot of computing horsepower to do that.
And the "Imagine What You Could Do With A Cray" poster on their walls was cool, but when there's only one broadband cable system to get data across campus, back when 10Mbps Ethernet was still fast, and a few thousand people imagining what they can do with that Cray, networking problems get really hairy.
But yeah, apparently somebody else had made the mushroom cloud joke a bit earlier than I had. Didn't say anything useful about their computing environment, though.
Yes, the author does rudely smack the original poster in the face, but as Captain Jack Sparrow said, he "may have deserved that."
The original poster didn't give us enough info. Aggregation from multiple ISPs is possible, but it's a lot dodgier performance-wise than aggregating multiple connections from the same ISP. On the other hand, your choices of possible solutions depend a lot on your problem - if you want to make a single fat TCP session go faster, for instance a big file transfer, that's a lot harder than load-balancing a bunch of smaller sessions. And of course most cheap consumer solutions are very asymmetric, so the upstream will be your limitation, and don't give you good Layer 2 feedback and probably aren't running TCP ECN either.
There are problems which really need high memory bandwidth and don't fit on smaller-than-super computers, so a time slice on a supercomputer can be worth far more than full-time access to dedicated fast conventional computers. But those problems become less and less common as regular computers get bigger and faster - your laptop probably has a graphics processor that's faster than a Cray-2 by now...
SETI works on what gets described in the trade as "embarrassingly parallel" problems - supercomputers deal with stuff that's harder to get good parallel speedup without throwing fancy hardware at it. DOE problems are often somewhere in between, and unfortunately the boinc/seti/screensaver approach to ad-hoc supercomputing isn't always good for applications like LINPAK, so it's hard to compare the real computing power. However, if you ignore that (:-), most of the top computers in the Top500 list are doing nuclear or military work, and some are for weather bureaus or Earth simulation, but about half of the last decade, the SETI people are volunteering 2-10x as much CPU just to look for little green men as the largest military supercomputers were providing. The supercomputer guys are back on top, and I don't know that we'll catch up with them again, but on the other hand some of them are now doing genetics and other Good Things instead of Evil.
This is the DOE. They mainly work on nukes, and if they're experimenting with cloud computing, it's not because they're trying to do commercial or academic work into cloud computing, it's because they're trying to find powerful and cost-effective ways to get the computing horsepower they need to work on more nukes, just as when they work with supercomputers, it's not just because they're "Imagining what they can do with a Cray", it's because they're trying to solve problems that are really that big.
So, like, back off man, they're scientists!
Or don't back off,. because they're developing weapons as well as civilian power applications, but don't confuse tool-building with pork barrel.
Back when 1GB was enough to hold Wikipedia, there was application for iPods with screens that could run Wikipedia. This sounds like basically an evolution of that, and of course today the iPods/iPhones have a lot more storage (and Wikipedia's a bit larger - not sure how many GB you need for 3M articles in English and .5-1M in each of another dozen languages? probably around 20GB before compression, and good compressibility?)
There are three reasons you run raid, other than because it's cookl
SSDs are about speed, but don't have high capacity unless you've got an infinite budget. That's not the problem here. The problems are that
SSDs don't really help you on time, because you can't afford to run your whole shop on SSDs if it's big enough that RAID's not reliable. The recovery problem is helped somewhat by disk block error-correcting codes: you can detect if a block has corruption and rewrite/remap it when you're first storing it, so the ratio of bad blocks on disk isn't as bad as you'd expect from raw error rates, and you may need to continually recheck your raid block and rebuild bad blocks proactively rather than finding them after failures.
The big problem with RAID on bigger disks is that you're splitting data across N+1 disks, so if one disk fails catastrophically and needs to be rebuilt, you need to have good data on the remaining N disks, and as the disks become larger, the chances that all that data is good are no longer near-100%. So you're likely to get some bad data even with RAID, though you're not going to have as much bad data as you'd have without RAID, and you're still avoiding the problem that if you disk fails, you lose everything; you're just not guaranteed to lose nothing.
I pulled into a parking lot in Milpitas to make a phone call and use my computer. I didn't need to be online for the call, just look at stuff, but I was pleased to see that there was a wireless signal, they've got tons of free access points all over Milpitas, and the signal was pretty good., It wasn't foolproof - they have a login-timeout browser window thingy, and connecting to my company VPN meant killed its connection so it cut me off after about 5 minutes, but that was enough to download any new email, and I could log back in without the VPN and see the web and my home email just fine.
NYtimes.com is usually on my exceptions list, but not today...
Anybody know what the malware sites are, either by DNS name or IP address?
You've accused all the physicists in the world of crackpottery, and while perhaps you spent some time writing your essay, it's not "a well-formed argument"; it's closer to "not even clear enough to be wrong". It does in fact speak for itself, which is why I recommended the "-1 Crackpot" rating for your earlier posting, and I'm at least as identified around here as you are, having been online since ~1981...
You appear to be confused about what an "ad hominem" attack is. If you remember your junior high school Latin or logic classes, it's arguing "to the man" instead of "to the topic". If I had said "Oh, that's just Louis Savain, so anything he says is bogus", or "He's a New Ager, so anything he says is bogus", that would be "ad hominem".
That's not what's happening here - I went and read your web page, looked at a bunch of your arguments, even got as far as quoting your own words in my posting. I'm not saying people should discount your fine technical work simply because you're a crackpot - I'm saying that your arguments display a hopelessly cracked misunderstanding of physics, and that they shouldn't waste their time reading it except for entertainment purposes, but that you appear to actually believe them so you don't deserve a "-1 Troll" down-rating. (And as far as your posting's first two sentences go, I'm also highly skeptical about whether space-based power makes sense or will work, but the rest of your article and your website makes it clear that you think so for hopelessly wrong reasons.) If you want to claim that I'm insulting you because of your work, and that that's rude, well perhaps it's a fair cop (:-), but it's just the opposite of insulting your work because of you, which is what an ad hominem argument is, and given that you start many of your paragraphs describing how you think normal physicists think the work works as "The Crackpottery", well, you're not in much position to argue - you not only get Newtonian physics wrong, you also get quantum wrong and don't even get Yin and Yang right.
And yes, there are many Bill Stewarts out there - feel free to Google me; I'm the one who's not the drummer and wasn't killed in Nicaragua. And I've been out of town for a couple of days so I didn't see your reply until today.