Understatement of the year, think high tens to hundreds of thousands. A little over 500,000 fibers can fit in a 6" pipe, assuming perfect packing. Even 1/2 full would be over 200k fibers. You can purchase 144 strand 1/4" diameter with a trace wire, for something like under $1/foot when you purchase more than 1km at a time.
I am not sure how practical that would be. I would not want to be the person fixing a cut on a main pipe like that. On the flip side of that, I would not want to be the person who cut the fiber because I see red and yellow flags right next to the orange. If you're off by even a few inches, expect a gas leak with a 1400v life wire sending sparks. F*ck'n Darwin your ass.
If the US Postal ran our Internet, we'd have 10gb FTTH, if you want to use an appropriate analogy. US postal is about 1/2 the price of FedEx of UPS when it comes to small parcel Next Day or 2 day, and it comes with more basic features without paying extra. I've also had a better experience with USPS when shipping to friends and family, than FedEx or UPS. USPS is self sustaining without tax support.
Wifi is too unreliable and slow. Wireless will become a replacement when the latency is sub 10ms and virtual 0 packet-loss while maintain a constant speed.
If I'm transferring 1gb/s and suddenly someone else starts transferring and the available bandwidth drops to 500mb, I'm going to get hit with a mixture of packetloss and latency for the duration of the TCP receive window. Can't have that. VOIP calls to 911 should be reliable, even with other users consuming large amounts of bandwidth, and with no QoS. This will be infeasible with wifi for the foreseeable future.
Airplanes need to use very special flexible joints because Aluminum is so brittle. Aluminum becomes very weak very fast when it flexes to the point of being deformed. Steel, not so much, unless you put a ton of carbon in it.
He was correct when it comes to the general properties of steel, but not all cases.
It's more like, everyone should performance tune their engine's timings themselves, instead of paying a professional to do it. Datacenters are not cheap to design correctly. With "the cloud", you pay for characteristics, instead of concerning yourself on how it's implemented. This isn't perfect, but it works "good enough" and quite a bit better than what many can do on their own.
Which experiment? I only see one experiment listed and " researchers in the past have been able to prove that it works". I thought it was common knowledge that we an already turn light into matter. It's a common issue with high power lasers for fusion research. Nothing worse than you laser output going down as you add more energy because you increase the chance of your photons turning into matter.
First off, that $5.7b is for the first quarter, you know, 3 months. Second, that only applies to the 11.3mil Uverse customers. How many industries can get $166/month average per customer?
And it uses that detection to fix the bad data from good data from your other RAID drives. Not just using one drive, right? Can't blame the FS for mistakes of the user, you also can't blame the FS for really bad luck. The Earth blew up when the Sun went super-nova. ZFS is a bad FS because it didn't protect my data.
If you don't like shiney, then there's always the console. For everyone else who likes desktops for stuff like videos or web browsing, annoyances like tearing is a distraction, like a 3 year old throwing a temper tantrum. Tearing was an issue fixed back in the 90s.
OpenZFS. They're getting rid of "versions" and just having "Feature flags". This will allow you to create ZFS pools on one system, and just make sure what ever features that are only supported on another target system, are enable when you create the pool.
btrfs is interesting, but it's taking a long time to get anywhere and it has some big backers. I've also read some really well written blogs from sysadmins who have been Unix admins since the beginning of time, and they had some really good examples of some "Features" of btrfs that a sysadmin should never-ever use under any circumstance, and some features that are half-asses that are nearly a requirement for any good sysadmin, but cannot be done because of those other "bad" features.
One such example is btrfs allows a volume to be mounted under multiple parents. In order to handle this "awesome" feature, they had to give up the ability to atomically snapshot across volumes. In ZFS, if you mount a volume under another volume and snapshot the parent, the children will automatically be atomically included, not so with btrfs, that's an impossibility to add a feature that should never be used.
It's a worthy goal to have. We need more competition in the FS sector. Many times competition is the inspiration for new features, even if some of these FS don't even make it off the ground. ZFS is great, but it's not perfect, and they only have so many resources to throw at new ideas to test. Monoculture is never a good thing.
If your customers are getting lag, either the ISP is not delivering quality bandwidth to the customer or the server operator is not paying for high quality bandwidth to their ISP/datacenter.
If a customer only has a 1.5mb line and someone wants to stream Netflix while someone else plays games, then they can shove off. They need more bandwidth, they don't need priority lanes to hide that fact.
Once you account for L1 bandwidth and load latencies, AVX FPU "sharing" still isn't an issue. L1 is only so fast and 32byte loads takes a bit, and the scheduler can let the hyper-thread do work while the other thread is loading.
It was a long while ago, but I remember watching about some nifty devices that used a very specific range of microwave radiation, and it could turn tires, plastic bags, and a huge range of plastic based objects back into a strange oily carbon soup that could be reused. For this device to work, they just threw garbage in it, and metal and oil would come out the other end, separated. We need that kind of device, but working large scale.
Except in this case the buffet is not the main cost, it's the gold plates, platinum forks, and entertainment, which is the analogy to infrastructure. Bandwidth has little to do with costs compared to infrastructure.
Slashdot Mirror
That is enough for thousands of fibers
Understatement of the year, think high tens to hundreds of thousands. A little over 500,000 fibers can fit in a 6" pipe, assuming perfect packing. Even 1/2 full would be over 200k fibers. You can purchase 144 strand 1/4" diameter with a trace wire, for something like under $1/foot when you purchase more than 1km at a time.
I am not sure how practical that would be. I would not want to be the person fixing a cut on a main pipe like that. On the flip side of that, I would not want to be the person who cut the fiber because I see red and yellow flags right next to the orange. If you're off by even a few inches, expect a gas leak with a 1400v life wire sending sparks. F*ck'n Darwin your ass.
If the US Postal ran our Internet, we'd have 10gb FTTH, if you want to use an appropriate analogy. US postal is about 1/2 the price of FedEx of UPS when it comes to small parcel Next Day or 2 day, and it comes with more basic features without paying extra. I've also had a better experience with USPS when shipping to friends and family, than FedEx or UPS. USPS is self sustaining without tax support.
Wifi is too unreliable and slow. Wireless will become a replacement when the latency is sub 10ms and virtual 0 packet-loss while maintain a constant speed.
If I'm transferring 1gb/s and suddenly someone else starts transferring and the available bandwidth drops to 500mb, I'm going to get hit with a mixture of packetloss and latency for the duration of the TCP receive window. Can't have that. VOIP calls to 911 should be reliable, even with other users consuming large amounts of bandwidth, and with no QoS. This will be infeasible with wifi for the foreseeable future.
Not trolling, but how would Google Fiber be illegal under common carrier?
Airplanes need to use very special flexible joints because Aluminum is so brittle. Aluminum becomes very weak very fast when it flexes to the point of being deformed. Steel, not so much, unless you put a ton of carbon in it.
He was correct when it comes to the general properties of steel, but not all cases.
Electric Heating coils, like what trains use.
It's more like, everyone should performance tune their engine's timings themselves, instead of paying a professional to do it. Datacenters are not cheap to design correctly. With "the cloud", you pay for characteristics, instead of concerning yourself on how it's implemented. This isn't perfect, but it works "good enough" and quite a bit better than what many can do on their own.
It's not a binary exclusive thing. Some politicians can be one type and some can be another type, but as a whole, they can all be both.
Which experiment? I only see one experiment listed and " researchers in the past have been able to prove that it works". I thought it was common knowledge that we an already turn light into matter. It's a common issue with high power lasers for fusion research. Nothing worse than you laser output going down as you add more energy because you increase the chance of your photons turning into matter.
Yeah, maybe Comcast or Verizon instead.
First off, that $5.7b is for the first quarter, you know, 3 months. Second, that only applies to the 11.3mil Uverse customers. How many industries can get $166/month average per customer?
I always assumed "brogramming" was similar to "Pair programming", but urbandictionary gives a less professional image.
And it uses that detection to fix the bad data from good data from your other RAID drives. Not just using one drive, right? Can't blame the FS for mistakes of the user, you also can't blame the FS for really bad luck. The Earth blew up when the Sun went super-nova. ZFS is a bad FS because it didn't protect my data.
If you don't like shiney, then there's always the console. For everyone else who likes desktops for stuff like videos or web browsing, annoyances like tearing is a distraction, like a 3 year old throwing a temper tantrum. Tearing was an issue fixed back in the 90s.
I've never heard of anyone ever losing data with ZFS, even people with petabytes of R&D data.
OpenZFS. They're getting rid of "versions" and just having "Feature flags". This will allow you to create ZFS pools on one system, and just make sure what ever features that are only supported on another target system, are enable when you create the pool.
btrfs is interesting, but it's taking a long time to get anywhere and it has some big backers. I've also read some really well written blogs from sysadmins who have been Unix admins since the beginning of time, and they had some really good examples of some "Features" of btrfs that a sysadmin should never-ever use under any circumstance, and some features that are half-asses that are nearly a requirement for any good sysadmin, but cannot be done because of those other "bad" features.
One such example is btrfs allows a volume to be mounted under multiple parents. In order to handle this "awesome" feature, they had to give up the ability to atomically snapshot across volumes. In ZFS, if you mount a volume under another volume and snapshot the parent, the children will automatically be atomically included, not so with btrfs, that's an impossibility to add a feature that should never be used.
It's a worthy goal to have. We need more competition in the FS sector. Many times competition is the inspiration for new features, even if some of these FS don't even make it off the ground. ZFS is great, but it's not perfect, and they only have so many resources to throw at new ideas to test. Monoculture is never a good thing.
If your customers are getting lag, either the ISP is not delivering quality bandwidth to the customer or the server operator is not paying for high quality bandwidth to their ISP/datacenter.
If a customer only has a 1.5mb line and someone wants to stream Netflix while someone else plays games, then they can shove off. They need more bandwidth, they don't need priority lanes to hide that fact.
Lnyx doesn't seem to load web pages any faster than it did 15 years ago on a 56k.
Once you account for L1 bandwidth and load latencies, AVX FPU "sharing" still isn't an issue. L1 is only so fast and 32byte loads takes a bit, and the scheduler can let the hyper-thread do work while the other thread is loading.
I thought part of the reason it was so far out was to make it so it wasn't moving(relatively speaking) and to remove light from the Earth and Moon.
It supports a small subset of cases that can be ACID.
It was a long while ago, but I remember watching about some nifty devices that used a very specific range of microwave radiation, and it could turn tires, plastic bags, and a huge range of plastic based objects back into a strange oily carbon soup that could be reused. For this device to work, they just threw garbage in it, and metal and oil would come out the other end, separated. We need that kind of device, but working large scale.
Except in this case the buffet is not the main cost, it's the gold plates, platinum forks, and entertainment, which is the analogy to infrastructure. Bandwidth has little to do with costs compared to infrastructure.