Huh? Cogent is a TIER 1 ISP, they have peering agreements with all the other TIER 1 providers. Sprint decided to treat them as a TIER 2 provider and charge them for peering. Cogent (correctly) thinks this is stupid and has refused to pay as there has always been a general agreement that all TIER 1 ISP's peer for free, it's what makes the Internet work. Cogent ASN's can reach all other ASN's except those that are single homed with Sprint. Cogent doesn't need Sprint to connect to the Internet, they are part of the Internet. The big thing with this type of showdown is that Sprint has a bunch of end users and Cogent has a bunch of content providers, Sprint thinks that the content providers are more likely to yell and scream about not having the percentage of customers that Sprint represents more than the Sprint customers will yell about not being able to get to their favorite site that is hosted on a Cogent customer. Sprint may be correct but it still doesn't make it right and the only reason they will get any sympathy is that Cogent has driven down the price of bandwidth to the point that all of the TIER 1 ISP's are a bit miffed at them. As a user I think that a competitor dropping prices is a GOOD thing =)
Companies. Specifically companies needing CAD workstation (though they'd use a card that costs almost as much as your estimate). Also I can't wait for Core i7 to come to the HP DL line, I expect I can finally use Intel for database work because it's been their very poor multicore memory bandwidth that has kept AMD in the lead up till now.
Nah, our SONET loop goes out to two different streets and to two different POP's which are routed out separate directions. I know that isn't typical or universally available but we did our homework =)
First it's not just single homed connections that suffer, due to cost many multihomed systems have such a large weighting to Cogent that they won't use the other connection unless the physical route to Cogent is down (I know that's stupid but it's the real world). Also the peering contracts Cogent has with their other peers might not allow for transit traffic and further those contracts are probably enforced through ACL's that would drop the traffic into the bit bucket anyways.
Actually unless a third party agrees to transit traffic between Cogent and Sprint the Internet IS broken, there are hundreds of AS's unable to communicate. It's not a technology problem, there is plenty of interconnection, it is a political and financial problem (Layer 8, often the most troublesome).
Our ISP is AT&T, I don't think they are going to get depeered =) Also the fact that a bunch of amateurs didn't know how to get STP under control has nothing to do with the reliability of Cisco hardware. It's like the condition of our network before the current staff got there. The company had a laser link between buildings that would go out with the slightest bit of precipitation, this would have been bad enough for the other building but whenever the network reconverged their was a spanning tree war because there was a switch in each building that was setup as a root bridge. Turning off the root bridge in the other building and setting spanning tree weights correctly eliminated the main network problems and digging a horizontal bore and running fiber eliminated the other buildings problems.
Even my very worst site out of 13 that I monitor has 99.8% availability, if you are getting much worse than that then I strongly suggest you change ISP's.
WAN gateway downtime, what? Our DS3 hasn't been down once since it was installed in September 2006 and the firewall cluster behind it has likewise never been down since it was installed in 2005. If you have significant internet outages you are doing something wrong. Forget LAN outages, it just doesn't happen. Of course that's why we paid the premium for Cisco chassis based switches with redundant supervisors for both the datacenter and the wiring closets.
Well, there is no 256 core x64 machine, but there is a 96 core machine (Unisys ES7600R) which should get them close enough to prove the algorithms work.
No, MS is just licensing in a way that scales with hardware which is the pro-consumer method of doing things. Besides there is a technical reason for doing so, in the x86 architecture there is an identifier that reports back the CPU socket number which is what MS has always used for licensing within the Windows OS.
Better than Oracle licensing, Oracle Enterprise is around $50K (list) per 2 cores on x86/x64. It's funny we had our Oracle reps out the other day and they talked about how cool hardware was getting with a quad 6-core server with 256GB or ram from Dell costing only $55K. My response was "yeah but the Oracle licensing would be over $300K at our costs, not so cheap now is it". They had no response to that.
I get consistent 750KB/s on popular torrents but that's because I limit it to that speed to leave enough upload for good VoIP performance, overnight I could leave it full blast and get a bit over 1MB/s. Storage is cheap, a 500GB drive is $59 on sale, a 1TB drive is around $129. I wouldn't call 300 movies for about half the cost of a BlueRay player expensive =)
NT Server had 3 editions, Normal, Enterprise, and Terminal Server and was available for four CPU architectures, x86, MIPS, PPC, and Alpha. Windows 2000 had four server flavors, Standard, Advanced, Datacenter edition, and Small Business Server. Windows 2003 has 8 flavors, Standard, Enterprise, Datacenter, Web, Compute, Storage, Home, and Small Business Server. Windows 2003 supports three different CPU architectures, x86, x64, and IA64. So it's really never been as simple as you make it out to be.
There are almost zero BD disks at 54Mbps. Only ones using MPEG2 would use the full possible bitrate. The much more common titles using MPEG4 or VC1 cap out at around 30Mbps.
Internet speeds are plenty fast for trading BD rips. Even at full quality they are only about 30Mbps on average. My download speed is 10Mbps so it would take me about 6 hours to download a 2 hour movie, certainly not realtime streaming but fast enough to not be too annoying. Just start the download one day and watch it the next.
As it turns out it was probably a GOOD thing that they didn't require that the original plan be followed closely. The direct injection of capital and the purchase of commercial paper were both MUCH better uses of money than the buying of toxic assets will be and they were much quicker ways to get that money into the market since we still don't even have people hired to valuate the assets.
Well, the $700B bailout really was just a gesture. The EU used almost $2T to back their banking system and the asian countries about $1T. Since each of those economic areas is about the same size as the US economy (~$14T/year) it really puts the US plan in perspective.
Like I said the ACH protects corporate accounts because their native bank will reject the transaction if there isn't a matching entry from the corporate upload. It shouldn't be difficult for them to extend that system to personal accounts though it would make it a bit less convenient (you would have to go online and enter the check number and amount within a day or two of writing the check or it would bounce).
Nah, telco/cellphone collections affect your credit score only slightly more than doctor bills do, ie not very much. The credit scoring companies know that those entities are about as likely as the customer to be responsible for billing problems. So long as you generally pay your bills ontime every month a single collection notice isn't going to change your score enough to affect your rates or ability to qualify for a loan. I do love how corporate America has convinced people that it's the end of the world if they don't pay every single bill justified or not.
At a previous employer we had the ultimate screwup from a telco. We moved offices and returned the equipment from the old office. We had a signed receipt for the return of the equipment, but because the person who had received it hadn't entered it into their computer they insisted that we owed them for the equipment. After months of arguing we finally gave in and paid the stupid fee. TWO years later their accounting system decided that it had applied that check to our monthly charges instead of to the fee and turned off our service without notification because it saw the automated letters that had been sent out ~30 months prior. That was the last straw for them, we switched to a fractional PRI with a CLEC and never looked back.
It doesn't HAVE to cost time and money to verify a transaction, we've had smartchips for well over a decade now and they are widely used in Europe. However, here in the US I had a smartcard in my credit card for 6 years and had a total of THREE places use the chip in that entire period. At two of those it took extra time because the cashier had never used the smartcard reader and so didn't understand the message the credit card reader was giving them.
Actually banks no longer transfer physical checks, they ship around images of the checks. The banks did this to reduce costs, but it obviously comes at the cost of security. Since it isn't their money they are protecting they just don't care, if they can reduce their costs and only risk the few small accounts that get hacked then it's definitely a net win for them.
The flipside of this is that Knuth is wrong when he says "Before long, companies will find it impossible to give out paychecks without exposing themselves to unacceptable risk." Corporate accounts are protected by double entry protection. In order for a corporate check to be considered valid the company has to upload a file to their bank with the check number and the amount, if the bank hasn't received a matching upload then they reject the check is invalid when it comes through the clearing house.
Slashdot Mirror
Huh? Cogent is a TIER 1 ISP, they have peering agreements with all the other TIER 1 providers. Sprint decided to treat them as a TIER 2 provider and charge them for peering. Cogent (correctly) thinks this is stupid and has refused to pay as there has always been a general agreement that all TIER 1 ISP's peer for free, it's what makes the Internet work. Cogent ASN's can reach all other ASN's except those that are single homed with Sprint. Cogent doesn't need Sprint to connect to the Internet, they are part of the Internet. The big thing with this type of showdown is that Sprint has a bunch of end users and Cogent has a bunch of content providers, Sprint thinks that the content providers are more likely to yell and scream about not having the percentage of customers that Sprint represents more than the Sprint customers will yell about not being able to get to their favorite site that is hosted on a Cogent customer. Sprint may be correct but it still doesn't make it right and the only reason they will get any sympathy is that Cogent has driven down the price of bandwidth to the point that all of the TIER 1 ISP's are a bit miffed at them. As a user I think that a competitor dropping prices is a GOOD thing =)
Companies. Specifically companies needing CAD workstation (though they'd use a card that costs almost as much as your estimate). Also I can't wait for Core i7 to come to the HP DL line, I expect I can finally use Intel for database work because it's been their very poor multicore memory bandwidth that has kept AMD in the lead up till now.
Nah, our SONET loop goes out to two different streets and to two different POP's which are routed out separate directions. I know that isn't typical or universally available but we did our homework =)
First it's not just single homed connections that suffer, due to cost many multihomed systems have such a large weighting to Cogent that they won't use the other connection unless the physical route to Cogent is down (I know that's stupid but it's the real world). Also the peering contracts Cogent has with their other peers might not allow for transit traffic and further those contracts are probably enforced through ACL's that would drop the traffic into the bit bucket anyways.
Actually unless a third party agrees to transit traffic between Cogent and Sprint the Internet IS broken, there are hundreds of AS's unable to communicate. It's not a technology problem, there is plenty of interconnection, it is a political and financial problem (Layer 8, often the most troublesome).
Our ISP is AT&T, I don't think they are going to get depeered =) Also the fact that a bunch of amateurs didn't know how to get STP under control has nothing to do with the reliability of Cisco hardware. It's like the condition of our network before the current staff got there. The company had a laser link between buildings that would go out with the slightest bit of precipitation, this would have been bad enough for the other building but whenever the network reconverged their was a spanning tree war because there was a switch in each building that was setup as a root bridge. Turning off the root bridge in the other building and setting spanning tree weights correctly eliminated the main network problems and digging a horizontal bore and running fiber eliminated the other buildings problems.
Even my very worst site out of 13 that I monitor has 99.8% availability, if you are getting much worse than that then I strongly suggest you change ISP's.
WAN gateway downtime, what? Our DS3 hasn't been down once since it was installed in September 2006 and the firewall cluster behind it has likewise never been down since it was installed in 2005. If you have significant internet outages you are doing something wrong. Forget LAN outages, it just doesn't happen. Of course that's why we paid the premium for Cisco chassis based switches with redundant supervisors for both the datacenter and the wiring closets.
There are plenty, they are just running in a VM after being P2V'd =)
Well, there is no 256 core x64 machine, but there is a 96 core machine (Unisys ES7600R) which should get them close enough to prove the algorithms work.
Wide Open West, they just upgraded me from 4/.5 to 10/1 =) Of course at work we have one site that was on Fios and they had 15/2 for $35/month!
No, MS is just licensing in a way that scales with hardware which is the pro-consumer method of doing things. Besides there is a technical reason for doing so, in the x86 architecture there is an identifier that reports back the CPU socket number which is what MS has always used for licensing within the Windows OS.
Better than Oracle licensing, Oracle Enterprise is around $50K (list) per 2 cores on x86/x64. It's funny we had our Oracle reps out the other day and they talked about how cool hardware was getting with a quad 6-core server with 256GB or ram from Dell costing only $55K. My response was "yeah but the Oracle licensing would be over $300K at our costs, not so cheap now is it". They had no response to that.
I get consistent 750KB/s on popular torrents but that's because I limit it to that speed to leave enough upload for good VoIP performance, overnight I could leave it full blast and get a bit over 1MB/s. Storage is cheap, a 500GB drive is $59 on sale, a 1TB drive is around $129. I wouldn't call 300 movies for about half the cost of a BlueRay player expensive =)
NT Server had 3 editions, Normal, Enterprise, and Terminal Server and was available for four CPU architectures, x86, MIPS, PPC, and Alpha. Windows 2000 had four server flavors, Standard, Advanced, Datacenter edition, and Small Business Server. Windows 2003 has 8 flavors, Standard, Enterprise, Datacenter, Web, Compute, Storage, Home, and Small Business Server. Windows 2003 supports three different CPU architectures, x86, x64, and IA64. So it's really never been as simple as you make it out to be.
There are almost zero BD disks at 54Mbps. Only ones using MPEG2 would use the full possible bitrate. The much more common titles using MPEG4 or VC1 cap out at around 30Mbps.
Internet speeds are plenty fast for trading BD rips. Even at full quality they are only about 30Mbps on average. My download speed is 10Mbps so it would take me about 6 hours to download a 2 hour movie, certainly not realtime streaming but fast enough to not be too annoying. Just start the download one day and watch it the next.
As it turns out it was probably a GOOD thing that they didn't require that the original plan be followed closely. The direct injection of capital and the purchase of commercial paper were both MUCH better uses of money than the buying of toxic assets will be and they were much quicker ways to get that money into the market since we still don't even have people hired to valuate the assets.
Well, the $700B bailout really was just a gesture. The EU used almost $2T to back their banking system and the asian countries about $1T. Since each of those economic areas is about the same size as the US economy (~$14T/year) it really puts the US plan in perspective.
Like I said the ACH protects corporate accounts because their native bank will reject the transaction if there isn't a matching entry from the corporate upload. It shouldn't be difficult for them to extend that system to personal accounts though it would make it a bit less convenient (you would have to go online and enter the check number and amount within a day or two of writing the check or it would bounce).
Nah, telco/cellphone collections affect your credit score only slightly more than doctor bills do, ie not very much. The credit scoring companies know that those entities are about as likely as the customer to be responsible for billing problems. So long as you generally pay your bills ontime every month a single collection notice isn't going to change your score enough to affect your rates or ability to qualify for a loan. I do love how corporate America has convinced people that it's the end of the world if they don't pay every single bill justified or not.
At a previous employer we had the ultimate screwup from a telco. We moved offices and returned the equipment from the old office. We had a signed receipt for the return of the equipment, but because the person who had received it hadn't entered it into their computer they insisted that we owed them for the equipment. After months of arguing we finally gave in and paid the stupid fee. TWO years later their accounting system decided that it had applied that check to our monthly charges instead of to the fee and turned off our service without notification because it saw the automated letters that had been sent out ~30 months prior. That was the last straw for them, we switched to a fractional PRI with a CLEC and never looked back.
It doesn't HAVE to cost time and money to verify a transaction, we've had smartchips for well over a decade now and they are widely used in Europe. However, here in the US I had a smartcard in my credit card for 6 years and had a total of THREE places use the chip in that entire period. At two of those it took extra time because the cashier had never used the smartcard reader and so didn't understand the message the credit card reader was giving them.
Not ANYprinter, any printer with magnetic MICR ink, at least if the merchant/bank is using even the most rudimentary of anti-fraud protections.
Actually banks no longer transfer physical checks, they ship around images of the checks. The banks did this to reduce costs, but it obviously comes at the cost of security. Since it isn't their money they are protecting they just don't care, if they can reduce their costs and only risk the few small accounts that get hacked then it's definitely a net win for them.
The flipside of this is that Knuth is wrong when he says "Before long, companies will find it impossible to give out paychecks without exposing themselves to unacceptable risk." Corporate accounts are protected by double entry protection. In order for a corporate check to be considered valid the company has to upload a file to their bank with the check number and the amount, if the bank hasn't received a matching upload then they reject the check is invalid when it comes through the clearing house.