$1k more for most PIXen for a 3DES key, much less on a 501
Why, other than marketing? Does it take them so much computing power to generate this key that they have to rent a mainframe somewhere?:)
Having a full box as a firewall/router has an advantage, you can arbitrarily upgrade it as needed. Some new whiz-bang VPN technology comes out, you just install it, no running back to Cisco for more expensive upgrades.
As you said, one can install as little or as much as one needs. If you are in a security sensitive environment, then run the OS on read only media, and don't even install a hard disk, just lots of RAM for a ramdisk. Have the thing reboot every night if you are paranoid, and don't install any services facing the internet that would need patching (unless you need to), and you would rarely need to recreate the media due to security patches, since very very few security flaws are in the kernel filters themselves. Only one I can recall with iptables, the FTP RELATED bug, only affected certain users anyway, who were using that rule.
My point is.... flexibility and freedom, at a trivial hardware cost, no lock-in, and performance that can easily match, or beat, Cisco offerings.
Someone else mentioned a failover mode that was seamless with Cisco. If you need something like that, that only Cisco can provide, then by all means go Cisco, but for 99% of the people out there, there are very good alternatives.
Cisco is a relic at this point, they are being commodized (is that a word?) into non-existance. The high end offerings are way overpriced, for example, several hundred dollars for a single GBIC for a fiber optic GB ethernet port, or $6000-$7000 for a blade of 5 or 6 copper gigabit ports for a Catalyst switch.... Those prices can't compete on the high end either, being 10X more expensive than commodity hardware is a killer, especially when the value you provide is nominal.
I think Cisco's financials lately strongly support this too.
Re:You're behind a hub? You ain't got a T1.
on
How to Test Your T1?
·
· Score: 1
You know, I think I forgot a zero somewhere:) I believe I meant to say over 500 times, which works out around 93Mbytes/sec.
Re:You're behind a hub? You ain't got a T1.
on
How to Test Your T1?
·
· Score: 1
If a PC requires tuning to sustain 1.5Mbps you have major problems, or you need to upgrade your 286. 1.5Mbps is extremely slow in computer terms. The PCI bus has over 50 times that capacity.
300-900kbps is most definitely not a real T1. He has likely oversold his upstream. We have a real qwest T1 at work, and most file downloads are in the 130-170KBytes/s range.
Extra hops add latency too. In any case, the salesman is probably mostly correct.
But, it still may be a better deal for you to go with the local guy. You need to ask yourself, are you getting the T1 and plan to use 100% of it bandwidth wise, or are you getting it for the freedom that comes with a T1? If it is more the latter, the local ISP might be a good deal.
You know, I always wonder how someone gets the job of "commentator". I often hear them with their 60 second or 2 minute pieces on NPR, and I wonder... Did they apply for the job? Does one become an apprentice commentator first? Do you just start bitching a lot and being nostalgic, and then people assume you are a commentator, and the rest follows?
I would have gotten a signed waiver that said that the employee signing it was representing that he had authority to allow such a thing. This guy will be lucky if he isn't sued.
Re:Netscape popularity a problem for webmasters ?
on
Netscape 7.0 is Out
·
· Score: 1
since any normal linux user will use a recent Mozilla (or galeon & co) or Konqueror.
Oh, yeah, no normal user would want to use a browser under 100 megs.
The market does not demand such a thing, if it did, C3 would be flying off the shelves.
Electricity to run the computer for 8 hours a day is negligible. Even if the faster chip only saves 10 minutes of employee time, for most reasonable salaries, it was worth it. Employees are too expensive and electricity is too cheap for people to be worried about 50 watts here and there.
I once heard of an individual who tried this and burped for an extended period time, an involuntary response that rid his body of gas produced as the nitrogen changed phases. He was concerned that he might burp long enough to be harmed by the lack of oxygen. That sounds more plausible to me. But I can believe portions of the digestive system might be frozen by the liquid nitrogen."
When I can buy a dual athlon board with PCI-X for less than $500, then it will be enough. It doesn't help until it becomes available, and cards are being made for it. It's possible this standard will be completely passed over if it doesn't get adoption quickly enough, in 5 or 10 years 1GB per second may seem slow.
As you said, you burn anything critical on CDR. Those are your backups, not the RAID1.
RAID1/3/5 protects against one thing only, physical hard disk failure. Backups protect not only against hard disk failure, but user error, viruses, crackers, kernel goes insane, software bugs, etc..
If your filesystem module, for example, goes nuts and corrupts your filesystem, RAID1 is no help, it is the same with the other cases I named, and many others. The RAID will happily corrupt/delete the mirror at the same time.
Redundancy in RAID is useful, but as I said, never a substitute for backups, it only protects against one small class of data loss.
This is a trouble, I had to replicate a server and send it offsite, putting the 600GB on it took two days at 100Mbit speeds. Even at the full speed of the RAID, 40MB/Sec, it takes several hours, PCI maxes around 150MB/sec for 64bit in the real world, assumedly as faster controllers come out, 64/66 could be pushed a little higher than that, since there is some room there before you hit the theoritical limit.
We need to take all this AGP technology and quit wasting it on toys for gamer kids, and apply it to a more generalized bus technology.
As I posted to another message, we are currently phasing out tape in favor of keeping many copies of data on various RAIDs for backup. You are correct, I'd call it a "backup crisis".
The biggest helical scan 200GB tapes are very very expensive compared to hard disk prices. We have over 4TB of disk space at work, most of it is redundancy (not counting RAID redundancy), but we do have almost 1TB of live data.
I've resorted to creative rsyncing for main backups, the Macs use retrospect (we are going to soon target that to a hard disk rather than tape), Veritas and the 1TB tape robot are still running, but too slow and cumbersome to be practical (if we ever needed to restore the full 1TB of data off that thing it would take weeks).
And really, who do we have to blame? I'd look at the MPAA... who has the most to lose from large removable media?
Scary, since we are phasing out most of the tape at work in favor of hard disks. Tape is just more expensive per GB these days, slower, less flexible, arguably less reliable.
Of course, if we are going to get bit rot in 5 years, then tape isn't looking so bad. At least we keep all important data in triplicate, or more.
I don't see anyone else posting this... It's true. If you want to run an OS from a rape-you-in-the-ass company that has no respect for its customers, you better not do it with my goddamn medical data. MS products are not fit for important uses. Running personal web pages from MS products is probably OK, but for any actual business use you need a real OS.
How do you understand that an encoder should pay, but not the decoder? That makes no sense. That's like saying people should pay to create ASCII text files, but not to read them. Software is software, and software patents are stupid no matter how they are applied.
Ross died of sudden adult death syndrome after a basketball match.
Oh come on! "Sudden adult death syndrome"?!!? WTF! This whole thing seems to be the same as the GHB scare... a bunch of lies and propaganda, funded by incumbent industries to crush disruptive products.
From what I remember of CNBC interviews with the startups that were trying this stuff years ago, apparently they had trouble segmenting the signal. Big chunks of the grid are electromagnetically coupled at any one time, so the difficulty was making the segments small enough to be able to provide meaningful bandwidth.
Slashdot Mirror
Men must have voted for this one.
$1k more for most PIXen for a 3DES key, much less on a 501
:)
Why, other than marketing? Does it take them so much computing power to generate this key that they have to rent a mainframe somewhere?
Having a full box as a firewall/router has an advantage, you can arbitrarily upgrade it as needed. Some new whiz-bang VPN technology comes out, you just install it, no running back to Cisco for more expensive upgrades.
As you said, one can install as little or as much as one needs. If you are in a security sensitive environment, then run the OS on read only media, and don't even install a hard disk, just lots of RAM for a ramdisk. Have the thing reboot every night if you are paranoid, and don't install any services facing the internet that would need patching (unless you need to), and you would rarely need to recreate the media due to security patches, since very very few security flaws are in the kernel filters themselves. Only one I can recall with iptables, the FTP RELATED bug, only affected certain users anyway, who were using that rule.
My point is.... flexibility and freedom, at a trivial hardware cost, no lock-in, and performance that can easily match, or beat, Cisco offerings.
Someone else mentioned a failover mode that was seamless with Cisco. If you need something like that, that only Cisco can provide, then by all means go Cisco, but for 99% of the people out there, there are very good alternatives.
Cisco is a relic at this point, they are being commodized (is that a word?) into non-existance. The high end offerings are way overpriced, for example, several hundred dollars for a single GBIC for a fiber optic GB ethernet port, or $6000-$7000 for a blade of 5 or 6 copper gigabit ports for a Catalyst switch.... Those prices can't compete on the high end either, being 10X more expensive than commodity hardware is a killer, especially when the value you provide is nominal.
I think Cisco's financials lately strongly support this too.
So only intelligent speech should be protected?
You know, I think I forgot a zero somewhere :) I believe I meant to say over 500 times, which works out around 93Mbytes/sec.
If a PC requires tuning to sustain 1.5Mbps you have major problems, or you need to upgrade your 286. 1.5Mbps is extremely slow in computer terms. The PCI bus has over 50 times that capacity.
300-900kbps is most definitely not a real T1. He has likely oversold his upstream. We have a real qwest T1 at work, and most file downloads are in the 130-170KBytes/s range.
Extra hops add latency too. In any case, the salesman is probably mostly correct.
But, it still may be a better deal for you to go with the local guy. You need to ask yourself, are you getting the T1 and plan to use 100% of it bandwidth wise, or are you getting it for the freedom that comes with a T1? If it is more the latter, the local ISP might be a good deal.
You know, I always wonder how someone gets the job of "commentator". I often hear them with their 60 second or 2 minute pieces on NPR, and I wonder... Did they apply for the job? Does one become an apprentice commentator first? Do you just start bitching a lot and being nostalgic, and then people assume you are a commentator, and the rest follows?
I would have gotten a signed waiver that said that the employee signing it was representing that he had authority to allow such a thing. This guy will be lucky if he isn't sued.
since any normal linux user will use a recent Mozilla (or galeon & co) or Konqueror.
Oh, yeah, no normal user would want to use a browser under 100 megs.
I use Opera, lots of other Linux users do also.
How about spend $300 and get an x86 machine that is 4 times faster than that boat anchor that cost $10,000.
Terminal velocity is not a constant, it depends on drag.
The market does not demand such a thing, if it did, C3 would be flying off the shelves.
Electricity to run the computer for 8 hours a day is negligible. Even if the faster chip only saves 10 minutes of employee time, for most reasonable salaries, it was worth it. Employees are too expensive and electricity is too cheap for people to be worried about 50 watts here and there.
Your "friend" eh?
0 0- 25.html
I once heard of an individual who tried this and burped for an extended period time, an involuntary response that rid his body of gas produced as the nitrogen changed phases. He was concerned that he might burp long enough to be harmed by the lack of oxygen. That sounds more plausible to me. But I can believe portions of the digestive system might be frozen by the liquid nitrogen."
http://www.darwinawards.com/personal/personal20
When I can buy a dual athlon board with PCI-X for less than $500, then it will be enough. It doesn't help until it becomes available, and cards are being made for it. It's possible this standard will be completely passed over if it doesn't get adoption quickly enough, in 5 or 10 years 1GB per second may seem slow.
As you said, you burn anything critical on CDR. Those are your backups, not the RAID1.
RAID1/3/5 protects against one thing only, physical hard disk failure. Backups protect not only against hard disk failure, but user error, viruses, crackers, kernel goes insane, software bugs, etc..
If your filesystem module, for example, goes nuts and corrupts your filesystem, RAID1 is no help, it is the same with the other cases I named, and many others. The RAID will happily corrupt/delete the mirror at the same time.
Redundancy in RAID is useful, but as I said, never a substitute for backups, it only protects against one small class of data loss.
This is a trouble, I had to replicate a server and send it offsite, putting the 600GB on it took two days at 100Mbit speeds. Even at the full speed of the RAID, 40MB/Sec, it takes several hours, PCI maxes around 150MB/sec for 64bit in the real world, assumedly as faster controllers come out, 64/66 could be pushed a little higher than that, since there is some room there before you hit the theoritical limit.
We need to take all this AGP technology and quit wasting it on toys for gamer kids, and apply it to a more generalized bus technology.
RAID1 is never a substitute for backup, ever.
Don't forget a great rant by Denis Leary.
As I posted to another message, we are currently phasing out tape in favor of keeping many copies of data on various RAIDs for backup. You are correct, I'd call it a "backup crisis".
The biggest helical scan 200GB tapes are very very expensive compared to hard disk prices. We have over 4TB of disk space at work, most of it is redundancy (not counting RAID redundancy), but we do have almost 1TB of live data.
I've resorted to creative rsyncing for main backups, the Macs use retrospect (we are going to soon target that to a hard disk rather than tape), Veritas and the 1TB tape robot are still running, but too slow and cumbersome to be practical (if we ever needed to restore the full 1TB of data off that thing it would take weeks).
And really, who do we have to blame? I'd look at the MPAA... who has the most to lose from large removable media?
Scary, since we are phasing out most of the tape at work in favor of hard disks. Tape is just more expensive per GB these days, slower, less flexible, arguably less reliable.
Of course, if we are going to get bit rot in 5 years, then tape isn't looking so bad. At least we keep all important data in triplicate, or more.
I don't see anyone else posting this... It's true. If you want to run an OS from a rape-you-in-the-ass company that has no respect for its customers, you better not do it with my goddamn medical data. MS products are not fit for important uses. Running personal web pages from MS products is probably OK, but for any actual business use you need a real OS.
Yeah, what would be even neater is some sort of Congestion Notification. It could be Explicitely sent when there was a congested area.
I have a feeling that it might prevent access to sites running some older, buggy versions of firewalls though.
Oh well, one could always dream.
How do you understand that an encoder should pay, but not the decoder? That makes no sense. That's like saying people should pay to create ASCII text files, but not to read them. Software is software, and software patents are stupid no matter how they are applied.
Ross died of sudden adult death syndrome after a basketball match.
Oh come on! "Sudden adult death syndrome"?!!? WTF! This whole thing seems to be the same as the GHB scare... a bunch of lies and propaganda, funded by incumbent industries to crush disruptive products.
From what I remember of CNBC interviews with the startups that were trying this stuff years ago, apparently they had trouble segmenting the signal. Big chunks of the grid are electromagnetically coupled at any one time, so the difficulty was making the segments small enough to be able to provide meaningful bandwidth.
IIRC, of course, it's been a while.