I've voted in elections where only 25% of registered voters showed up. That means my vote was worth 4 votes. Assuming you're registered, my voice had 4 times the clout because people like you didn't bother voting. But that's only counting registered, not *eligible* voters who never bothered to even register! In that case, my vote was worth much more than 4.
When you don't vote, my voice gets heard even more, and I'm more likely to get what I want. You may agree with my views, but what if you don't?
I have owned domains for around 8 years now, and have used WHOIS even longer. Sure there is fake information out there, but legitimate domains will have working contact information 99.99% of the time. These days, I've mainly been looking up information for abuse contacts at large e-mail provider domains, but generally it's been useful just to find some sort of live contact for website problems, or any other failed or crippled service.
Back in '99, I planned ahead before I registered my domains. I rented a P.O. box. When I registered, I gave my P.O. box number and the telephone number of the company I worked for. The work number doesn't give complete privacy, but at least it wasn't my residential line. With the advent of Skype, I bought a Skype-In telephone number with voicemail. Now I don't even have to worry about disconnected or company phone numbers. The e-mail address for the domain contact is a special one I set aside. All of my information is correct, and all of my personal details are hidden. I don't care who sees my name.
No one has ever had a problem with my contact details. My e-mail and box addresses have always been live, and, now my telephone number is also live and direct. I may not answer a message right away, but I'll get it eventually. And I don't really care about spam. My domain contact's inbox is not linked to my personal inbox, so I don't see it everyday, but it's not like I'm going to run out of space. Deleting is easy too. As far as my P.O. box address, I hardly ever get junk mail. I have more problems getting the post office to stop delivering mail addressed to the guy who had the box several years ago. And now with the new Skype number, I can just deal with the voicemail like e-mail.
For those who say they couldn't afford a P.O. box (and now a voicemail number), I say they can't afford their own vanity domain. And today some registrars even offer private registration for a small fee. If you want privacy, there are options. Sure it will cost a bit extra, but why buy a house if you can't afford to maintain it? For all the rest, free e-mail accounts are being offered. And if you're running a business, there's no excuse for not having correct information.
Those who are complaining about the privacy of WHOIS are just complaining. The root of the problem lies elsewhere.
Yahoo has been junking all e-mail from my domain. Yet, my domain has been around since '99, has an SPF record, and has not been on a spam blacklist ever. I don't run any lists, and usually these e-mails are only directed at one recipient.
When I contacted Yahoo, I was referred to a broken web form that supposedly would direct me to a place where I could whitelist my domain, or at least make it less spammy-looking to Yahoo. Upon further attempts to reach them, I only received automated responses, but no answers to my questions.
I am not the only one who has had this problem sending e-mail to Yahoo accounts. Ironically, just Google for all the discussions on how Yahoo doesn't care.
Sending e-mail to GMail accounts works just fine for me. None of my messages show up in the spam folder. This is an indicator that the problem lies with Yahoo, and not with my domain.
It's easy to delete a domain. Delete the zone file on the primary name server and reload the daemon. That's only two steps, and it would really ruin your day if you didn't keep a backup on hand. Why would you think it's not hard to do?
Huh? I never slammed any SAN islands. You posted about redundancy with more than a dual fabric. If you are talking about SAN islands, you need to get your terminology straight. You also need to understand that if you have a SAN island, and there isn't more than one path (like through a dual fabric setup), like you say, you're taking down your entire BU anyway as you have no redundancy.
SAN islands can be dual fabric, as well as single fabric. I'm not sure you're grokking my posts.
Of course they can easily have more than 2 fabrics. They can have 3, 7, 12 redundant fabrics, if they were that paranoid. I never said it wasn't possible.
But you're just picking and choosing and making up strawmen now. If you have three fabrics, and one goes down because you just botched the upgrade, and the other goes down because a circuit breaker blows, well, now you're back to one fabric and vulnerable again. You're also totally flipping insane if you're worried about this, because the power company can have an extended outage or a maintenance person can short circuit the building with a new power line cut-in for a UPS. So, now you're saying you need three separate power companies, with three separate connections, and three separate UPS systems capable of running the whole building? What's the risk/return ratio on this?
Just as a simplistic example, director switches have dual power supplies. When one fails, it's reduced to running on one. Do you now want three redundant power supplies to guard against the infinitesimal risk (assuming you've designed properly) that the other will fail? You've just upped the cost, which will probably be more than 50%.
If you had read some of my other posts in a parallel thread, you'd see that I mentioned having redundant datacenters in different geographical locations.
You say that SAN design should only be directed by business need, and I've never said it shouldn't. However, your argument is a vendor's dream scenario, and not dictated by sound logic.
Now you're making the argument that these companies need redundancy but are too cheap to spring for it, or the existing topology won't allow it because of the limitations of the design. No one can help that.
Fabric merges and other configuration commands can have their potential disruptions limited by using zoning and VSANs. But no one can help you if you decide to log on and reboot the switch.
And yes I do know about those multi-vendor switch fabrics. While no one vendor seems to follow the standards the same way, these differences are mapped out and are well understood by each vendor. If you're just dumping a new vendor's switch on the fabric without research and a plan, then no one can help that.
It seems the situations you describe are either brought about by unwillingness or being unable to spend the correct amount of money, or just a poorly designed SAN.
And to that I ask: How paranoid are you, and how much money do you have?
You also talk about copies of data as if a disk went bad, you'd lose the data. These storage arrays have multiple redundancies (RAIDs of VDisks which are RAIDs themselves) as well as having live replication capability to remote sites -- at which point you likely have a copy (or copies) of an entire datacenter in a different geographic location that is running as a hot spare.
Within a datacenter, you would not have more than dual fabrics. Your fabrics' switches will also be redundantly connected within themselves. And if you're killing an entire fabric with an upgrade, you're doing it wrong.
You'll also have service contracts with lockers of disks, switches, linecards, etc., *on site* with field technicians from the vendors on-call 24/7.
Fibre Channel installations are not like some small company's Ethernet LAN.
Zoning, Virtual SANs (VSANs), and Inter-VSAN Routing (IVR) solve that problem. And why would you need more than a dual fabric setup? Do you like buying expensive HBAs for your servers? Just what kind of hostile environment are you deploying into?
I've seen the current-gen fiber extremely abused and it still works. Ninety-degree turns, tight coils, stepped on constantly, no slack... (No, it's not ideal, and these conditions should be avoided.) But, for a short-haul connection of a few meters, you will likely not see any problems unless you snap the fiber in half.
Hi, what state do you live in?...Well, it has everything to do with it.
Let's pretend the President is elected by a direct popular vote. Now say you live in North Dakota. How would you feel that New York, California, and Florida are consistently electing the President (based on population), and it doesn't, and won't ever matter, whom North Dakota votes for. In this scenario, it's always the big cities who are electing the President.
Currently, every state gets a vote for each senator and a vote for each congressional district. For you in North Dakota, this makes your state's vote go from 0 to 3. You should be able to see what effect this has on candidates (and Presidents) paying attention to you. A candidate just can't simply campaign in a few populous states, as he/she will need the other states (maybe not individually, but in aggregate) to win the election.
Most people that I've seen complain usually live in a populous state, in a large urban area, and are not taking what I said above into account. Usually when I explain it this way, at the very least their argument becomes more focused on refinements rather than scrapping the system.
Personally, I believe that if one wants to scrap the Electoral College, that person inherently wants the bicameral Congress scrapped too.
A spokeswoman for contractor SRA International Inc., where the AP found a document the Defense Department said could let hackers access military computer networks, said the company wasn't concerned because the unclassified file was on an FTP site that's not indexed by Internet search engines. "The only way you could find it is by an awful lot of investigation," said SRA spokeswoman Laura Luke.
a lot of your criticisms are significantly reduced when using Blade Servers instead of 1U pizza boxes
His comparison was 60 Dell rackmounts were equal in *price* to 1 AS/400. I believe my criticism was accurate, and not mere nitpicking.
Incidentally, if anything your analysis was generous with regards to physical connectivity.
I know. I was intentionally being generous because it shows that the point is valid even at a minimum configuration of the Dell rackmounts.
No need for 60 KVM ports, however, even with individual servers, just use the onboard DRAC management cards - albeit at the cost of another 60 switchports - then you also get handy features like virtual CDROM and floppy drives.
HP has iLO, and that uses an ethernet connection. The way you describe DRAC, it looks like the same thing. So, if the network goes away, how do you administer 60 rackmount servers? And, yes, I left out serial consoles, but not all 1U rackmounts (or most, if any?) have a serial console. If they did, then you'd need a 64 port terminal server which itself requires another ethernet port.
You'll pay a lot more than Dell, but HP's blades will let you get 64 8-core, 32G RAM blades into a single rack. For probably half the cost, Dell will get you 50 blades.
I've put this last, since your new example has veered away from comparing on the basis of price (60 for 1) into how many one can cram into a 48U rack *at any price*. HP's C-class will get you 64 half-height blades with 4 chassis in one rack, but is that feasible? Let's fully load it like you say. Each chassis has 6 13.8A @ 240V inputs which means 24 power drops requiring a total of 240V/360A service, just for that one rack. There are two ethernet switches @ 8 ports each on the chassis, plus another 4 ports for the OA/iLO (we're going for redundancy here, right?) for a total of 68 GigE drops. For fibre channel, since you're using the half-height blades, we'll only count 4 FC switches per chassis (only the full-heights have 3 HBAs) @ 8 outbound ports each multiplied by 4 chassis which equals 128 4Gb FC cable drops. By the way, (I know its anecdotal as I don't have an empirical BTU figure) those chassis run HOT when fully loaded and all are running IO.
So yeah, you can beat an AS/400 in raw computing power per dollar and per rackspace, but I never tried to disprove that. The argument I responded to was reliability per dollar and my counterpoint was to consider all the hidden costs in having true reliability.
I could buy something like 60 nice Dell rackmount servers for the same price and make a small Linux cluster of them. I'd end up with about 30 times the throughput, 100 times the storage, and 0% of the software cost. I cannot believe that the AS/400, solid as it is, has better uptimes than a 60-machine cluster (given that only about one tenth of those machines had to be online to exceed the AS/400's performance). Heck, for half the price, you could have two smaller clusters in geographically distinct locations with a high-speed link between them.
I don't doubt your numbers, but I believe you're leaving a lot out. Let's analyze this:
1) You equate an AS/400's price with 60 Dell rackmount servers. Although you didn't specify *which* Dell rackmount servers, assuming 1U each, this is two racks of Dell, minimum. The AS/400 takes about half a rack, but we'll just generalize to 1 rack. The Dells cost at least twice as much in floor space. Data center space: AS/400 wins.
2) Power. Minimum 60 AC-DC converting power supplies for Dell. How much is wasted in the conversion for the Dells? AS/400 wins. Minimum 60 power drops needed for Dell. How many does the AS/400 need? AS/400 wins.
3) Cooling. 1 rack for the AS/400 vs 2 or more for the Dells. AS/400 wins. I will guess that all the extra heat from having so many power supplies will just make the Dells more of a loser.
4) Network access. 60 individual NICs to configure for the Dells, and 60 different network sessions. AS/400 wins. 60 individual network drops for the Dells, and that would be at least a 48 port and a 12 port switch combo --maybe three 24 port switches? AS/400 wins again.
5) Storage access. You have 2.5 options. 60 individual disks for the Dells, 60 individual Fibre Channel HBAs for the Dells, or 60 saturated NICs running iSCSI for the Dells. AS/400 wins on either not needing 60 disks, or 60 HBAs. iSCSI could be a wash.
6) Console access. If the network fails, you will need to get onto the console. All 60 of them for the Dells; 1 for the AS/400. AS/400 wins. Good luck with 60 KVM ports. I recommend Avocent. If you can "get by" on one console at a time for the Dells, you'll need to pay someone to switch the cables, or physically be there yourself. AS/400 wins.
7) Sys admins. You only need 1 for the AS/400 -- and still have time for 59 more AS/400 servers. Good luck with the Dells -- you'll be bogged down with just that one cluster while the AS/400 admin is busy with the equivalent of your 60th (just an educated guess). AS/400 wins.
8) Fault tolerance: See #1-7. Simplification allows for easier problem resolution and time for other tasks. AS/400 wins.
9) Service contract: 60 machines for Dell vs 1 for IBM. Does the AS/400 support cost 60 times more for the same level? I'm guessing, no. AS/400 wins. Dell might not even offer service contracts for the machines you're comparing (hard to tell since you never mentioned which ones).
Hey, I'm not an IBM shill, nor am I short on Dell (I'm only expanding on your comparison). But you have to seriously consider the application here. And you will run into scalability issues with that many machines -- and scalability means money. Along the same lines, you might not want a whole lot of reliability, but you'll sure be spending the money you save (and likely more than that) for all the hidden costs I've listed above.
Weather Underground, because of the Weather Service Scandal is a suspect source of information. They did their best to cripple free updates from the national weather service and I'm still angry at them for it.
Why are you mad at Weather Underground? You're thinking of Accuweather and the other people. If you follow a link from that slashdot story you'll see that Weather Underground is not listed as a member of CWSA.
I have a hard time finding any references to this guy's employer anywhere in his blog.
I didn't. With one command I found out his name, e-mail address, street address, and phone number. I don't know what contact info he had listed before, but if he used his company's street address, phone number, or e-mail address, I could see a potential for a problem.
I'm guilty of that one. But that was back in '91, '92. I don't know where I picked it up from, but when one's typing in a real-time teleconference on a BBS, one tends to shortcut as many words as possible to save time (without going overboard). I didn't do it to be cute, and I never said "prolly" out loud.
To see people type it now is like hearing people speak Valley almost thirty years later.
There is no particular reason to censor TV just so YOU can subscribe.
Who reads comments anymore? Yours is the second to ascribe some motive to my statement other than the one I clearly stated -- which that I wasn't commenting on the article in general, just pointing out that the "everyone can receive broadcast TV" type of statement was uninformed. Those that would say that I've encountered an extremely small probability of this happening are grossly understating just how often this happens for people. Maybe people here need to travel to different regions and states and actually live on their own in cheap apartment communities for a few years (and I don't mean a college dorm).
And if you don't watch TV, then good for you. You're actually in a much smaller minority. My guess is you have a high speed Internet connection (which is mostly geographically determined), a computer, and you watch your movies and other TV type videos using them -- which means you're not really giving up TV at all; you're just changing the medium.
Unlike broadcast television, which is available to anyone with a TV and an antenna, people subscribe to and pay for cable/satellite.
The author has never lived in a concrete apartment building with nothing but cable available. When *I* lived in such places (and a few others that had bad broadcast reception for other reasons), I had the option of not subscribing, which meant absolutely no TV, or maybe a couple of snowy channels.
I'm not commenting on the article in general. I just thought that particular statement was ill-informed.
Slashdot Mirror
You're going to hate me when I say this.
:)
I've voted in elections where only 25% of registered voters showed up. That means my vote was worth 4 votes. Assuming you're registered, my voice had 4 times the clout because people like you didn't bother voting. But that's only counting registered, not *eligible* voters who never bothered to even register! In that case, my vote was worth much more than 4.
When you don't vote, my voice gets heard even more, and I'm more likely to get what I want. You may agree with my views, but what if you don't?
See? You hate me now.
I have owned domains for around 8 years now, and have used WHOIS even longer. Sure there is fake information out there, but legitimate domains will have working contact information 99.99% of the time. These days, I've mainly been looking up information for abuse contacts at large e-mail provider domains, but generally it's been useful just to find some sort of live contact for website problems, or any other failed or crippled service.
Back in '99, I planned ahead before I registered my domains. I rented a P.O. box. When I registered, I gave my P.O. box number and the telephone number of the company I worked for. The work number doesn't give complete privacy, but at least it wasn't my residential line. With the advent of Skype, I bought a Skype-In telephone number with voicemail. Now I don't even have to worry about disconnected or company phone numbers. The e-mail address for the domain contact is a special one I set aside. All of my information is correct, and all of my personal details are hidden. I don't care who sees my name.
No one has ever had a problem with my contact details. My e-mail and box addresses have always been live, and, now my telephone number is also live and direct. I may not answer a message right away, but I'll get it eventually. And I don't really care about spam. My domain contact's inbox is not linked to my personal inbox, so I don't see it everyday, but it's not like I'm going to run out of space. Deleting is easy too. As far as my P.O. box address, I hardly ever get junk mail. I have more problems getting the post office to stop delivering mail addressed to the guy who had the box several years ago. And now with the new Skype number, I can just deal with the voicemail like e-mail.
For those who say they couldn't afford a P.O. box (and now a voicemail number), I say they can't afford their own vanity domain. And today some registrars even offer private registration for a small fee. If you want privacy, there are options. Sure it will cost a bit extra, but why buy a house if you can't afford to maintain it? For all the rest, free e-mail accounts are being offered. And if you're running a business, there's no excuse for not having correct information.
Those who are complaining about the privacy of WHOIS are just complaining. The root of the problem lies elsewhere.
Yahoo has been junking all e-mail from my domain. Yet, my domain has been around since '99, has an SPF record, and has not been on a spam blacklist ever. I don't run any lists, and usually these e-mails are only directed at one recipient.
When I contacted Yahoo, I was referred to a broken web form that supposedly would direct me to a place where I could whitelist my domain, or at least make it less spammy-looking to Yahoo. Upon further attempts to reach them, I only received automated responses, but no answers to my questions.
I am not the only one who has had this problem sending e-mail to Yahoo accounts. Ironically, just Google for all the discussions on how Yahoo doesn't care.
Sending e-mail to GMail accounts works just fine for me. None of my messages show up in the spam folder. This is an indicator that the problem lies with Yahoo, and not with my domain.
It's easy to delete a domain. Delete the zone file on the primary name server and reload the daemon. That's only two steps, and it would really ruin your day if you didn't keep a backup on hand. Why would you think it's not hard to do?
Huh? I never slammed any SAN islands. You posted about redundancy with more than a dual fabric. If you are talking about SAN islands, you need to get your terminology straight. You also need to understand that if you have a SAN island, and there isn't more than one path (like through a dual fabric setup), like you say, you're taking down your entire BU anyway as you have no redundancy.
SAN islands can be dual fabric, as well as single fabric. I'm not sure you're grokking my posts.
Of course they can easily have more than 2 fabrics. They can have 3, 7, 12 redundant fabrics, if they were that paranoid. I never said it wasn't possible.
But you're just picking and choosing and making up strawmen now. If you have three fabrics, and one goes down because you just botched the upgrade, and the other goes down because a circuit breaker blows, well, now you're back to one fabric and vulnerable again. You're also totally flipping insane if you're worried about this, because the power company can have an extended outage or a maintenance person can short circuit the building with a new power line cut-in for a UPS. So, now you're saying you need three separate power companies, with three separate connections, and three separate UPS systems capable of running the whole building? What's the risk/return ratio on this?
Just as a simplistic example, director switches have dual power supplies. When one fails, it's reduced to running on one. Do you now want three redundant power supplies to guard against the infinitesimal risk (assuming you've designed properly) that the other will fail? You've just upped the cost, which will probably be more than 50%.
If you had read some of my other posts in a parallel thread, you'd see that I mentioned having redundant datacenters in different geographical locations.
You say that SAN design should only be directed by business need, and I've never said it shouldn't. However, your argument is a vendor's dream scenario, and not dictated by sound logic.
Now you're making the argument that these companies need redundancy but are too cheap to spring for it, or the existing topology won't allow it because of the limitations of the design. No one can help that.
Fabric merges and other configuration commands can have their potential disruptions limited by using zoning and VSANs. But no one can help you if you decide to log on and reboot the switch.
And yes I do know about those multi-vendor switch fabrics. While no one vendor seems to follow the standards the same way, these differences are mapped out and are well understood by each vendor. If you're just dumping a new vendor's switch on the fabric without research and a plan, then no one can help that.
It seems the situations you describe are either brought about by unwillingness or being unable to spend the correct amount of money, or just a poorly designed SAN.
And to that I ask: How paranoid are you, and how much money do you have?
You also talk about copies of data as if a disk went bad, you'd lose the data. These storage arrays have multiple redundancies (RAIDs of VDisks which are RAIDs themselves) as well as having live replication capability to remote sites -- at which point you likely have a copy (or copies) of an entire datacenter in a different geographic location that is running as a hot spare.
Within a datacenter, you would not have more than dual fabrics. Your fabrics' switches will also be redundantly connected within themselves. And if you're killing an entire fabric with an upgrade, you're doing it wrong.
You'll also have service contracts with lockers of disks, switches, linecards, etc., *on site* with field technicians from the vendors on-call 24/7.
Fibre Channel installations are not like some small company's Ethernet LAN.
Zoning, Virtual SANs (VSANs), and Inter-VSAN Routing (IVR) solve that problem. And why would you need more than a dual fabric setup? Do you like buying expensive HBAs for your servers? Just what kind of hostile environment are you deploying into?
Do tell. I'm curious.
And while we're at it, "people" did not apply to women or those who did not own land. Originally, only male landowners were allowed to vote.
If anything, that book would have lessened any scrutiny (as it was banned in many Islamic countries, and the author received death threats from Iran).
You might as well have been flashing around the King James Bible.
I've seen the current-gen fiber extremely abused and it still works. Ninety-degree turns, tight coils, stepped on constantly, no slack... (No, it's not ideal, and these conditions should be avoided.) But, for a short-haul connection of a few meters, you will likely not see any problems unless you snap the fiber in half.
Use the solar panel for shade. Problem solved.
Hi, what state do you live in? ...Well, it has everything to do with it.
Let's pretend the President is elected by a direct popular vote. Now say you live in North Dakota. How would you feel that New York, California, and Florida are consistently electing the President (based on population), and it doesn't, and won't ever matter, whom North Dakota votes for. In this scenario, it's always the big cities who are electing the President.
Currently, every state gets a vote for each senator and a vote for each congressional district. For you in North Dakota, this makes your state's vote go from 0 to 3. You should be able to see what effect this has on candidates (and Presidents) paying attention to you. A candidate just can't simply campaign in a few populous states, as he/she will need the other states (maybe not individually, but in aggregate) to win the election.
Most people that I've seen complain usually live in a populous state, in a large urban area, and are not taking what I said above into account. Usually when I explain it this way, at the very least their argument becomes more focused on refinements rather than scrapping the system.
Personally, I believe that if one wants to scrap the Electoral College, that person inherently wants the bicameral Congress scrapped too.
Holy crap! I've been trying to find out the name of that movie for years. And I found it through reading slashdot comments....
Gopher... No one looks there!
Even HP doesn't use it any more.
They don't? Better tell HP that.
a lot of your criticisms are significantly reduced when using Blade Servers instead of 1U pizza boxes
His comparison was 60 Dell rackmounts were equal in *price* to 1 AS/400. I believe my criticism was accurate, and not mere nitpicking.
Incidentally, if anything your analysis was generous with regards to physical connectivity.
I know. I was intentionally being generous because it shows that the point is valid even at a minimum configuration of the Dell rackmounts.
No need for 60 KVM ports, however, even with individual servers, just use the onboard DRAC management cards - albeit at the cost of another 60 switchports - then you also get handy features like virtual CDROM and floppy drives.
HP has iLO, and that uses an ethernet connection. The way you describe DRAC, it looks like the same thing. So, if the network goes away, how do you administer 60 rackmount servers? And, yes, I left out serial consoles, but not all 1U rackmounts (or most, if any?) have a serial console. If they did, then you'd need a 64 port terminal server which itself requires another ethernet port.
You'll pay a lot more than Dell, but HP's blades will let you get 64 8-core, 32G RAM blades into a single rack. For probably half the cost, Dell will get you 50 blades.
I've put this last, since your new example has veered away from comparing on the basis of price (60 for 1) into how many one can cram into a 48U rack *at any price*. HP's C-class will get you 64 half-height blades with 4 chassis in one rack, but is that feasible? Let's fully load it like you say. Each chassis has 6 13.8A @ 240V inputs which means 24 power drops requiring a total of 240V/360A service, just for that one rack. There are two ethernet switches @ 8 ports each on the chassis, plus another 4 ports for the OA/iLO (we're going for redundancy here, right?) for a total of 68 GigE drops. For fibre channel, since you're using the half-height blades, we'll only count 4 FC switches per chassis (only the full-heights have 3 HBAs) @ 8 outbound ports each multiplied by 4 chassis which equals 128 4Gb FC cable drops. By the way, (I know its anecdotal as I don't have an empirical BTU figure) those chassis run HOT when fully loaded and all are running IO.
So yeah, you can beat an AS/400 in raw computing power per dollar and per rackspace, but I never tried to disprove that. The argument I responded to was reliability per dollar and my counterpoint was to consider all the hidden costs in having true reliability.
The meter is in the pump, not the nozzle. It's your gas that's in the hose; you paid for it.
I could buy something like 60 nice Dell rackmount servers for the same price and make a small Linux cluster of them. I'd end up with about 30 times the throughput, 100 times the storage, and 0% of the software cost. I cannot believe that the AS/400, solid as it is, has better uptimes than a 60-machine cluster (given that only about one tenth of those machines had to be online to exceed the AS/400's performance). Heck, for half the price, you could have two smaller clusters in geographically distinct locations with a high-speed link between them.
I don't doubt your numbers, but I believe you're leaving a lot out. Let's analyze this:
1) You equate an AS/400's price with 60 Dell rackmount servers. Although you didn't specify *which* Dell rackmount servers, assuming 1U each, this is two racks of Dell, minimum. The AS/400 takes about half a rack, but we'll just generalize to 1 rack. The Dells cost at least twice as much in floor space. Data center space: AS/400 wins.
2) Power. Minimum 60 AC-DC converting power supplies for Dell. How much is wasted in the conversion for the Dells? AS/400 wins. Minimum 60 power drops needed for Dell. How many does the AS/400 need? AS/400 wins.
3) Cooling. 1 rack for the AS/400 vs 2 or more for the Dells. AS/400 wins. I will guess that all the extra heat from having so many power supplies will just make the Dells more of a loser.
4) Network access. 60 individual NICs to configure for the Dells, and 60 different network sessions. AS/400 wins. 60 individual network drops for the Dells, and that would be at least a 48 port and a 12 port switch combo --maybe three 24 port switches? AS/400 wins again.
5) Storage access. You have 2.5 options. 60 individual disks for the Dells, 60 individual Fibre Channel HBAs for the Dells, or 60 saturated NICs running iSCSI for the Dells. AS/400 wins on either not needing 60 disks, or 60 HBAs. iSCSI could be a wash.
6) Console access. If the network fails, you will need to get onto the console. All 60 of them for the Dells; 1 for the AS/400. AS/400 wins. Good luck with 60 KVM ports. I recommend Avocent. If you can "get by" on one console at a time for the Dells, you'll need to pay someone to switch the cables, or physically be there yourself. AS/400 wins.
7) Sys admins. You only need 1 for the AS/400 -- and still have time for 59 more AS/400 servers. Good luck with the Dells -- you'll be bogged down with just that one cluster while the AS/400 admin is busy with the equivalent of your 60th (just an educated guess). AS/400 wins.
8) Fault tolerance: See #1-7. Simplification allows for easier problem resolution and time for other tasks. AS/400 wins.
9) Service contract: 60 machines for Dell vs 1 for IBM. Does the AS/400 support cost 60 times more for the same level? I'm guessing, no. AS/400 wins. Dell might not even offer service contracts for the machines you're comparing (hard to tell since you never mentioned which ones).
Hey, I'm not an IBM shill, nor am I short on Dell (I'm only expanding on your comparison). But you have to seriously consider the application here. And you will run into scalability issues with that many machines -- and scalability means money. Along the same lines, you might not want a whole lot of reliability, but you'll sure be spending the money you save (and likely more than that) for all the hidden costs I've listed above.
Weather Underground, because of the Weather Service Scandal is a suspect source of information. They did their best to cripple free updates from the national weather service and I'm still angry at them for it.
Why are you mad at Weather Underground? You're thinking of Accuweather and the other people. If you follow a link from that slashdot story you'll see that Weather Underground is not listed as a member of CWSA.
I have a hard time finding any references to this guy's employer anywhere in his blog.
I didn't. With one command I found out his name, e-mail address, street address, and phone number. I don't know what contact info he had listed before, but if he used his company's street address, phone number, or e-mail address, I could see a potential for a problem.
I'm guilty of that one. But that was back in '91, '92. I don't know where I picked it up from, but when one's typing in a real-time teleconference on a BBS, one tends to shortcut as many words as possible to save time (without going overboard). I didn't do it to be cute, and I never said "prolly" out loud.
To see people type it now is like hearing people speak Valley almost thirty years later.
There is no particular reason to censor TV just so YOU can subscribe.
Who reads comments anymore? Yours is the second to ascribe some motive to my statement other than the one I clearly stated -- which that I wasn't commenting on the article in general, just pointing out that the "everyone can receive broadcast TV" type of statement was uninformed. Those that would say that I've encountered an extremely small probability of this happening are grossly understating just how often this happens for people. Maybe people here need to travel to different regions and states and actually live on their own in cheap apartment communities for a few years (and I don't mean a college dorm).
And if you don't watch TV, then good for you. You're actually in a much smaller minority. My guess is you have a high speed Internet connection (which is mostly geographically determined), a computer, and you watch your movies and other TV type videos using them -- which means you're not really giving up TV at all; you're just changing the medium.
Unlike broadcast television, which is available to anyone with a TV and an antenna, people subscribe to and pay for cable/satellite.
The author has never lived in a concrete apartment building with nothing but cable available. When *I* lived in such places (and a few others that had bad broadcast reception for other reasons), I had the option of not subscribing, which meant absolutely no TV, or maybe a couple of snowy channels.
I'm not commenting on the article in general. I just thought that particular statement was ill-informed.