If Bill is no longer "the face of Microsoft", perhaps we can change the Bill of Borg icon that's associated with Slashdot stories about Microsoft with one of Ballmer throwing a chair?
Back in the mid-90's a research student in a south-east Asian country decided to do a similar experiment. They started pinging 0.0.0.0, 0.0.0.1,...etc... When they got to 1.0.0.0 they took down BBN's network and upstream ISPs because the routers would negative-cache host routes of failed pings, thereby flushing out all the other working routes. My ISP got hosed when they got to 3.0.0.0 (Merit) since they were our customer. The attack moved up through 4.0.0.0 , then, back to 4.0.0.0 BBN, and up through other networks. On that day, the Internet suffered a rolling blackout because everyone was using Cisco routers affected by the same problem. When the source was identified and blocked, the problem stopped.
It's better to measure who is _using_ the Internet at central resources (root DNS servers, google, time.windows.com) rather than who can respond to a ping. Back when I was young, people didn't use NAT or firewalls and everything responded to a ping. Today, millions (billions?) of people don't really have public address space, and are separated from the IPv4 Internet by one or more levels of NAT or proxy servers. Clusters of web servers are mostly virtualized behind a single address served by load balancers and/or firewalls. A "ping" census is worth less today compared to prior to the rise of NAT firewalls in the late 90's. It's still interesting, but not at all accurate.
Aside: When ISPs and corporations are forced to pay equitably for the addresses (and routes!) they use, the IPv4 "crisis" will solve itself.
Ummm, OC-192 is 9.6Gbps I think they are a little shy of the speed record. Maybe I missed something.
Within a data center or a metro area, it's commercially viable to pump tens of gigabits per second of bits from point A to point B using many parallel fiber circuits between the two locations. What makes the Internet2 land speed record (http://www.internet2.edu/lsr/) interesting is adding distance to the problem by multiplying the speed times the distance. The unit of measurement they use is "terabit-meters per second" (Tbmps?). The current record is 272,400 Tbmps, or ~9Gbps over 30000km (1km=1000000m). The transfer rate is really a function of 1) latency adjustments in the data transfer protocol, 2) the minimum transfer speed capable between all points on the network (currently OC192=10Gbps), and 3) the speed of the sending and receiving computers. While OC192 might theoretically be 9.6Gbps, getting the various vendors
switches on different continents to all send packets at line speed for a long period of time with minimal packet overhead can be challenging.
What makes this pointless, though, is that the sending and receiving equipment is in the same location. In their documentation they send the bits from a computer in Tokyo through Chicago through Amsterdam and back through Seattle to the same lab in Tokyo. It would be much easier to put a 10GigE fiber between the two machines, but that's not he "point" of the exercise.
Someone has to pay for this. Usually its the country's taxpayers or a company's stockholders.
I'd much rather see benchmarks for transferring N terabytes of real data from one site with lots of disks to another far-away site with lots of disks. Real companies can use that data for pontificating disaster recovery and content/database replication technologies. I'd reckon that Google can beat the multiple stream Internet2 LSR any day they want by pumping petabytes of data between its data centers over multiple 10GigE backbones. Andy Tanenbaum's (or Hal Stern's?) station wagon full of tapes is also a fine competitor.
I don't see any reference to "one click" nor "Amazon" nor anything remotely technical.
I wonder if this story is partly a hoax. I can't imagine that lawyers would actually send someone to look at a document like that. If so, perhaps they're just seeing if the patent attorneys even pay attention?
I was fortunate enough to get a tour of the google campus once and i asked them about thier server and what they run,... The dead part simply get tossed in the trash
It must have been some time ago. Dumping electronics in the trash is now illegal in Santa Clara, California.
If I had the buying power of Google, I would make my vendors replace and recycle any and all dead components for me at no cost to me. Forced recycling might also give vendors more incentive to make their components last longer.
The following applies to my personal experience in California's most recent election using electonic Diebold machines with register-roll paper trail.
and if a person is completely blind, how in the name of whatever Deity you believe in is a touch screen that they can't see going to help?
A Diebold HAVA-capable machine audibly reads the whole ballot and choices to the blind voter using a headset. The voter presses their choices into a keypad with a dimple on the #5 key. While the process takes much much longer than someone who can use the touchscreen, it is designed to work for someone who cannot see at all. They can go backwards, forwards, change their vote selection, review their ballot, change the speed of the speaker to be faster or slower, etc.
and just how disabled are you if you can't put an X in a 1.25" circle? even if you have tourette's or something and you screw up your ballot, you can get another one as many times as you need to until you get it right.
There's "legally blind" (someone's grandma), and then there's totally blind (nothing but black). The latter can't vote with an "X" without someone to "help" them. The latter still has a right to vote (with privacy, without "help").
faster? results before you go to bed (well, results when you wake up if you live in the east) isn't fast enough for you?
Electronic voting saves all poll workers about 30-45 minutes of time not having to count all of the paper ballots, so the results get to the central counting facility much sooner. It certainly isn't enough to make a huge difference to most people who get the results the next day. While there would be a huge media benefit and convenience to "modem in" preliminary results (get 95% reporting by 10pm), there aren't enough safeguards (yet) to prevent tampering with that process in CA.
FYI: Sony claimed 11 billion bits per square inch quite some time ago.
It's always good news when someone figures out how to store more bits into the same amount of space, and I'm sure that companies like IBM and Sony will keep pushing the limits.
I don't host anything of my own at work. Take a look at the Personal Co-location Registry. You'll find a bunch of inexpensive providers for your servers or apps.
Regardless of whether you use SCSI or SATA or EIDE or FC, you need to know and trust your RAID controller as if your job depended in it. Drives will fail. Some drives will fail more than others. As long as your RAID controller can recover quickly on failed drives, not lose any data, and can be monitored from your OS so that you can detect when a drive fails so that you can replace your hot-spare ASAP, you'll be fine.
Consider how long it will take to rebuild a 400GB drive as part of a RAID5 array compared to a faster-spinning 73GB drive part of a RAID1+0 array. While your controller is rebuilding, your array is at risk for data loss if another drive fails.
If you cannot monitor your RAID controller form the operating system, you ma never know that a drive failed and rebuilt using the spare. You could go days or weeks without knowing that you have no spare available for the next failure.
A RAID controller without a cache battery is slower than one with a cache battery. If your RAID controller safely buffers writes, your operating system won't have to wait for the slow disks to write the data. Turn off write buffering on the hard drives. The write buffer on a hard drive is an area where data can get lost on a power failure that neither your operating system nor RAID controller can recover from.
Keep your disks cool and power-conditioned. Don't put your file server in an 80-degree closet. Don't just stick the drives in a cheap case that doesn't have proper airflow over the drives. If the data is important to you, you will make sure your file server is on a UPS protected from brown-outs, power spikes, idiot employees or even cleaning staff. If you can monitor your UPS from the file server, so much the better.
All of this is more important than SATA-vs-SCSI and applies to both.
When it comes to SATA vs SCSI, it is a matter of low-cost capacity versus performance. If you need high-RPM drives to support low latency times for random read access, you're stuck with SCSI. If just having enough capacity for your data-hungry staff is all you need, then SATA might be a better fit for your budget.
I've been deployed in Iraq for the last 9 months...
BTW: We hope you get to look forward to something as mundane as Windows viruses real soon now. In case anyone hasn't mentioned it yet, "Thank you for the job you're doing."
Back in the early nineties I managed domain registrations at a relatively small ISP. This was back when people filled out registration forms using a text editor and emailed them in. I'd do about 10 a day, and since they were free, I thought I'd slip another one in to see if it'd work - "z.com". I thought, "Well, they've registered x.org already, so why not?" To my surprise, I got it. I also took a stab at registering a one-letter UUCP name - "z". I got that too. So you could email me at eric@z.com or even z!z (shortest email address). Joy!
Someone had this idea that serving the com/net/org domains was going to be too large for any single nameserver to handle and that we should perhaps start hashing domains. If a resolver was going to lookup domain.com, it would look it up first at domain.d.com, and they'd distribute the letters out to multiple servers. Moore's law and load balancers proved that you could create some beefy root-server clusters, and the idea never materialized. Besides x.com, q.com, z.com, x.org were already in use (and I think one other), and I'm sure no one wanted to give them back.
When domain prospecting came along, I had many offers for real money to buy my domain. I turned down many of them, but one day someone made me a good enough offer. My elite-ness was gone, but I used the proceeds as a down-payment on my first house.
Portals were all the rage, and my buyers tried to turn z.com into one. The best part about this one-letter domain at the time was that if you simply entered the letter "z" in Mozilla or Explorer, you'd go straight to the z.com page. The project didn't seem to go anywhere, though. Those people sold it to some people from IdeaLab (founders of Goto.com). I don't think they ever thought of anything to do with it, so the domain stayed in limbo for a while. One day at the movies, I saw an ad/trailer for the new Nissan 350Z sports sedan. For more information, you had to go to "z.com". Surely, I suspect those guys at Idealab got alot more money than me, but at least the domain was being used for something useful now.
A single-letter domain without good branding and advertising isn't worth much, and perhaps the people at ICANN are seeing that they're now on the falling side of the value curve. Can anyone thing of a reason why new domains would be released, aside from money? I could only hope that ICANN, a non-profit organization, would use the proceeds to help fund IETF (ietf.org) and DNS infrastructure research, but they'll probably go to fund more "meetings" in far-off places.
If you can grab a hold of a person's email account (sniff their non-SSL web/pop/imap login and password), you have access to:
- eBay login (buy stuff online)
- paypal account (get access to a credit card or back account for paying for things)
- bank account (look at all of their recent transactions - do bill-pay)
- Amazon (order some books and gear)
- most eCommerce sites (buy more stuff)
- domain names (take their domain names and web sites and blogs)
- tax returns (get them from their Intuit login - - - get SSN here?)
- cell phone (see who they call; order yourself a phone)
... because they all email your password to your registered email account when you claim you lost your password or account information.
Make sure your email provider offers encrypted (SSL) access to your email.
Has anyone tried this card with Linux or BSD yet? Something on an Anandtech web forum say it works just like any SATA disk (use fdisk, etc.), and the article shows some BIOS screen shots of the SATA disk, but I haven't seen anyone try it on a non-Windows operating system.
Here's is a link to some interesting test info to show how reliable the disk performs in different failure modes.
I want one for my mail spool, MySQL bin-logs,/tmp, Apache logs, small MySQL transaction-driven tables, and any other place where I want reliable fast random writes. If I could cut down the number of writes to disks, I could use slower disks (or get more use out of my current disks).
Slashdot Mirror
If Bill is no longer "the face of Microsoft", perhaps we can change the Bill of Borg icon that's associated with Slashdot stories about Microsoft with one of Ballmer throwing a chair?
Back in the mid-90's a research student in a south-east Asian country decided to do a similar experiment. They started pinging 0.0.0.0, 0.0.0.1, ...etc... When they got to 1.0.0.0 they took down BBN's network and upstream ISPs because the routers would negative-cache host routes of failed pings, thereby flushing out all the other working routes. My ISP got hosed when they got to 3.0.0.0 (Merit) since they were our customer. The attack moved up through 4.0.0.0 , then, back to 4.0.0.0 BBN, and up through other networks. On that day, the Internet suffered a rolling blackout because everyone was using Cisco routers affected by the same problem. When the source was identified and blocked, the problem stopped.
It's better to measure who is _using_ the Internet at central resources (root DNS servers, google, time.windows.com) rather than who can respond to a ping. Back when I was young, people didn't use NAT or firewalls and everything responded to a ping. Today, millions (billions?) of people don't really have public address space, and are separated from the IPv4 Internet by one or more levels of NAT or proxy servers. Clusters of web servers are mostly virtualized behind a single address served by load balancers and/or firewalls. A "ping" census is worth less today compared to prior to the rise of NAT firewalls in the late 90's. It's still interesting, but not at all accurate.
Aside: When ISPs and corporations are forced to pay equitably for the addresses (and routes!) they use, the IPv4 "crisis" will solve itself.
Ummm, OC-192 is 9.6Gbps I think they are a little shy of the speed record. Maybe I missed something.
Within a data center or a metro area, it's commercially viable to pump tens of gigabits per second of bits from point A to point B using many parallel fiber circuits between the two locations. What makes the Internet2 land speed record (http://www.internet2.edu/lsr/) interesting is adding distance to the problem by multiplying the speed times the distance. The unit of measurement they use is "terabit-meters per second" (Tbmps?). The current record is 272,400 Tbmps, or ~9Gbps over 30000km (1km=1000000m). The transfer rate is really a function of 1) latency adjustments in the data transfer protocol, 2) the minimum transfer speed capable between all points on the network (currently OC192=10Gbps), and 3) the speed of the sending and receiving computers. While OC192 might theoretically be 9.6Gbps, getting the various vendors
switches on different continents to all send packets at line speed for a long period of time with minimal packet overhead can be challenging.
What makes this pointless, though, is that the sending and receiving equipment is in the same location. In their documentation they send the bits from a computer in Tokyo through Chicago through Amsterdam and back through Seattle to the same lab in Tokyo. It would be much easier to put a 10GigE fiber between the two machines, but that's not he "point" of the exercise.
Someone has to pay for this. Usually its the country's taxpayers or a company's stockholders.
I'd much rather see benchmarks for transferring N terabytes of real data from one site with lots of disks to another far-away site with lots of disks. Real companies can use that data for pontificating disaster recovery and content/database replication technologies. I'd reckon that Google can beat the multiple stream Internet2 LSR any day they want by pumping petabytes of data between its data centers over multiple 10GigE backbones. Andy Tanenbaum's (or Hal Stern's?) station wagon full of tapes is also a fine competitor.
-ez
Mea culpa - mod me down "redundant". I should have clicked all of the links before posting.
Those who are curious what was on the "Norm" quote page was but are too lazy to type the URL by hand will enjoy this time saver:
d .compuserve.com/homepages/wildkingdom/normb.htm .
Click here or copy and paste http://web.archive.org/web/20010702004226/ourworl
I don't see any reference to "one click" nor "Amazon" nor anything remotely technical.
I wonder if this story is partly a hoax. I can't imagine that lawyers would actually send someone to look at a document like that. If so, perhaps they're just seeing if the patent attorneys even pay attention?
Some people analyzed weaknesses in the bill and made recommendations for changes....e d2HR811.pdf
http://electionarchive.org/ucvInfo/US/ChangesNeed
I was fortunate enough to get a tour of the google campus once and i asked them about thier server and what they run, ... The dead part simply get tossed in the trash
It must have been some time ago. Dumping electronics in the trash is now illegal in Santa Clara, California.
If I had the buying power of Google, I would make my vendors replace and recycle any and all dead components for me at no cost to me. Forced recycling might also give vendors more incentive to make their components last longer.
-ez
In the US, move your servers to Arizona or eastern Indiana or Hawaii or Alaska, so you don't have to deal with time zone changes anymore.
-ez
The following applies to my personal experience in California's most recent election using electonic Diebold machines with register-roll paper trail.
and if a person is completely blind, how in the name of whatever Deity you believe in is a touch screen that they can't see going to help?
A Diebold HAVA-capable machine audibly reads the whole ballot and choices to the blind voter using a headset. The voter presses their choices into a keypad with a dimple on the #5 key. While the process takes much much longer than someone who can use the touchscreen, it is designed to work for someone who cannot see at all. They can go backwards, forwards, change their vote selection, review their ballot, change the speed of the speaker to be faster or slower, etc.
and just how disabled are you if you can't put an X in a 1.25" circle? even if you have tourette's or something and you screw up your ballot, you can get another one as many times as you need to until you get it right.
There's "legally blind" (someone's grandma), and then there's totally blind (nothing but black). The latter can't vote with an "X" without someone to "help" them. The latter still has a right to vote (with privacy, without "help").
faster? results before you go to bed (well, results when you wake up if you live in the east) isn't fast enough for you?
Electronic voting saves all poll workers about 30-45 minutes of time not having to count all of the paper ballots, so the results get to the central counting facility much sooner. It certainly isn't enough to make a huge difference to most people who get the results the next day. While there would be a huge media benefit and convenience to "modem in" preliminary results (get 95% reporting by 10pm), there aren't enough safeguards (yet) to prevent tampering with that process in CA.
Every baby "goes wireless" when they get their cord cut at birth.
article
While Googling for "bits per square inch world record", I found a company claiming 515 gigabits per square inch. Here come the terabyte discs!
FYI: Sony claimed 11 billion bits per square inch quite some time ago.
It's always good news when someone figures out how to store more bits into the same amount of space, and I'm sure that companies like IBM and Sony will keep pushing the limits.
Get step-by-step instructions for contructing such a wallet as well as background information about RFID in consumer objects.
I'd love to be able to buy one at thinkgeek.com (they already sell a duct tape wallet).
Foiled again! Those Slashdot readers never would never have figured it out if it weren't for you darn snooping kids!
I don't host anything of my own at work. Take a look at the Personal Co-location Registry. You'll find a bunch of inexpensive providers for your servers or apps.
Ok, then can I choose Al Gore?
He invented the Internet, you know.
Regardless of whether you use SCSI or SATA or EIDE or FC, you need to know and trust your RAID controller as if your job depended in it. Drives will fail. Some drives will fail more than others. As long as your RAID controller can recover quickly on failed drives, not lose any data, and can be monitored from your OS so that you can detect when a drive fails so that you can replace your hot-spare ASAP, you'll be fine.
Consider how long it will take to rebuild a 400GB drive as part of a RAID5 array compared to a faster-spinning 73GB drive part of a RAID1+0 array. While your controller is rebuilding, your array is at risk for data loss if another drive fails.
If you cannot monitor your RAID controller form the operating system, you ma never know that a drive failed and rebuilt using the spare. You could go days or weeks without knowing that you have no spare available for the next failure.
A RAID controller without a cache battery is slower than one with a cache battery. If your RAID controller safely buffers writes, your operating system won't have to wait for the slow disks to write the data. Turn off write buffering on the hard drives. The write buffer on a hard drive is an area where data can get lost on a power failure that neither your operating system nor RAID controller can recover from.
Keep your disks cool and power-conditioned. Don't put your file server in an 80-degree closet. Don't just stick the drives in a cheap case that doesn't have proper airflow over the drives. If the data is important to you, you will make sure your file server is on a UPS protected from brown-outs, power spikes, idiot employees or even cleaning staff. If you can monitor your UPS from the file server, so much the better.
All of this is more important than SATA-vs-SCSI and applies to both.
When it comes to SATA vs SCSI, it is a matter of low-cost capacity versus performance. If you need high-RPM drives to support low latency times for random read access, you're stuck with SCSI. If just having enough capacity for your data-hungry staff is all you need, then SATA might be a better fit for your budget.
I've been deployed in Iraq for the last 9 months...
BTW: We hope you get to look forward to something as mundane as Windows viruses real soon now. In case anyone hasn't mentioned it yet, "Thank you for the job you're doing."
Back in the early nineties I managed domain registrations at a relatively small ISP. This was back when people filled out registration forms using a text editor and emailed them in. I'd do about 10 a day, and since they were free, I thought I'd slip another one in to see if it'd work - "z.com". I thought, "Well, they've registered x.org already, so why not?" To my surprise, I got it. I also took a stab at registering a one-letter UUCP name - "z". I got that too. So you could email me at eric@z.com or even z!z (shortest email address). Joy!
Someone had this idea that serving the com/net/org domains was going to be too large for any single nameserver to handle and that we should perhaps start hashing domains. If a resolver was going to lookup domain.com, it would look it up first at domain.d.com, and they'd distribute the letters out to multiple servers. Moore's law and load balancers proved that you could create some beefy root-server clusters, and the idea never materialized. Besides x.com, q.com, z.com, x.org were already in use (and I think one other), and I'm sure no one wanted to give them back.
When domain prospecting came along, I had many offers for real money to buy my domain. I turned down many of them, but one day someone made me a good enough offer. My elite-ness was gone, but I used the proceeds as a down-payment on my first house.
Portals were all the rage, and my buyers tried to turn z.com into one. The best part about this one-letter domain at the time was that if you simply entered the letter "z" in Mozilla or Explorer, you'd go straight to the z.com page. The project didn't seem to go anywhere, though. Those people sold it to some people from IdeaLab (founders of Goto.com). I don't think they ever thought of anything to do with it, so the domain stayed in limbo for a while. One day at the movies, I saw an ad/trailer for the new Nissan 350Z sports sedan. For more information, you had to go to "z.com". Surely, I suspect those guys at Idealab got alot more money than me, but at least the domain was being used for something useful now.
A single-letter domain without good branding and advertising isn't worth much, and perhaps the people at ICANN are seeing that they're now on the falling side of the value curve. Can anyone thing of a reason why new domains would be released, aside from money? I could only hope that ICANN, a non-profit organization, would use the proceeds to help fund IETF (ietf.org) and DNS infrastructure research, but they'll probably go to fund more "meetings" in far-off places.
-ez
If you can grab a hold of a person's email account (sniff their non-SSL web/pop/imap login and password), you have access to:
... because they all email your password to your registered email account when you claim you lost your password or account information.
- eBay login (buy stuff online)
- paypal account (get access to a credit card or back account for paying for things)
- bank account (look at all of their recent transactions - do bill-pay)
- Amazon (order some books and gear)
- most eCommerce sites (buy more stuff)
- domain names (take their domain names and web sites and blogs)
- tax returns (get them from their Intuit login - - - get SSN here?)
- cell phone (see who they call; order yourself a phone)
Make sure your email provider offers encrypted (SSL) access to your email.
-ez
Anyone notice which $15 million donor is missing from the Hall of Fellows awards? Too bad they already picked their 2005 winners.
Quick, cheap, right -- pick one.
I think "pick two" works here.
Quick & Cheap - it's not likely to work.
Quick & Right - it'll cost alot.
Cheap & Right - it'll take forever.
Suggest PPTP as an alternative.
Has anyone tried this card with Linux or BSD yet? Something on an Anandtech web forum say it works just like any SATA disk (use fdisk, etc.), and the article shows some BIOS screen shots of the SATA disk, but I haven't seen anyone try it on a non-Windows operating system.
/tmp, Apache logs, small MySQL transaction-driven tables, and any other place where I want reliable fast random writes. If I could cut down the number of writes to disks, I could use slower disks (or get more use out of my current disks).
Here's is a link to some interesting test info to show how reliable the disk performs in different failure modes.
I want one for my mail spool, MySQL bin-logs,