Hmm, well the interface for my brand new Brocade 5100 switch is a java app using java web start which requires the browser plugin AFAIK. Would you rather they have coded it as a Windows only app so you couldn't admin the switch from Linux?
Climbing Everest without oxygen is incredibly difficult, only 60 people of ~3,000 that have summited have done it without supplemental oxygen. 2% of people who have accomplished one of the toughest human feats have done it while making it even harder, that's a pretty elite group.
I'll tell you from personal experience that available oxygen for the brain has a HUGE impact on altitude sickness. While climbing Mt Whitney in California I got to high basecamp (12,000 ft) without any real problems, just a minor headache. We were there for a few hours, setting up tents, securing our food in a way the Marmit's couldn't get to it, and filter water. Around sundown we finally got to eating dinner, about half way through the meal I started vomiting violently and my headache went from a dull ache to something worse then most migraines I have had. In addition my lips turned blue and my eyes became bloodshot. All of this was due to the fact that my stomach was using oxygen to digest food which took away from the supply available to my brain. We quickly broke camp and headed down, once I got below 10,000ft I was fine. It turns out it was fortunate I became sick because on the way down we ran into a father with two boys who had been day hiking and had vastly underestimated their task. They had no provisions and only one light between the three of them. We helped them get down through the more technical areas to where they could safely walk the path with just the one light source.
An interesting factoid I just discovered, Georgia Moffett is the real life daughter of the fifth doctor Peter Davison. I guess it was a bit of an in joke to have her portray the doctor's daughter =)
Interesting factoid I just discovered, Georgia Moffett is the real life daughter of the fifth doctor Peter Davison! I guess it's kind of an in joke to have her portrayed as the doctor's daughter =)
Yep, over Thanksgiving we watched 32 year old wedding footage with my wife's grandmother and parents, the glimpses of her grandfather and great-grandfather brought up a lot of emotions for them all.
DLT is 24 years old and tapes written on 14 year old DLTIV tapes are still readable both physically and logically today (assuming tar format for backups). If you had tapes from 1989 onward you would need 2 drives to read every tape, something that can read DLTIII and a new SDLT 320 drive to read DLTIV onwards. It's not the cheapest solution for home backups but if you are serious about longterm archives it's about the only solution that doesn't require refreshing data.
SCSI has been replaced by SAS (Serial Attach SCSI). You won't find huge capacity SAS drives because most applications need IOPS and if you need more storage you add additional spindles which gives you more of both. There's also the problem with 1TB drive rebuild time, heck we're concerned enough about rebuild times on our 300GB fibrechannel drives that we are going with 5% hotspares to reduce the chances of a double fault causing data loss (despite a 1.5% annual failure rate for similar drives in our array).
If that's a troll it's a good one, but you can provide 7.5x more power over ethernet then the USB spec allows (15.4W vs 2.5W). 802.3at pushes things even further to 24W max.
Hmm, a crappy host based implementation at 5Gbps or a storage targeted, low everhead one at 3Gbps (possibly going to 6Gbps if the next gen spec gets extended to eSATA), I know which one I will use =)
have you ever seen DMA to memory ever work properly on consumer grade hardware anyway? Yeah, every day. PIO mode HDD's suck terribly, DMA is necessary to achieve decent performance.
Ok, so they are going to use 200KW for 10 minutes or about 34kWhrs. These guys make a system that is probably currently the best bet for power density vs orbital safety, the EEV version has 1306 Wh and weighs 15.6kg. At 80% discharge that's about 1KWhr. That means that you need ~530Kg of batteries, at $10K/lb that comes out to $12M to launch the batteries for this thing. I was going to suggest ultracapacitors but it turns out they suck for energy density, on the order of less than 6Wh/Kg! Reading on wikipedia (yeah, I know) Lithium Ion batteries with nanowires achieve almost 15kWHrs/Kg (.75 * 4,2000 mAh/g @ a nominal 3.6V per cell) so developing them would save almost the entire launch cost AND get us better batteries for all sorts of terrestrial applications!
Nope, cable co's are forced to carry the ATSC signal for the locals. My cableco does it the cheap way and just encapsulates them in clear QAM and puts em on the wire, not sure how the SDV folks are handling it. This is why I haven't upped to the digital+HD package (about $30/month more after all the receiver rentals) which only adds 14 non-local HD channels.
I know with Drivesavers you pay an analysis fee upfront, if they can't recover anything they refund it, if they can they will apply it to your recovery bill, if you choose not to pay for recovery they keep the fee to pay for the techs time. This is completely fair in my eyes as just getting to that stage may have involved swapping controllers or moving platters to a new enclosure.
Many BIOS's support timed power on events so it could go like this:
7:50a power on
8A-5P startup script enables profile with full power and dynamic clock throttle (AMD PowerNow/Cool'n'Quiet or Intel SpeedStep)
5P scheduled task switches power profile to mobile or custom
3A WOL from management server wakes up PC if updates are needed (I believe you can even do this with a scheduled task for standalone PC's)
4A previously changed profile puts PC back to sleep
You keep talking about codes, WHAT codes? The payload that activates the feature (at least in the Lenovo implementation) is a cryptographically signed message, there is no default code! It's just like Blackberries, a cryptographically signed message is received by the device and it initiates a wipe of the device, it the case of the Blackberry it wipes the RAM and flash areas, in the case of the Lenovo it wipes the storage keys from the TPM chip.
From what I've read it doesn't work, a scan with an MFT aware app still shows fragmentation afterwards. I haven't tested this myself as I figure Diskkeeper (the guys who wrote MS defrag) know enough that I trust them to defrag the MFT online.
Hmm, on second thought after doing some more research it MIGHT matter:
CRC is somewhat expensive. 200-300 MB/s on a modern CPU looking into ways to optimize
SSE4 will have a CRC instruction (any poly) linky(pdf)
Not sure what he means by CPU there, if he means core then it's not so bad (other than freaking expensive per core Oracle licensing), if that's per processor then that's very bad. Guess I might only be able to implement this on the new Shanghai beast we got last week or a new Corei7 Xeon system we will probably get next year once they are available.
Taking the HDD out gains you NOTHING, in theory it's already fully encrypted with 256 bit AES which is uncrackable by any currently known method. The idea is that there is only one real vulnerability in a TPM based system and that is the TPM chip's keystore and the databus that the TPM chip uses to talk to the CPU, if you erase the keystore and thus makes sure that both those pathways are neutralized there should be no possible way to retrieve the data off the disk. There's still the cooled RAM trick and possibly a trace of the key left in the disk controller's cache, but those are both VERY sophisticated attacks that have a very low chance of working even in lab conditions. Oh and I just thought of something, if the TPM keystore is wiped then the TPM trust web collapses and the machine should reboot thus flushing the key from ram.
Slashdot Mirror
Hmm, well the interface for my brand new Brocade 5100 switch is a java app using java web start which requires the browser plugin AFAIK. Would you rather they have coded it as a Windows only app so you couldn't admin the switch from Linux?
I'm already running computers with a quarter terabyte of ram you insensitive clod.
Seriously, the data warehouse server I'm implementing this week is an HP DL585 G5 with 4x AMD 8378 and 256GB of ram using 2x Brocade 815 8Gb HBA's =)
Climbing Everest without oxygen is incredibly difficult, only 60 people of ~3,000 that have summited have done it without supplemental oxygen. 2% of people who have accomplished one of the toughest human feats have done it while making it even harder, that's a pretty elite group.
Yeah war kept me from doing Kilamanjaro several years ago as well, that part of Africa sure has a lot of violence =(
Weather picking up quickly happens on ALL tall peaks, even in beautiful Hawaii you can die from a sudden snowstorm on the volcano's.
I'll tell you from personal experience that available oxygen for the brain has a HUGE impact on altitude sickness. While climbing Mt Whitney in California I got to high basecamp (12,000 ft) without any real problems, just a minor headache. We were there for a few hours, setting up tents, securing our food in a way the Marmit's couldn't get to it, and filter water. Around sundown we finally got to eating dinner, about half way through the meal I started vomiting violently and my headache went from a dull ache to something worse then most migraines I have had. In addition my lips turned blue and my eyes became bloodshot. All of this was due to the fact that my stomach was using oxygen to digest food which took away from the supply available to my brain. We quickly broke camp and headed down, once I got below 10,000ft I was fine. It turns out it was fortunate I became sick because on the way down we ran into a father with two boys who had been day hiking and had vastly underestimated their task. They had no provisions and only one light between the three of them. We helped them get down through the more technical areas to where they could safely walk the path with just the one light source.
An interesting factoid I just discovered, Georgia Moffett is the real life daughter of the fifth doctor Peter Davison. I guess it was a bit of an in joke to have her portray the doctor's daughter =)
Interesting factoid I just discovered, Georgia Moffett is the real life daughter of the fifth doctor Peter Davison! I guess it's kind of an in joke to have her portrayed as the doctor's daughter =)
Yep, over Thanksgiving we watched 32 year old wedding footage with my wife's grandmother and parents, the glimpses of her grandfather and great-grandfather brought up a lot of emotions for them all.
DLT is 24 years old and tapes written on 14 year old DLTIV tapes are still readable both physically and logically today (assuming tar format for backups). If you had tapes from 1989 onward you would need 2 drives to read every tape, something that can read DLTIII and a new SDLT 320 drive to read DLTIV onwards. It's not the cheapest solution for home backups but if you are serious about longterm archives it's about the only solution that doesn't require refreshing data.
SCSI has been replaced by SAS (Serial Attach SCSI). You won't find huge capacity SAS drives because most applications need IOPS and if you need more storage you add additional spindles which gives you more of both. There's also the problem with 1TB drive rebuild time, heck we're concerned enough about rebuild times on our 300GB fibrechannel drives that we are going with 5% hotspares to reduce the chances of a double fault causing data loss (despite a 1.5% annual failure rate for similar drives in our array).
If that's a troll it's a good one, but you can provide 7.5x more power over ethernet then the USB spec allows (15.4W vs 2.5W). 802.3at pushes things even further to 24W max.
Remove the soundcard and get a USB3 one, it's better to have the DAC outside the RF noisy PC case anyways.
Thanks, you just explained why I always got way better throughput with Ymodem-g back in the BSS days!
Get an iSCSI HBA (or license the feature if you have gen 5 HP), you should get very close to 110MB/s with jumbo frames.
Hmm, a crappy host based implementation at 5Gbps or a storage targeted, low everhead one at 3Gbps (possibly going to 6Gbps if the next gen spec gets extended to eSATA), I know which one I will use =)
have you ever seen DMA to memory ever work properly on consumer grade hardware anyway?
Yeah, every day. PIO mode HDD's suck terribly, DMA is necessary to achieve decent performance.
Ok, so they are going to use 200KW for 10 minutes or about 34kWhrs. These guys make a system that is probably currently the best bet for power density vs orbital safety, the EEV version has 1306 Wh and weighs 15.6kg. At 80% discharge that's about 1KWhr. That means that you need ~530Kg of batteries, at $10K/lb that comes out to $12M to launch the batteries for this thing. I was going to suggest ultracapacitors but it turns out they suck for energy density, on the order of less than 6Wh/Kg! Reading on wikipedia (yeah, I know) Lithium Ion batteries with nanowires achieve almost 15kWHrs/Kg (.75 * 4,2000 mAh/g @ a nominal 3.6V per cell) so developing them would save almost the entire launch cost AND get us better batteries for all sorts of terrestrial applications!
Nope, cable co's are forced to carry the ATSC signal for the locals. My cableco does it the cheap way and just encapsulates them in clear QAM and puts em on the wire, not sure how the SDV folks are handling it. This is why I haven't upped to the digital+HD package (about $30/month more after all the receiver rentals) which only adds 14 non-local HD channels.
I know with Drivesavers you pay an analysis fee upfront, if they can't recover anything they refund it, if they can they will apply it to your recovery bill, if you choose not to pay for recovery they keep the fee to pay for the techs time. This is completely fair in my eyes as just getting to that stage may have involved swapping controllers or moving platters to a new enclosure.
Many BIOS's support timed power on events so it could go like this: 7:50a power on
8A-5P startup script enables profile with full power and dynamic clock throttle (AMD PowerNow/Cool'n'Quiet or Intel SpeedStep)
5P scheduled task switches power profile to mobile or custom
3A WOL from management server wakes up PC if updates are needed (I believe you can even do this with a scheduled task for standalone PC's)
4A previously changed profile puts PC back to sleep
You keep talking about codes, WHAT codes? The payload that activates the feature (at least in the Lenovo implementation) is a cryptographically signed message, there is no default code! It's just like Blackberries, a cryptographically signed message is received by the device and it initiates a wipe of the device, it the case of the Blackberry it wipes the RAM and flash areas, in the case of the Lenovo it wipes the storage keys from the TPM chip.
From what I've read it doesn't work, a scan with an MFT aware app still shows fragmentation afterwards. I haven't tested this myself as I figure Diskkeeper (the guys who wrote MS defrag) know enough that I trust them to defrag the MFT online.
Hmm, on second thought after doing some more research it MIGHT matter:
CRC is somewhat expensive. 200-300 MB/s on a modern CPU looking into ways to optimize
SSE4 will have a CRC instruction (any poly)
linky(pdf)
Not sure what he means by CPU there, if he means core then it's not so bad (other than freaking expensive per core Oracle licensing), if that's per processor then that's very bad. Guess I might only be able to implement this on the new Shanghai beast we got last week or a new Corei7 Xeon system we will probably get next year once they are available.
Taking the HDD out gains you NOTHING, in theory it's already fully encrypted with 256 bit AES which is uncrackable by any currently known method. The idea is that there is only one real vulnerability in a TPM based system and that is the TPM chip's keystore and the databus that the TPM chip uses to talk to the CPU, if you erase the keystore and thus makes sure that both those pathways are neutralized there should be no possible way to retrieve the data off the disk. There's still the cooled RAM trick and possibly a trace of the key left in the disk controller's cache, but those are both VERY sophisticated attacks that have a very low chance of working even in lab conditions. Oh and I just thought of something, if the TPM keystore is wiped then the TPM trust web collapses and the machine should reboot thus flushing the key from ram.