The facts is, data density on solid state has already surpassed data density on magnetic/spinning media. We already have 4TB SSDs at SanDisk, and 8TB and 16TB drives are on the horizon. I highly doubt you'll see HDD do 16TB in a 2.5" package anytime soon. Also, when you factor in power requirements, the economics start to look very favorable to SSD, not 10K or 15K RPM drives which are used in latency-sensitive applications. And, in another unexpected and non-intuitive twist, SSDs have a much higher MTBF then HDDs, around 2.5M hours vs. 1.1M hours. Wear-out has become a non-issue, even in the harshest environments. There are SSDs that are rated for anything from less than 1 Drive Write Per Day (DWPD) on up to 45 DWPD during the length of warranty (usually between 3 and 5 years).
Bottom line, SSDs are a far better economical choice in many cases. Of course, this only matters where economics count. Enthusiasts are already seeing the benefit in lower prices and higher densities, but the OP is not about them. Economics is not about "can the consumer afford it" as much as it is about "does this enable me to lower my TCO in the data center". We are there already.
I think a lot of commenters on this are missing the point that 16 lanes of PCI-e 3.0 directly on-die is going to be a massive boost to Native PCI-e NAND Flash implementations (Fusion-io, for example). One of the biggest hurdles to getting more productivity out of faster CPUs and the proliferation of sockets/cores is feeding data to those CPUs. The disparity here is staggering... CPUs have improved by over a million times where storage interfaces and devices have only improved perhaps 100x (being generous) in the same timeframe. This change puts many terabytes of native PCIe NAND flash memory in very close proximity to the CPU complex and will enable vastly more efficient applications.
This type of reader has very good specifications regarding false-positives and false-negatives.
The basis of the technology is that it constructs an image of the fingerprint based on variations in the electrical resistance over the surface of the living finger - please note the word "living." The exterior surface of your skin, including the finger, is dead. Strip off the upper few layers of dead skin cells and you find the first layer of living skin cells. These skin cells have specific qualities of electrical resistance. They also composed into specific shapes over the surface of the skin. The combination of specific electrical qualities in the cells and specifically how the cells are arranged results in measurable and unique variations of electrical resistance over the surface of the finger. That is what a capacitive fingerprint sensor does - it reads variations in the electrical resistance of the living surface of the finger and constructs a map of the finger that shows those variations.
I don't think the variations caused by stress or illness would be enough to skew the reader to get a false-negative reading (user fails to login with correct finger). If you think about the fact that the user is dragging the finger across the reader at infinitely variable speeds (even though only a relatively small range of speeds will work), you'll realize that the mapping technology is pretty advanced, so the very slight variations introduced by stress or illness would not interfere.
I am the guy they quoted in the original press release. I have one of these babies in my hands and let me tell you... pretty cool stuff.
My 2 cents...
The fingerprint reader is of a type that has not been 'fooled' yet. Yes, contact readers are easy to fool. This is not a contact reader. It reads the capacitive properties of the ridges and valleys that make up your finger print. This is actually quite cool since a severed finger does not have the same capacitive properties, and the reading is of live tissue *under* the skin, not your dead skin at the surface. So, a minor injury isn't going to be a big deal and the mafia cannot cut your finger off and use it. Furthermore, the extra small footprint of the reader is nice because there is less opportunity to damage the reader with scratches.
The idea is to register more than one finger and fingers from both hands. Of course, nothing is foolproof, but the idea here was to include a low cost yet effective way to provide biometric access control to the laptop. The embedded security system (ESS) protects a lot of things including a password vault. Password vaults have their drawbacks, the most obvious of which is if you have the 'master' password, you now have *all* of the passwords that user has stored in the vault. Average users tend to use simple master passwords, making the password vault a huge risk. This is a way to provide the functional equivalent of a strong password to unlock the vault without making the user have to remember a complicated password or some hardware key.
I am very impressed with the entire package. I think it will make it much simpler for IT to deploy things like ESS without destroying all of the value in ESS because users choose crappy passwords. There are a number of add-ons that make it very appropriate for enterprise deployment, including centralized key storage and disaster recovery software.
My biggest problem to date with this kind of software was it hasn't been real reliable in the recovery category. I could make it very secure, but God help me if I had a hard drive crash or an OS go belly up. The 'backups' of this data were often times 'too secure' to be recovered. This latest package of hardware/software has many of the previous holes filled in and I am happy to report success in all of the tests I have conducted so far.
Of course, anybody can implement this poorly. However, IBM has done a stellar job with it this time. I feel privileged to get to play with stuff like this.
Slashdot Mirror
The facts is, data density on solid state has already surpassed data density on magnetic/spinning media. We already have 4TB SSDs at SanDisk, and 8TB and 16TB drives are on the horizon. I highly doubt you'll see HDD do 16TB in a 2.5" package anytime soon. Also, when you factor in power requirements, the economics start to look very favorable to SSD, not 10K or 15K RPM drives which are used in latency-sensitive applications. And, in another unexpected and non-intuitive twist, SSDs have a much higher MTBF then HDDs, around 2.5M hours vs. 1.1M hours. Wear-out has become a non-issue, even in the harshest environments. There are SSDs that are rated for anything from less than 1 Drive Write Per Day (DWPD) on up to 45 DWPD during the length of warranty (usually between 3 and 5 years). Bottom line, SSDs are a far better economical choice in many cases. Of course, this only matters where economics count. Enthusiasts are already seeing the benefit in lower prices and higher densities, but the OP is not about them. Economics is not about "can the consumer afford it" as much as it is about "does this enable me to lower my TCO in the data center". We are there already.
No, but maybe I should start writing them.
I think a lot of commenters on this are missing the point that 16 lanes of PCI-e 3.0 directly on-die is going to be a massive boost to Native PCI-e NAND Flash implementations (Fusion-io, for example). One of the biggest hurdles to getting more productivity out of faster CPUs and the proliferation of sockets/cores is feeding data to those CPUs. The disparity here is staggering... CPUs have improved by over a million times where storage interfaces and devices have only improved perhaps 100x (being generous) in the same timeframe. This change puts many terabytes of native PCIe NAND flash memory in very close proximity to the CPU complex and will enable vastly more efficient applications.
This type of reader has very good specifications regarding false-positives and false-negatives.
The basis of the technology is that it constructs an image of the fingerprint based on variations in the electrical resistance over the surface of the living finger - please note the word "living." The exterior surface of your skin, including the finger, is dead. Strip off the upper few layers of dead skin cells and you find the first layer of living skin cells. These skin cells have specific qualities of electrical resistance. They also composed into specific shapes over the surface of the skin. The combination of specific electrical qualities in the cells and specifically how the cells are arranged results in measurable and unique variations of electrical resistance over the surface of the finger. That is what a capacitive fingerprint sensor does - it reads variations in the electrical resistance of the living surface of the finger and constructs a map of the finger that shows those variations.
I don't think the variations caused by stress or illness would be enough to skew the reader to get a false-negative reading (user fails to login with correct finger). If you think about the fact that the user is dragging the finger across the reader at infinitely variable speeds (even though only a relatively small range of speeds will work), you'll realize that the mapping technology is pretty advanced, so the very slight variations introduced by stress or illness would not interfere.
I am the guy they quoted in the original press release. I have one of these babies in my hands and let me tell you... pretty cool stuff.
My 2 cents...
The fingerprint reader is of a type that has not been 'fooled' yet. Yes, contact readers are easy to fool. This is not a contact reader. It reads the capacitive properties of the ridges and valleys that make up your finger print. This is actually quite cool since a severed finger does not have the same capacitive properties, and the reading is of live tissue *under* the skin, not your dead skin at the surface. So, a minor injury isn't going to be a big deal and the mafia cannot cut your finger off and use it. Furthermore, the extra small footprint of the reader is nice because there is less opportunity to damage the reader with scratches.
The idea is to register more than one finger and fingers from both hands. Of course, nothing is foolproof, but the idea here was to include a low cost yet effective way to provide biometric access control to the laptop. The embedded security system (ESS) protects a lot of things including a password vault. Password vaults have their drawbacks, the most obvious of which is if you have the 'master' password, you now have *all* of the passwords that user has stored in the vault. Average users tend to use simple master passwords, making the password vault a huge risk. This is a way to provide the functional equivalent of a strong password to unlock the vault without making the user have to remember a complicated password or some hardware key.
I am very impressed with the entire package. I think it will make it much simpler for IT to deploy things like ESS without destroying all of the value in ESS because users choose crappy passwords. There are a number of add-ons that make it very appropriate for enterprise deployment, including centralized key storage and disaster recovery software.
My biggest problem to date with this kind of software was it hasn't been real reliable in the recovery category. I could make it very secure, but God help me if I had a hard drive crash or an OS go belly up. The 'backups' of this data were often times 'too secure' to be recovered. This latest package of hardware/software has many of the previous holes filled in and I am happy to report success in all of the tests I have conducted so far.
Of course, anybody can implement this poorly. However, IBM has done a stellar job with it this time. I feel privileged to get to play with stuff like this.
-Shawn