Slashdot Mirror
Resisting the PGP Whole Disk Encryption Craze
alaederach writes "I run a lab in a non-profit academic life sciences research institute. Our IT recently decided it would be a good idea to use PGP whole disk encryption on all of our computers, laptops and servers and picked PGP's suite of software. The main reason is that a small subset of our researchers work with patient information which we obviously are mandated to keep confidential. My lab does a lot of high-performance computational work (on genes from Tetrahymena, no humans here) and I am concerned that the overhead of complying with our ITs new security policy will be quite detrimental to my research program. For example, dynamically reallocating a partition on a PGP encrypted disk is apparently not possible. Furthermore, there is some evidence that certain forms of compression are also incompatible with PGP whole disk encryption. Interestingly, it is hard to find any negative articles on PGP, probably because most of them are written by IT pros who are only focused on the security, and not usability. I therefore ask the Slashdot community, what are the disadvantages of PGP in terms of performance, Linux, and high-performance computational research?"
what are the disadvantages of PGP in terms of high-performance computational research?
O(1) ;)
Here's a brief experiment I ran: dd if=/dev/zero of=/home/jonas/zeroes bs=1048576 count=1024; that is, writing one gig of zeroes to a disk encrypted with ubuntu's disk encryption from the 8.04 alternative installer.
I saw a roughly constant ~30% CPU usage from kcryptd, going from 25% to 35%, on a 2.13GHz Pentium M (in a thinkpad t43p). So I have 1.5 GHz worth of cycles left.
Hard disk write speed was about 30 megs per second, but oscillating in big leaps. I did my observations with conky, sampling in one-second intervals, but conky is known to sometimes merge two samples. That's probably not the only factor, disk writes are most efficient when clumped together into one big (much preferably sequential) write, so I'd assume the kernel does this.
You haven't told us what your disk usage patterns are. But if you're doing one big read, one big computation, and then one big write, there's going to be zero impact (almost): there was lots of CPU capacity left.
Another low impact scenario is that you have a server that reads work units from disk, hand them to clients, gets results and writes the results back [I assume clients don't need any disk activity]. There you can read a bunch of work units in advance while the server is idle, then hand them out instantaneously when needed.
Aside: bugger, fault in my experiment: I didn't look at the CPU usage of kernel code that's not in the process table. Take what I say with a grain of salt.
But: do the measurement in your own world. My software, hardware and artificial measured usage pattern may differ from yours, subtly but enough that my conclusion doesn't transfer. Be scientific about it :)
You really want blanket encryption because you to worry about such things as swap space, scratch copies made and then deleted and people forgetting to encrypt files. /boot encrypted during install.
If the encryption is done at the block device level (such as dmcrypt on linux) the impact is minimal on how things work and overhead and you are fairly well protected (unless the machine is accessed while powered up by someone wants the data as opposed to just the machine).
Fedora can make all partitions except
Sorry, they "claim" that.
But on my core 2 2.4 Ghz machine, windows boottime more than doubled after encoding the system partition.
Yeah, i can get 100Mbyte/s linear reads and writes.
But for some reason, random or semi random access get hosed quite a bit.
Maybe it messes with the comand queueing, or the internal prefetch alorithmns, i dont know. Never had a problem on data partitions, but the performance impact on the system drive was enourmous (up to the point that even with 6Gbyte RAM, it wasnt fun anymore)
Ah, and i forgot one thing: the 100Mbyte/s is nearly 100% cpu load on both cores. I dont know where you get 1% overhead from... Even the in-memory benchmark only gets about 150Mbyte under full load on two cores.
S
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
Positive:
- added security
Negative:
- worse performance
- you may forget the password (it has happened before.)
- has to be mounted manually (or at least type in password each time you need access to the data.)
- it's painful to backup
- it's painful to do a proper file systems check
- if the discs are somehow taken by the authorities you might have to give up your password (or be sentenced for whatever they think you have on the discs.)
- discs are only secure if they are not mounted.
There are a few negative sides, but usually they make up for the positive, i.e. if you really need the security then of course this is the way to go. Also remember to secure the other aspects of the machine, like physical access (including fire/theft), software protection (anti malware and virus) and network protection (firewalls, etc.)
If you do encrypt why use PGP? It costs money and its proprietary. Use Truecrypt which is free and open source, does whole disk encryption which according to this can sometimes actually *boost* performance. I use Truecrypt daily and its awesome. http://en.wikipedia.org/wiki/Truecrypt#Performance http://www.truecrypt.org/
The numbers on my machine are about 20% slower read and 30% slower write. I'm using 256 bit LUKS with serpent-xts-essiv:sha256.
Might I also suggest hardware encryption? Seagate (and others I believe) make drives that do AES128 (good enouhg for this sort of thing I believe) in hardware. Zero performance hit. No software required. Set a drive password and go.
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.
I second that.
If you're looking for an excuse not to protect the data, that's one thing. But TrueCrypt has lots of support and does a good job. PGP in general is well-known and has been refined frequently. That's the reason you don't find a lot of negative criticism-- there isn't any because it works fairly seemlessly. You'll find hard disk controllers don't help the process much, but if the machine does work in batches, and you backup frequently (presuming you're backing up an encrypted partition) and you use a UPS (or your controller supports battery-backed write cache), you can use various write cacheing driver options and techniques to boost performance dramatically. What write cacheing *can* do is to also cause transactional integrity problems if there's a machine hickup. Otherwise, writes are queued up and get batched onto disk. Performance can be 10x, so long as you understand the potential evils involved. It takes the sting out of the disk I/O degradation, but how much will vary with the duty cycles of your application's I/O profile.
---- Teach Peace. It's Cheaper Than War.
That is interesting - if the overhead was really 1%, then why even bother with optimizations for multi cores?
The other thing I cannot understand is why anyone would want to run whole-disk encryption on a compute server. Even the US DoD machines that are used for classified research do not do this!
My workplace recently mandated that all laptops/portable media be encrypted. The impact to the system cpu usage isn't that significant to be honest, except when attempting to access, say, USB drives.
What's more important is the reliability of the disk itself.
As everyone knows, drivers shipped with laptops tend to be the first casualties of boot-sector-loading programs, like disk encryption and certain virus scanners.
Guess what happens when your encrypted disk can't be booted? You can't boot under a windows/emergency restore disk, because your partition is not readable. You can't boot off anything other than the hard drive. Guess what happens if the corruption doesn't allow you to run the encryption app's boot loader? Only solution is to format the disk.
Some of us who have been hit by this already have gone through the trouble of ensuring that any data we want to keep is stored on a shared drive, and that all work is done in a VM, which is occasionally uploaded to the shared drive as well. Since any given windows or driver-affecting update could kill our machine at any minute and make it entirely unrestorable, that's what's required.
So in essence, we're switching back to storing the media on a non-encrypted device because the loss of the data is more important than the security of the data.
This reminds me of the policies surrounding passwords I've seen at many companies; limiting the set of choices by making password creation requirements, and forcing them to change so often that people end up writing them down and leaving them on their desk. Defeats much of the purpose of having them in the first place.
I hate to disagree, but I have to. IT policy is a law that must be followed. What the problem here is the people creating the law sees only the end goal and not the road that needs to be traveled. Talk to them and show them what is required of you during the research. Tell them that other ways need to be looked at in achieving the goal before this is implemented. More harm is done and time lost by people trying to circumvent the policies then it is by sitting down with them and stating the procedures that are done and stating why a different method is needed.
There are two major products that come out of Berkeley: LSD and BSD. We don't believe this to be a coincidence.
I'm with Smertrios on this one.. IT policy is just that.. a corporate policy. It's not subject to end-user interpretation, it's a definition of how IT resources are to be deployed and utilized. The written policy itself is what gives the company the "teeth" to discipline employees who choose to make their own interpretations and NOT comply.
Now back on topic: Whole disk encryption? For removable / transportable media, ABSOLUTELY! For enterprise data backups, ABSOLUTLEY! For live data on active servers, meh.. not as critical. If your data center employs appropriate physical, network and host security, your data is reasonably safe. If someone compromises your network -> system security, they've got your data.. encrypted or not. It's wonderful that your IT department has the desire to achieve the highest level of security possible, but there is always a balance that needs to be struck between the holy grail of ultimate security and the ability to do business. The OP needs to help everyone find that balance. A good place to start would be his local neighborhood HIPAA expert to make sure that no "business needs" prevent the company from maintaining regulatory compliance. Once the specific requirements for his continues compliance have been identified, then anything beyond that becomes somewhat negotiable.
chown -R us
Read the FAQ; drives usually have larger block sizes than the block size used for encryption, so there is not much difference.
I can tell you that when we ran a PGP encrypted disk partition on a 12 disk raid 50 I had MAJOR performance losses compared to a standard raid 50. This was on older hardware, I had tested it on a 8 processor Xeon PIII 800 system with only 4 gig of ram installed, but it had a significant impact on data transfer rate.
and yes I like re-purposing the Killer SQL servers of yester-year into a "Holy CRAP THAT's YOUR NAS??!??!"
The hit was NOT on the drives, it was on the processors. It was enough of a hit to slow down data transfer rate out the GB connection to be as slow as a consumer single disk NAS.
Do not look at laser with remaining good eye.
The patient information is a pretty serious concern. Any breach or loss of data covered under HIPAA, SOX, FERPA, or Privacy Act can result in some pretty severe expenses. The cost of notification to the individuals whose data was lost or exposed can run to more than $1,500 per individual, depending on the size of the breach. Base expenses start at $1-2M and go up fast. Litigation and fines can cost millions more. Anything that gets hacked or breached, that has information that should be protected, could put a company these days on the wrong side of the balance sheet.
I think the strategy should be to perform some speed comparison tests, to see if your research can be done with full disk crypto. Setup some vmware or other virtual machines.. and your test physical server.. Plug in a spare Hard drive, install a fresh OS, do testing of some virtual machines _with_ and without full disk encryption (on both host and on the VM), and tell them that the full disk encryption is slow if it is: reduces the effectiveness of disk cache, wastes memory, bogs down the CPU of machines that are needing to be used 100%, and better hardware is needed to run full disk encryption.
You're in research, and such a major change to your environment deserves to be looked at a little before you implement it...
I suspect with full disk crypto on your hardware backing the virtual disk, VM I/O performance will tank.
Show them nice graphs of research computing productivity on the same equipment WITHOUT full disk decryption, and WITH it.
Use "full disk encryption" policy as immediate justification for additional better hardware to compensate for the fact that the encryption is parasitic.
And note the migration costs and loss of research time that results in having to make such drastic changes.
Once you show the extra cost involved, they perhaps rethink the full-disk encryption blanket policy.
Just make sure the cost you show is high... (much higher than any imagined savings through simplified policy and assured security)
If you can't so much as justify a position against it, then why is PGP such a problem? If it doesn't hurt you... it certainly makes your research more secure from being stolen.
1% overhead is still a hit if you are using your equipment 100%.
But actually, I don't believe for a second that TrueCrypt or PGP is limited merely 1% overhead, the figure is deceptive in that running disk encryption has effects other than measurable disk I/O slowdown.
There is also CPU usage of the encryption, and memory and reduces page cache effectiveness.
i.e. The heavy cost of encryption must now in all likelihood be performed before data can be written to the page cache. This reduces system throughput.
You may measure simple operations as only impacted by 1%, but in reality, there are certain write patterns that this will hurt severely.
Just plain SELinux has overhead in excess of 10%.
I would expect full-disk encryption of 30% or higher.
It may be difficult to measure its true overhead if you don't fully use your hardware.
Parent is on the right track, imo. Submitter should work with the IT dept to assess the impact of this.
Setup two machines running the same processing task that is actual work that he does, one with encryption and one without. Compare the difference in processing. If the performance loss is acceptable, all done. If it's not acceptable, submitter needs to start agitating now that this will seriously hamper his/her ability to do the job, and push IT to come up with a different solution.
A previous employer rolled this out, and after my work productivity got killed, i found their assessment consisted of two guys opening MS Word, making some edits, saving, and exiting word.
Fine, as long as you work my hours. I work in a job where I may be setting up at 0500 for a multi-person network-heavy presentation scheduled to go at 0630, and I have zero time for argument.
Sounds like you may very well be the kind of user that makes IT staff bitter and hostile. If you didn't make arrangements with the IT staff for your presentation before your presentation is scheduled to start, how is that my problem? One thing that never fails to draw the ire of IT staff is a user who consistently doesn't tell anyone what they need until it's time to go live, then expects said IT staff to drop everything to accommodate their needs at the last minute.
I've had great support and lousy support, and yes, I bring my own network hardware in case the local admin doesn't have what I need.
That's reasonable. However, depending upon what you bring, the local admin may or may not be willing to plug your gear into his network. As the local admin, I am the guy that gets called on the carpet when rogue equipment takes down the network. If I don't know what the gear is or what it will do on my network, it doesn't get installed until it's been tested in a sandbox first. In the past, I've had rogue equipment cause routing loops which in turn caused spanning tree on my switch to turn off the port to the offending network drop, taking out most of a department (they had installed a SOHO switch because they needed more ports, but never told us). I've seen rogue equipment replying to DHCP requests causing conflicting IP addresses or IP addresses that were in an entirely different subnet than the main network. I could go on, but you get the picture. So please excuse me if I don't just take your word for it that your equipment won't break things, unless I know you and know that you really do know what you are talking about.
That said, I almost never have a problem, because good network admins do indeed work with me, and lousy ones either (a) aren't there to complain, or (b) trust me far more than they should.
As long as you are reasonable in your requests and are willing to compromise with the admins on your network, most admins, IMHO, will do their best to find a solution that both they and you can live with. From what I've seen, while the BOFH does indeed exist, he is more of an exception than the rule.
Oh, and I ask (and explain and discuss and compromise) long before any equipment sees power. It's only polite.
That goes a very long ways towards earning the trust of the local admin. I withdraw my first comment about how you sound like the type of user who causes IT staff to become bitter and hostile. I'll bend over backwards -- even at the last minute -- for someone who tries to work with me and/or has shown me that I can trust them.
I've never (ten years or so) had a local hardware issue extend into the host network. It seems to be fairly hard to do that if you're not an idiot (and if your own equipment is truly solid, which mine is).
There's the catch. While most people are reasonably intelligent, there are enough people who aren't to make network admins suspicious of others, if we don't know their technical competency. There are many users who think they know more about networking than the admins who built the network. Sometimes this is true, and sometimes it isn't. At my current job, there is a user that I trust very, very much. He held my job before I did, and still probably knows more about the network than I do (he left for a different department because he got fed up with the guy who used to manage the department). OTOH, there is another user who thinks he is God's gift to networking. While he does have a little knowledge...well, a little knowledge is a dangerous thing.
MCSE? No, sir...I don't do Windows. Yes, I am an idealist. What's your point?