What are the Limitations of Data Loss Prevention Technologies?
Answers
James,
Depends on what you are asking about. Note I've played in this space but I'm not a pro (although I have written relevant contracts around this stuff).
First example, it is rare that you will find someone who will promise that you won't lose your data. There are just too many ways for it to vanish.
Beyond that, it is a "get what you pay for" market. Fully redundant live failover is the best, and it is honestly getting better. Companies like Violin are making really robust systems for trading (speed + reliability). The software that goes with these systems is getting better too (Nutanix is an interesting play here, as is QuorumLabs, the latter a mil-tech spinout). This can get expensive however.
If you're asking about holes:
1) You never get the data. This is a big one, and is a logical-construct problem if you have two dumb systems conversing. It is possible to have a situation where you cannot verify transmission. See: http://en.wikipedia.org/wiki/Two_Generals'_Problem
2) The data blows up in processing. Think "I bought 100 shares of Facebook, but now you say I have none." In other words, some point of failure loses the event itself. Loss prevention (for example, fully redundant failover) can take time, and even worse fail gracefully (losing the excess transactions, not all). Failover time can easily be in the hours, which in some circumstances is useless, add to that you need to notice that you've failed.
3) The list goes on....
If you're asking about theft (as in 100 people just booked flights using my Amex), what I call "data leakage"...totally different question (so apologies if I tried to address the wrong one).
Cheers,
KP