Paranoia, money, and research

TalkBack 1 of 2:

Next »

• Thread View
• Flat View

Interesting bunch of problems

Convincing a resourceful determined intelligent paranoid and their lawyers that the mirror system is identical, down to every last line of every last config file, every last piece of hardware, e.g. CPU stepping, is identical. Every last hard drive (including the ones swapped out when they failed) has or had exactly the same firmware on it? Need a (independent, third party) team of experts to do a big audit? Before and after each run? And for there to be a lot of logging etc? Prove that no-one messed with anything deliberately or accidentally (even with good intentions) when the CCTV was down, months ago?

Containers etc, fine in many ways, but once you?re into virtualisation, how will you prove anything?

Keeping it secret, throughout a large organisation, that each department?s simulation system is duplicated. "You?ve explained what you do on those x racks. What do those other x/2 racks do then - they seem to be running flat out, they must be using about two-thirds of your power supply?" "Um," "I was just over at the other department, they have a similar set-up. Are you guys doing something for the government or something then?"

Won?t encryption slow things down?

Rather than duplication would the money and effort be better directed at:

1. Getting some disclaimers drafted? Ten pages of ambiguous stuff that seems to say we make best efforts and if it goes wrong the undersigned certifies that they were warned it probably would because we are not fully competent, and they swore they wouldn?t complain anyway, not even if there was deliberate sabotage?

2. As far as possible including self-checking in simulation software. And doing things in modules and steps, periodically backing up data at defined stages from where the run can be re-started.

3. In particular, buying really fast hardware and people who can program it to best use the speed (taking us back to previous topics

Further re duplication - what if you get different results? Does our would-be laureate choose the one which is "obviously" right and keep quiet and hope? Or are they just as angry, because it will all have to be done again? Another argument for 2. above

Posted by: Ross44   Posted on: 12/17/07 You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

SponsoredWhite Papers, Webcasts, and Downloads

• Three Steps You Need to Know to Stop Data Loss Varonis Sensitive data exposed to misuse or loss... it is the stuff of nightmares ... Download Now
• Reducing Server Total Cost of Ownership with VMware Virtualization Software VMware VMware virtualization enables customers to reduce their server TCO and ... Download Now
• Why Isn't Server Virtualization Saving Us More? A Few Small Changes May Dramatically Increase Your Efficiency VMware Companies have rapidly adopted server virtualization over the past few ... Download Now