Planning - 1/19/17

[twood02]

Implementation:

• [ ] Wrapper for VMI / Remus Interface: @nks5295
• [x] Hash table for bad process list: @leonizhen (@nks5295 will send hash table code). Then get sr/send_mfns code onto your machines
• [x] Modifying memory allocator to add canaries inside VM (in kernel or wrapper for malloc or data structure library) @harpreetsc1992
• [ ] Event monitoring - send system dump to volatility - @nks5295
• [ ] Event monitoring - fix stepping issue - @nks5295

Experiment:

• [x] Set up parsec and provide scripts to use it: @nks5295
• [ ] Set up opensuse guests: @nks5295
  • [x] pv guest
  • [ ] hvm guest
• [ ] @SunnyRaj should add whatever he is doing/needs done for experiments and evaluation
  • [x] Fix timers - Add timestamp instead of calculating time difference
  • [ ] Write scripts to calculate per checkpoint statistics and entire run statistics
• [x] Run experiments using different checkpoints and different workloads for
  • [x] w/ vs. w/o netbuff
  • [x] w/ vs. w/o LibVMI
• [x] Run exps. for a set checkpoint and workload for
  • [x] writev vs memcpy

Paper: We will meet Mon/Tuesday to talk about paper outline

• [ ] @SunnyRaj
• [ ] @harpreetsc1992
• [ ] @nks5295's parts
  • [ ] write experiment system setup (node info, versions, etc.)
  • [ ] write about libvmi event monitoring
  • [ ] write about interaction with volatility
  • [ ] write about parsec and netalytics experiments
• [ ] @leonizhen