Open alexpatel opened 7 years ago
Start with getting a Vagrant box off of phase 1 work, get mock server/receiver up, hack on docker/barrelfish within vagrant first and then bintools
https://github.com/Harvard-PRINCESS/documents/tree/master/eval/phase2 https://github.mit.edu/brass/mitll-mock-application https://github.mit.edu/brass/mitll-harness/blob/master/docs/api-developer-guide.md https://github.mit.edu/brass/mitll-harness https://github.mit.edu/brass/mitll-harness/blob/master/docs/api-doc.md
phase 2 evaluation
need to modify diagrams to show api instead of filesystem
produce a much more detailed interaction diagram
2 diagrams
one for bintools perturbation
one for machine desc perturbation
make a script little state machine that will execute the calls
have mostly a shared harness and then fork everything off
m's fantasy - within the next few weeks, can run the full interaction with CRA test harness
it is okay to use docker for barrelfish, do that first, then figure out bintools
want to tell lincoln that we have integrated eval in continuous integration
put pressure on lincoln to give us code so that we can test what we're doing
margo will write up tonight, everyone should review by tomorrow
margo will submit to CRA/Mike with change-tracking on
Start with getting a Vagrant box off of phase 1 work, get mock server/receiver up, hack on docker/barrelfish within vagrant first and then bintools
resources
https://github.com/Harvard-PRINCESS/documents/tree/master/eval/phase2 https://github.mit.edu/brass/mitll-mock-application https://github.mit.edu/brass/mitll-harness/blob/master/docs/api-developer-guide.md https://github.mit.edu/brass/mitll-harness https://github.mit.edu/brass/mitll-harness/blob/master/docs/api-doc.md
7/17/17 meeting notes
phase 2 evaluation
need to modify diagrams to show api instead of filesystem
produce a much more detailed interaction diagram
2 diagrams
one for bintools perturbation
one for machine desc perturbation
make a script little state machine that will execute the calls
have mostly a shared harness and then fork everything off
m's fantasy - within the next few weeks, can run the full interaction with CRA test harness
it is okay to use docker for barrelfish, do that first, then figure out bintools
want to tell lincoln that we have integrated eval in continuous integration
put pressure on lincoln to give us code so that we can test what we're doing
margo will write up tonight, everyone should review by tomorrow
margo will submit to CRA/Mike with change-tracking on