Version 3 (modified by Josh Smift, 11 years ago) (diff)


Experimenter Requirements

GENI experimenters provide major input to the meso-scale experiments requirements that drive this project. Following is a list of requirements as understood at this time.

The following requirements MUST be met within the Plastic Slices Project:

Requirement Met Known Issues
Include resources from multiple interoperating aggregates in a slice Y
Allow experimenters to control end-to-end data planes supporting a slice Y
Allow multiple experiment team members to share resources in a slice P Multiple team members can share resources in an experiment, but they're in different slices
Operate experiments continuously for three or more months P Experiments were started and stopped at various points during the project
Operate many experiments simultaneously Y
Support non-IP protocol experiments N MyPLC plnodes only support IP
Support comparison of IP and non-IP path results within an experiment N MyPLC plnodes only support IP
Support video and audio applications P The types of traffic sent are consistent with audio and video applications, but actual A/V apps were not tested

The following requirements MAY be met within the next meso-scale experiment project:

Requirement Met Known Issues
Support control and data connections from mobile endpoints N
Support use of cloud services with GENI slices N
Support up to 1000 hosts in an experiment N
Support BGPMux in an experiment N
Support experimenter-controlled load sharing N
Support bandwidth enforcement (QoS-like functions) N

The following general operations requirements from campuses, operators, and experimenters MAY be met:

Requirement Met Known Issues
Ensure that GENI experiments are compatible with campus security policies P Some recommendations made, no comprehensive review
Allow campuses to manage access to the resources they contribute to GENI P Some access control, but not very granular
Ensure that the level of effort required from campus IT staff to manage GENI experiments is reasonable, compared to similar research programs N
Make experiments easier to set up and take down than they were at GEC9 Y
Make experiments easier to change (add/delete resources, change data paths) than they were at GEC9 Y
Try out operations procedures and policies that can transition to production at-scale use P Emergency Stop testing is still in progress