...
Short Description: Scaling the Test Infrastructure using Docker Containers
- Detailed Description: Currently, most of the sanity/regression in the community is done at limited scale. This would not catch performance/scale penalties introduced as part of the bug fix, as the system would not be sufficiently stressed in this environment. The goal of the presentation is to provide a method to scale the regression setups to test at 60-100DPN scale, without incurring significant hardware costs.
- Topic Leader: Ashvin Lakshmikantha
- Estimated Audience Size (1-15 / 15-30 / 30-50 / >50):
- Interested In Attending: If you are interested in this discussion and would like to participate in it, please add your name and email here (one name/email per line please). We'll use this information when building the schedule so that we minimize overbooking people where possible.
- Ashvin Lakshmikantha, Dayavanti Kamath
- Ashvin Lakshmikantha, Dayavanti Kamath
Path Computation in BGPCEP
Short Description: Presentation of our Path Computation implementation in BGPCEP
- Detailed Description: The presentation will show how we add Path Computation to BGPCEP project in order to add conformity to RFC 5440. In particular, the new version is able to reply to a PCE Request message with a PCE Response message conveying a computed path that respect the requested constraints. We have implemented the well known SAMCRA algorithm for the Path Computation. The Traffic Engineering Database (TED) is automatically computed by learning network topology through BGP-LS. The TED has been modelled in yang and a connected and oriented graph is automatically derived from it. This graph is used by the SAMCRA algorithm to compute path. Bandwidth, Delay, Loss and other parameters could be used as constraints for Path Computation. The presentation ended with a demo.
- Topic Leader: Olivier Dugeon (Orange Labs)
- Estimated Audience Size (1-15 / 15-30 / 30-50 / >50):
- Interested In Attending: If you are interested in this discussion and would like to participate in it, please add your name and email here (one name/email per line please). We'll use this information when building the schedule so that we minimize overbooking people where possible.