| ViNePerf | Sridhar Rao | - Support for Kubernetes dataplane benchmarking - east-west traffic - is added.
- Two new tools (stress generator for kubernetes dataplane benchmarking and cloud information collector tool to automate the testing) are added.
- Support for building using newer version of DPDK/OVS-DPDK is added.
- Minor bug fixes (Qemu build, VPP Build, T-Rex)
| - The DPDK-Based East-West dataplane benchmarking, using would be verusing traffic generators such as T-Rex and DPPD-Prox, will be extremely valuable for testing
- The two tools support automation of dataplane benchmarking in kubernetes environment.
- Support for newer versions of DPDK is very important to evaluate the switching based on that (DPDK)
|
|
| Thoth | Sridhar Rao | - 6 Models under VM failure prediction is included (4 LSTM-variations, CNN and Decision Tree).
- Two research studies (ML Problems and Techniques for NFV and OSS projects for ML in NFV) are included.
- Two tools (Model Selector and Data Extractor) are included.
| - ModelSelector suggests the best model to start with, based on the data and problem in hand. This can be very useful and save significant time to the users.
- The research studies gives the thorough information about the state-of-art for ML in NFV.
- The Virtual-Machine FP models improves the performance compared to the existing models.
|
|