- 17 Jan, 2022 40 commits
-
-
Antoine Kaufmann authored
All devices will somehow need this, might as well stick it in the base class.
-
Antoine Kaufmann authored
For largescale experiment we otherwise run out of IPs
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
As described in paper
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
Includes source & dest mac/port, and ARP ips
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
Reduces the risk of getting the same mac in multi-host clusters
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
A bit too short for a fast qemu timing.
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
this allows the shared memory regions to be created in /dev/shm/ e.g. instead of the work dir.
-
Antoine Kaufmann authored
This breaks large mtu experiments, but when proxied makes a huge difference. Should make this a runtime parameter.
-
Antoine Kaufmann authored
Can now select between sockets (default) or rdma.
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
Instead of static ordering of simulator types. This simpliefies the experiment runners but also simplifies adding additional features. Immeadiate goal is adding network-to-network connections
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
These were too large until now.
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
Otherwise on a full queue we might overwrite entries repeatedly causing all sorts of mayhem.
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-
Antoine Kaufmann authored
-