GitLab

Updated congestion control in dragonfly: now messages are saved in a waiting list at the routers in case of buffer overflows

Updating torus network model: now generates one chunk at a time and holds off another until the previous one is sent

Adding group level mapping in the dragonfly: multiple routers can now be specified in the config file

1) Adding dragonfly adaptive and prog-adaptive routing algorithms--- still testing with more workloads and crosschecking the computation 2) Two bug fixes in credit-based flow control

Introduced back when I added a "last-hop" queue to allow prioritized
messages to go to the front of the set of msgs loggp is to receive. Haven't yet
put it in other models (simple*), so RNG bug doesn't affect there yet.

same var name used for the "sequencing" macros

Tired of explaining that it's not a good representation of a WAN. Instead, it's
now a simple point-to-point latency/bandwidth model.

NOTES:
- torus does not use it, as it has some routing logic preventing direct usage.
  It probably can be done, but I'll leave that for later.
- collectives in torus, dragonfly don't use it yet.

Rather than model-net LPs directly sending messages to other model-net LPs, LPs
can route the intended message through the scheduler interface to be queued up
for reception by the receiver (see the diff of loggp.c). This has the benefit of
    enabling things like priority and fairness for N->1 communication patterns.
    Currently, no packetizing is supported, and I haven't yet wrote checks for
    it - beware.

Loggp is currently the only supported model. simplenet could also be supported
without much trouble, but I doubt there's any demand for it at the moment.  This
should NOT be used by the dragonfly/torus models, as they have their own routing
backend.

- previously, packet issues were done without any consideration for device
  availability - within epsilon, preventing any meaningful scheduling
- enabled for loggp only, other networks will be incorporated shortly

- N priorities processed in increasing order
- queue for priority i not touched until 0...i-1 queues are empty
- example injection API usage shown in test program
  - currently, only the priority scheduler has need of it

All configuration now proceeds on a per-LP level, and requires separate
registration and configuration calls, as seen in the test programs.
model_net_set_params is no longer used, and is replaced by model_net_register
and model_net_configure. The dragonfly network, having two LP types bundled in
the same code-path, is special-cased in the registration code.

LP-mapping in model-net now has the following defaults:
- counts via codes_mapping_get_lp_count are now with respect to the calling
  network LP's annotation.
- when looking up network LPs via
  codes_mapping_get_lp_info/codes_mapping_get_lp_id, the annotation of the
  calling network LP is used. Hence, routing now occurs only between LPs of the
  same annotation. If the destination LP's group specified by model_net_*event
  does not contain a modelnet LP with the same annotation as the modelnet LP in
  the sender's group, then an error will occur (in codes_mapping).

Known Issues:
- modelnet users currently cannot specify which modelnet LP to use in the case
  of multiple modelnet LPs in the sender's group. This will be fixed in future
  commits after a consensus is reached on the best way to expose this
  information.

- "modelnet" parameter in cfg is now a no-op
- "modelnet_order" parameter in cfg is required,
  listing order in which networks are indexed to
  the model
- modified "model_net_set_params" signature
- updated tests to use the new interface

Adding collective modeling functions in model-net (currently only dragonfly/torus models support it)

- NOTE: shim just forces packet size to a huge number