Add more detaled descriptions and correct sample configurations

README.md

@@ -1,5 +1,111 @@
-# ns3-rdma
-NS3 simulator for RDMA over Converged Ethernet v2 (RoCEv2), including the implementation of DCQCN, TIMELY, PFC, ECN and shared buffer switch
+# NS-3 simulator for RDMA
+This is an NS-3 simulator for RDMA over Converged Ethernet v2 (RoCEv2). It includes the implementation of DCQCN, TIMELY, PFC, ECN and Broadcom shared buffer switch.
 
-# Note
-TIMELY module has not been merged into this yet. We are working on merging it. We will also add descriptions for this project soon. 
+It is based on NS-3 version 3.17, and ported to Visual Studio environment, as explained [here](https://www.nsnam.org/wiki/Ns-3_on_Visual_Studio_2012).  
+
+## Note
+TIMELY module has not been merged into this yet. We are working on merging it.
+
+## Quick Start
+
+### Build
+To compile it out-of-the-box, you need Visual Studio. 
+People have successfully built it with *free* version, 
+which can be downloaded [here](https://www.microsoft.com/en-us/download/details.aspx?id=48146). 
+Open windows/ns-3-dev/ns-3-dev.sln, just build the whole solution.
+
+You may try building it with the original Makefile, etc. We have done it a while back, but now you probably need to edit a few things to make it work.
+
+### Run
+The binary will be generated at windows/ns-3-dev/x64/Release/main.exe.
+We include a sample configuration file at windows/ns-3-dev/x64/Release/mix/config.txt
+Execute main.exe in windows/ns-3-dev/x64/Release/:
+```
+cd windows\ns-3-dev\x64\Release\
+main.exe mix\config.txt
+```
+
+It runs a 2:1 incast at 40Gbps for 1 second. Please allow a few minutes for it to finish.
+The trace will be generated at mix/mix.tr, as defined by mix/config.txt
+
+There are quite a few options in mix/config.txt. We will gradually add documentation.
+For your own convenience you can just check the code, 
+project "main" -- source files -- "third.cc", and see how these options are parsed.
+You can also raise issues if you have any questions.
+
+## What did we add exactly?
+
+**point-to-point/model/qbb-net-device.cc** and all other qbb-* files: 
+
+DCQCN and PFC implementation.
+It also includes go-back-to-N and go-back-to-0 that handle packet drop due to corruption.
+
+In 2013, we got a very basic NS-3 PFC implementation somewhere, and developed based on it. 
+We cannot find the original repository anymore.
+
+**network/model/broadcom-node.cc** and **.h**: 
+
+This implements a Broadcom ASIC switch model, which
+is mostly doing all kinds of buffer threshold-related operations. These include deciding 
+whether PFC should be triggered, ECN should be marked, buffer is too full so packets should
+be dropped, etc. It supports both static and dynamic thresholds for PFC.
+
+*Disclaim: this module is purely based on authors' personal understanding of Broadcom ASIC. It does not reflect any official confirmation from either Microsoft or Broadcom.*
+
+**network/utils/broadcom-egress-queue.cc** and **.h**: 
+
+This is the actual MMU buffering packets.
+It also includes switch scheduler, i.e., when upper layer ask for a packet to send, it will
+decide which queue to be dequeued. Strategies like strict priority and round robin are supported.
+
+**applications/model/udp-echo-client.cc**: 
+
+We implement the RDMA client here, which aligns 
+with the fact that RoCEv2 includes UDP header. In particular, original UDP client has troubles
+when PFC pause the link. Original UDP client keeps sending packets at line rate, soon
+it builds up huge queue and memory runs out. Here we throttle the sending rate if it gets
+pushed back by PFC.
+
+**internet/model/seq-ts-header.cc** and **.h**: 
+
+We didn't implement the full InfiniBand
+header. Instead, what we really need is just the sequence number (for detecting corruption
+drops, and also help us understand the throughput) and timestamp (required by TIMELY.) 
+This is where we encode this information into packets.
+
+**main/third.cc**: 
+
+The main() function.
+
+There may be other edits here and there, especially the trace generation is scattered
+among various network stacks. But above are the major ones.
+
+## Q&A
+
+**Q: Why do you port it to Windows?**
+
+A: This is a Microsoft project. Visual Studio, including the free version, works well.
+
+**Q: Fine. What if I want to run it on Linux, and do not want to spend time changing the build process?**
+
+A: You can build it using Visual Studio and run the .exe using WINE. We have tested WINE 1.6.2 and it works well.
+
+**Q: I don't understand ... (some part of the code or configuration)**
+
+A: Raise issues on GitHub, so that your questions can also help others. If you really do
+not want others know you are working on this, you can email [email protected]
+
+**Q: What papers should I cite, if I also publish?**
+
+A: Below are the ones you should definitely check. They are ranked from most relevant to 
+less. That said, all of them are quite relevant:
+
+*ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY*, CoNEXT'16 (this project is released with this paper, we ask you to at least cite this paper if you use this code.)
+
+*Congestion Control for Large-scale RDMA Deployments*, SIGCOMM'15 (DCQCN)
+
+*TIMELY: RTT-based Congestion Control for the Datacenter*, SIGCOMM'15 (TIMELY)
+
+*RDMA over Commodity Ethernet at Scale*, SIGCOMM'16 (discussed go-back-to-N)
+
+*Deadlocks in Datacenter Networks: Why Do They Form, and How to Avoid Them*, HotNets'16 (PFC deadlock analysis, directly used this simulator.)

windows/ns-3-dev/mix/config.txt → windows/ns-3-dev/x64/Release/mix/config.txt

@@ -1,33 +1,39 @@
 ENABLE_QCN 1
 USE_DYNAMIC_PFC_THRESHOLD 1
 PACKET_LEVEL_ECMP 0
-FLOW_LEVEL_ECMP 0
+FLOW_LEVEL_ECMP 1
 
 PAUSE_TIME 5
 PACKET_PAYLOAD_SIZE 1000
 
-TOPOLOGY_FILE C:\ns-3-win2\windows\ns-3-dev\mix\topology.txt
-FLOW_FILE C:\ns-3-win2\windows\ns-3-dev\mix\flow.txt
-TCP_FLOW_FILE C:\ns-3-win2\windows\ns-3-dev\mix\flow_tcp.txt
-TRACE_FILE C:\ns-3-win2\windows\ns-3-dev\mix\trace.txt
-TRACE_OUTPUT_FILE Z:\mix.tr
+TOPOLOGY_FILE mix/topology.txt
+FLOW_FILE mix/flow.txt
+TCP_FLOW_FILE mix/flow_tcp_0.txt
+TRACE_FILE mix/trace.txt
+TRACE_OUTPUT_FILE mix/mix.tr
 
 SEND_IN_CHUNKS 0
 APP_START_TIME 1.0
 APP_STOP_TIME 10.0
-SIMULATOR_STOP_TIME 2.05
+SIMULATOR_STOP_TIME 3.01
 
 CNP_INTERVAL 50
 ALPHA_RESUME_INTERVAL 55
 NP_SAMPLING_INTERVAL 0
 CLAMP_TARGET_RATE 1
 CLAMP_TARGET_RATE_AFTER_TIMER 0
 RP_TIMER 60
-BYTE_COUNTER 300000
+BYTE_COUNTER 300000000
 DCTCP_GAIN 0.00390625
-KMAX 1000
-KMIN 40
-PMAX 1.0
+KMAX 1000
+KMIN 40
+PMAX 1.0
 FAST_RECOVERY_TIMES 5
 RATE_AI 40Mb/s
 RATE_HAI 200Mb/s
+
+ERROR_RATE_PER_LINK 0.0000
+L2_CHUNK_SIZE 4000
+L2_WAIT_FOR_ACK 0
+L2_ACK_INTERVAL 256
+L2_BACK_TO_ZERO 0

windows/ns-3-dev/mix/flow.txt → windows/ns-3-dev/x64/Release/mix/flow.txt

@@ -1,9 +1,7 @@
-4
+2
 2 1 3 10000000 2.0 9.5
 3 1 3 10000000 2.0 9.5
-4 1 3 10000000 2.0 9.5
-5 1 3 10000000 2.0 9.5
 
 
 First line: flow #
-src dst pg packet#
+src dst priority packet# start_time end_time

windows/ns-3-dev/x64/Release/mix/topology.txt

@@ -0,0 +1,11 @@
+4 1 3
+0
+0 1 40Gbps 0.001ms 0
+0 2 40Gbps 0.001ms 0
+0 3 40Gbps 0.001ms 0
+
+First line: total node #, switch node #, link #
+Second line: switch node IDs...
+src0 dst0 rate delay error_rate
+src1 dst1 rate delay error_rate
+...