Added questions, test cases, metrics, and diagnostics

Author: bwbush

docs/protocol-testing.md

@@ -19,14 +19,47 @@ behavior, it is a red flag.
 ## Questions
 
 ### Input Block (IB) Parameters
+
 - Are smaller-sized, more frequent IBs better or worse than larger-sized, less frequent IBs?
 - How does IB generation frequency relate to IB network dissemination time?
-  - How does this time relate to the time described in the network functionality in the Leios paper?
-  - What is the induced network capacity?
+    - How does this time relate to the time described in the network functionality in the Leios paper?
+    - What is the induced network capacity?
+- How rapidly or gradually does efficiency degrade as throughput increases?
+- Do RBs propagate back to IB producers in time for the ledger update to be available when the IB is produced?
+
+### Endoser Block (EB) Parameters
+- How does EB generation frequency relate to having zero or duplicate EBs as candidates for certification in the RB?
+
+### Resources
+
+- What are the peak CPU resources required as a function of throughput?
+    - What operations create CPU bottlenecks?
 
 ## Test Cases
 
 | No. | Description                  | [Parameter](../data/simulation/config.d.ts)                                      |
 | --- | ---------------------------- | -------------------------------------------------------------------------------- |
 | 1   | Impact of increasing IB size | `ib-body-max-size-bytes`, `ib-body-avg-size-bytes`, `tx-size-bytes-distribution` |
 | 2   | Vary IB production rate      | `ib-generation-probability`                                                      |
+| 3   | Vary EB production rate      | `eb-generation-probability`                                                      |
+| 4   | Maximize throughput          | `ib-body-avg-size-byte` * `ib-generation-probability`                            |
+
+## Metrics
+
+- Duplicate EBs
+- EBs not making into a RB (short Leios only)
+- IBs sent or received outside of required window for the pipeline
+- EBs sent or received outside of required window for the pipeline
+- Votes sent or received outside of required window for the pipeline
+- "Expired" EBs
+- "Expired" IBs
+- RBs arriving after an IB is produced in a new pipeline
+- Peak CPU usage
+- Average CPU usage
+- I/O operations per second
+
+## Diagnostics (i.e., faithfulness to the protocol)
+
+- IB, EB, and vote production matches protocol parameters
+- IB, EB, and vote production is proportional to a node's stake
+- Propagation times are consistent with DeltaQSD estimates