Problem in displaying statistics after batch execution

[LuccianoSileo]

When using the experimentation tab and running the Batch Execution function, the following is observed:

• Discrepancy between the graph and the JSON: the graph shown in the Stats section does not correspond to the data received in the output JSON.
• Number of executions reflected: while 20 executions are expected, the graph only shows the result of what appears to be two execution cycles (twice as many as a single cycle).
• Inconsistent results in the second run: the data for the second run appears to be meaningless, suggesting that there is some error in how the results are collected or interpreted during that cycle.

I have used the example programme and the default memory. Using Google Chrome browser 135.0.7049.41 (Official Build) (64 bit). Here are the results.

First, a normal execution

We now show the results after a batch execution of 20.

And if we do a run, but add page fault. In this case, with 10% cache miss. And we run ‘Batch Execution’ several times, the final result is as if it were a single execution.

Here is a video showing the above.
https://github.com/user-attachments/assets/0a498bc5-571a-4ce6-b069-2452512d8c87

[oxcabe]

Thanks for the thorough explanation. Will try to find some time to investigate and resolve it today.