I could use some help troubleshooting some strange behavior...  It seems like the ECU is getting confused momentarily.

When under cruise at highway speeds, I get a momentary surge when transitioning from a lower load (flat/downhill) to a higher load (uphill).  It is slight, but noticeable.

A review of logs shows (screenshot attached) the following...
1) Items remaining constant in the event...  Engine speed is 2,850 rpm.  WGDC = 0, boost set point is idle vacuum, actual boost is slightly positive.  Cam is set at 30 degree overlap.
2) Requested Load spikes 20% from one log entry to the next, which is followed by a return to the near original level in the subsequent log entry.  Since I'm using Durametric, this "round trip" takes place over about half a second.  Actual torque change mirrors the requested load change -- up 20%, then immediately down 15%.  Injector time (te) spikes about 10% and lamdba drops a similar 10%.
3)  Prior to the load spike, the throttle is gradually opening, but drops 15% (4 degrees) when the load spike occurs.  Unlike other factors throttle doesn't recover to near original value.  Note throttle value is well below WDKUGDN.  Ignition angle behaves like throttle.
4)  Here is the really interesting part...  HFM (time averaged/filtered by ME7) drops by 40%, BUT calculated mass flow increases by 20%.  Both converge immediately following the large deviation.

One thing I can't log in Durametric...  Valve height: low vs. high.  So, there could be a valve height changeover happening, but I don't know.

Anyone have any thoughts?  Encountered anything like this before?

I was trying to make it easy for folks to review.  You actually want to review a Durametric log?  I don't, but I don't have much choice.

I tried attaching the log, but it was too big at 8MB.  Durametric logs to an excel format (but stores them as text fields).