Why the doming happens
When you heat leavened dough, two things happen:
- leavening agent creates bubbles, causing the soft dough to rise. For chemically leavened doughs (baking powder or baking soda), the amount of lift mostly depends on the time the bubble creating reaction goes on and the concentration of non-spent leavening agent.
- The gluten in the dough sets, building a sturdy 3-d mesh of long, branchey molecules. When the mesh is strong enough, further bubbles cannot stretch it more, despite the fact that the leavening reaction is still going on. The setting of the mesh depends mostly on heat and the amount of gluten present in flour.
The pattern you are seeing means that your cake gets hotter on the sides than in the middle. The sides get set early and stop rising, while the middle is still soft and continues rising.
This happens because the walls of the pan conduct heat to the sides of the cake quicker than heat is conducted inside the cake. You see the phenomenon in an exaggarated form in a muffin, which is usually higher than wide: it is always rounded on top, and often split, because the liquid core from the middle continues to rise after a crust has formed on the surface.
How to prevent it
One good solution could involve a slower, more even heating method.
This can be accomplished by using a pan which insulates well. With an insulating pan, the sides will stay liquid longer and rise more. Unfortunately there are no insulated drop-out-bottom pans.
I have had good results in a porcelain or Pyrex pan. The pan insulates the cake from the bottom and the sides, but not from the top, so you will need to bake it on a lower rack and/or reduce temperature on the upper heater in order to not burn the upper surface.
Getting a cake layer out of the pan in one piece can be hard.
To deal with this problem, if your batter is not too sticky, the bottom of the pan can be lined with parchment paper. The sides can be lined with a strip of parchment or well lubricated with fat. This approach is a hassle.
Insulate a metal pan
Perhaps, when combined with other approaches, a metal cake pan with thicker walls will work well enough.
You could also try a more DIY solution, like attaching 2-3 layers of a cut-to-fit silicone rolling mat stripes to the outside of the walls of a metal pan. However, it is quite hard to come up with a good attachment method (glue could release toxic fumes at oven temperatures, if it holds at all). So this method should work, but is somewhat hard to apply.
Glass and silicone pans
There are pans with glass bottoms and silicone walls but I think these would not help because the glass would insulate better than the thin silicone.
Perhaps preheating the detached bottom of a metal pan with the oven could help, but I haven't tried it.
Minimize the center
A second approach is to minimize the liquid core by making a thin cake.
If the middle of the cake is heated quickly enough from above and below, it will set shortly after the walls do, and there will be no unset center to "bloom".
Use a bigger pan.
I think that Americans mostly use 9 inch pans, but if you used an original sacher recipe from Austria, it is probably meant for a 26 or a 28 cm pan.
Bake the two layers separately
Divide the batter into two pans instead of making a single layer and cutting it. This is unorthodox, as you get more crust but it's better than a hunchbacked sacher. You should use a scale or at least a measuring jug to divide the batter, or you'll end up with different thicknesses. Also remember to reduce the baking time, as the core will heat up faster.
Use a toothpick for probing doneness.
Reduce the leavening
The third idea is to reduce the amount of leavening agent.
The walls will always set before the core and, if there is enough baking powder, the core will still expand a little bit more. If the concentration of baking powder is low you'll get less bubbles, so less lift.
Of course, using too little baking powder will also ruin the cake, so you'll have to be cautious and experiment a bit before hitting the correct amount.
The fourth approach is to give the outer portions of the cake more time to rise by baking at lower heat.
This is somewhat risky, as it can result in a different texture of the final product, due to the different rate at which water will evaporate from the dough and a longer baking time. Also, if your temperature is too low, you won't get a golden crust on a light-coloured dough. This should not be a problem for a frosted cake like the sacher.
The fifth approach is to use a flour with less gluten.
With less gluten you get a mesh which is less dense and needs more time until it gets firm enough to prevent rising. Use cake flour instead of all purpose flour. A bit more fat will also help to inhibit gluten development but too much will change the taste and texture.
All methods I described should attribute to a solution, but probably none of them will be sufficient by itself. You'll have to pick a combination of them and see what works best for you.
General Cake Techniques
You should also apply all the usual methods for getting a good cake:
- measure with a scale
- use room-temperature ingredients
- sift your flour
- only combine dry mix with fluid mix at the last moment before putting it into the oven
- preheat the oven well
These techniques ensure a better batter texture, which means a more even heating. They will also ensure a more consistent leavening process because:
- the ratio of leavening to other ingredients will be correct
- the batter will be more thoroughly mixed
- the leavening reaction will not start early
Not following these techniques is more likely to result in a lopsided cake or a big bubble. I realize that this hasn't happened in the case you describe, but it would be too bad to get an asymmetrically risen cake after you took all the precautions against a disproportionally rising center.