The epicentral data I have seen places the main shock right in the trench area, southwest of the island of Sumatra. The aftershock sequence seems to be spread out along the trend of the trench. This would seem to be most certainly subduction related.
However, I have also seen conflicting data concerning the depth of the hypocenter. Some listing reports it at the relatively shallow depth of 10 kiliometers, which is unusual for a subduction zone. For instance, the other recent great quake near New Zealand was at a depth of over 60 km - way down in the subducting plate.
On the other hand, I have also seen a report of a hypocentral depth of 40 km - which is more like it - but still on the shallow side. The amount of energy that be released in a subduction-related earthquake (in the area of two colliding crustal plates) is phenominal.
Today's shallow, thrust-type earthquake occurred off the west coast of northern Sumatra at the interface between the India and Burma plates. In this region, the Burma plate is characterized by significant strain partitioning due to oblique convergence of the India and Australia plates to the west and the Sunda and Eurasian plates to the east. Off the west coast of northern Sumatra, the India plate is moving in a northeastward direction at about 5 cm per year relative to the Burma plate. Preliminary locations of larger aftershocks following today's earthquake show that approximately 1000 km of the plate boundary slipped as a result of the earthquake. Aftershocks are distributed along much of the shallow plate boundary between northern Sumatra (approximately 3 degrees north) to near Andaman Island (at about 14 degrees north).
I'm a bit confused over the "Burma Plate;" I think most sources call it the Andaman Plate.