This study evaluated the hemodynamic performance of arterial and venous cannulae in a compliant pediatric extracorporeal membrane oxygenation (ECMO) mock circuit in pulsatile and non-pulsatile flow conditions. The ECMO setup consisted of an oxygenator, diagonal pump, and standardized-length arterial/venous tubing with pressure transducers. A validated left-heart mock loop was adapted to simulate pediatric conditions. The pulsatile flow was driven by a computer-controlled piston pump set at 120 bpm. A roller pump was used for non-pulsatile conditions. The circuit was primed with 40% glycerol-based solution. The cardiac output was set to 1 L/min and the aortic pressure to 40-50 mmHg. Four arterial cannulae (8Fr, 10Fr, 12Fr, 14Fr) and five venous cannulae (12Fr, 14Fr, 16Fr, 18Fr, 20Fr) (Medtronic, Inc., Minneapolis, MN, USA) were tested at increasing flow rate in 12 combinations. The pulsatile condition required lower ECMO pump speeds for all cannulae combinations at a given flow rate, inducing a significantly smaller increase of flow in the mock loop. Under non-pulsatile conditions, the aortic and arterial pressures in the cannulae were higher (p < 0.01) while no significant differences in pressure drop and pressure-flow characteristics (M-number) were observed. The total hemodynamic energy was higher in case of non-pulsatile flow (p < 0.01). Under non-pulsatile conditions, the system was characterized by overall higher pressures, resulting in higher support to the patient. The consequent increase of potential energy compensates for increases of kinetic energy, leading to a higher total hemodynamic energy. Pressure gradients and M number are independent of the testing conditions. Pulsatile testing conditions led to more physiological testing conditions, and it is recommended for ECMO testing.