AWS Certified Big Data - Specialty (#49)

How should an Administrator BEST architect a large multi-layer Long Short-Term Memory (LSTM) recurrent neural network (RNN) running with MXNET on Amazon EC2? (Choose two.)

Use data parallelism to partition the workload over multiple devices and balance the workload within the GPUs.
Use compute-optimized EC2 instances with an attached elastic GPU.
Use general purpose GPU computing instances such as G3 and P3.
Use processing parallelism to partition the workload over multiple storage devices and balance the workload within the GPUs.