Systems, apparatus, and methods related to parallel processing in a spiking neural network are described. In some examples, parallel processors may compute a time delta vector based on post-synaptic timestamp vector and a current timestamp. The processors may calculate a long-term depression (LTD) value based on the time delta vector and load synaptic weights from memory based on at least the time delta vector. The processors may compute a second time delta vector using various inputs, such as a pre-synaptic timestamp vector, the current timestamp, and pre-synaptic timestamps. The processors may calculate a long-term potentiation (LTP) value based on the second time delta vector and adjust a current synaptic weight vector based on the LTD value and LTP value to generate an updated synaptic weight vector. The updated synaptic weight vector may be written to volatile memory (e.g., DRAM) or non-volatile (e.g., NAND Flash).