Authors: Theodoros D. Verykios, Domenico Balsamo, Geoff V. Merrett Contact: {T.Verykios, D.Balsamo, gvm}@ecs.soton.ac.uk Group: Electronic and Software Systems Department: Electronics and Computer Science Faculty: Physical Sciences and Engineering Date of collection: June-November 2017 Tab 1: Figure 3: Experimental results showing the time and energy oveerhead of Allocated State with FRAM and Flash memories. Tab 2: Figure 4: Modelled energy consumption for restoring the system state using the Allocated State Policy, applied to a system featuring a symmetric memory without erase cost and a system featuring an asymmetric memory with erase cost, using typical parameters of FRAM and Flash from Table 2. Tab 3: Figure 5: Modelled energy requirement for saving the system state using the Allocated State policy applied to a system featuring NVM with erase cost (Equation 5), using typical parameters of Flash from Table 2. Tab 4: Figure 6: Modelled energy consumption for saving the system state using the Multiple Allocated State Images policy, applied to a system featuring NVM with erase cost (Equation 7), using typical parameters of Flash from Table 2. Tab 5: Figure 9: Modelled energy consumption for saving the system state using the Updated Blocks policy, applied to a system featuring an asymmetric NVM without erase cost (Equation 9), using typical parameters of PCM from Table 2. Tab 6: Figure 11: Modelled energy consumption for saving the system state using the Multiple Updated Blocks policy, applied to a system featuring an asymmetric NVM with erase cost (Equation 11), using typical parameters of Flash from Table 2. Tab 7: Figure 14: Experimental results showing the energy required to save the system state, when using the Allocated State policy on FRAM. Tab 8: Figure 15: Experimental results showing the energy required to save the system state, when using the Allocated State policy on Flash memory. Tab 9: Figure 16: Experimental results showing (a) the energy required to save the system state and (b) the number of iterations, when using the Multiple Allocated State Images policy on Flash memory. Tab 10: Figure 17: Experimental results showing the restore energy against the number of stored images in NVM, when using the Multiple Allocated State Images policy on Flash memory. Tab 11: Figure 18: Experimental results showing (a) the energy required to save the system state and (b) the number of iterations, when using the Multiple Updated Blocks policy on Flash memory. Tab 12: Figure 19: Experimental results showing the energy required to save the system state, when using the Updated Blocks policy on asymmetric NVM without erase cost. Policy validated using the available hardware as a proof of concept (Flash, negating erase cost). Tab 13: Figure 20: Experimental results showing the average energy required to save each block against percentage of memory contiguously changed. Policy validated using the available hardware as a proof of concept (Flash, negating erase cost). Tab 14: Figure 21: Experimental results showing the energy required to save the system state, when using the Allocated State policy on asymmetric NVM without erase cost. Policy validated using the available hardware as a proof of concept (Flash, negating erase cost). Tab 15: Table 1: Comparison of the properties of different NVM technologies. Tab 16: Table 2: Typical values for reading/writing data from/to different NVM technologies. University of Southampton, UK