Sentence examples for a simulation instance from inspiring English sources

A simulation instance is illustrated in Figure 1.

For each system size, we launched concurrently on the GPU an increasing number of stochastic simulations; that is, we launched the simulation kernel with an increasing number of blocks (as explained before, the evolution of a simulation instance is computed entirely by a single block of threads independently from the others).

A direct comparison between the schemes shows that the Hysteresis scheme outperforms the k-Handover and the Greedy scheme in 75% of all simulation instances.

Figure 8 shows the average PSNR plot, averaged over all simulation instances and over the whole sequence duration, for the received sequence of the first user for,, and subcarriers, where increasing the number of subcarriers is synonymous to an increase in the available bandwidth.

With reference to Figure 6, the inner simulation area Au where the UEs are distributed consists of a random number of macrocells and picocells in each simulation instance.

(a) Simulation result instances (C=Nma), (b) The effect of a on the power law coefficients of the distributions of social relationships strengths, (c) Simulation result instances (the distributions of social relationships strengths).

For this reason, the kernel can compute a maximum of 256 simulation instances for the smallest system on the small GTX-580 device and between 64 and 84 for the largest one when considering the three different devices.

The implementation scales well when augmenting the number of compartments: in Table 3 we can see that, for a fixed number of simulation instances (i.e., the same number of blocks as 64-32), increasing the system dimension does not affect the speedup achievable.

As larger systems occupy more space in the global memory of the device and we avoid any communication between host and device during the simulation (which is performed running a single kernel as explained), the maximum number of simulation instances that is possible to launch decreases as the number of compartments increases.

For example, if the shared memory is 48 Kbytes (the maximum amount for each stream multiprocessor of an Nvidia GPUs available nowadays on the market) and we want to run concurrently 16 simulation instances (i.e., we launch a grid of 16 blocks), then the maximum number of membranes each block can process would be 48 Kbytes/(16 × 4) = 768 bytes, rather small for our purposes.

The maximum values obtained using the simple branching operation for different problem sizes were compared to the optimal solutions obtained through the branch-and-bound method over 1000 different simulation instances.