How can synapses change the probability of neurotransmitter release during synaptic plasticity? While synaptic release probability depends on the number of release-ready vesicles and the release probability of each vesicle, a dissection of both parameters during synaptic plasticity is difficult. Here we use the well established presynaptic homeostatic compensation upon interference with postsynaptic glutamate receptors at the Drosophila neuromuscular junction as a model for plastic alterations of synaptic release probability. As a genetic tool to induce synaptic homeostasis, we used animals lacking the glutamate receptor subunit IIA and expressing only the IIB-type receptor (GluRIIB). Combining paired-pulse analysis, cumulative excitatory postsynaptic current analysis and quantal short-term plasticity modeling, we found an increase in the number of release-ready vesicles during homeostatic compensation. Consistently, in fluctuation analysis, the number of release-ready vesicles was almost doubled in GluRIIB compared to controls (690 ± 128 and 359 ± 58, n = 6 and 8 in GluRIIB and control, respectively; Mann-Whitney rank sum test: P = 0.03). Quantitative confocal image analysis revealed an increase in the amount of the active zone protein Bruchpilot and stimulated emission depletion (STED) microscopy showed an enlargement of the presynaptic cytomatrix structure during homeostatic compensation. Furthermore, we analysed homeostatic compensation on a shorter timescale by incubation with the glutamate receptor blocker Philanthotoxin for 10 min. Interestingly, we found again an increase in the number of release-ready vesicles as well as presynaptic structural alterations. Our results demonstrate that synaptic homeostasis regulates the number of release-ready vesicles on the time scale of minutes to weeks by structural protein redistributions.