Massive stars (M ≳ 10 M☉) form from collapse of parsec-scale molecular clumps. How molecular clumps fragment to give rise to massive stars in a cluster with a distribution of masses is unclear. We search for cold cores that may lead to future formation of massive stars in a massive (>103 M☉), low luminosity (4.6x102 L☉) infrared dark cloud (IRDC) G030.88+00.13. The NH3 data from the Very Large Array (VLA) and Green Bank Telescope reveal that the extinction feature seen in the infrared consists of two distinctive clumps along the same line of sight. The C1 clump at 97 km/s coincides with the extinction in the Spitzer 8 and 24 µm. Therefore, it is responsible for the majority of the IRDC. The C2 clump at 107 km/s is more compact and has a peak temperature of 45 K. Compact dust cores and H2 O masers revealed in the Submillimeter Array and VLA observations are mostly associated with C2, and none are within the IRDC in C1. The luminosity indicates that neither the C1 nor C2 clump has yet to form massive protostars. But C1 might be at a precluster forming stage. The simulated observations rule out 0.1 pc cold cores with masses above 8 M☉ within the IRDC. The core masses in C1 and C2 and those in high-mass protostellar objects suggest an evolutionary trend that the mass of cold cores increases over time. Based on our findings, we propose an empirical picture of massive star formation that protostellar cores and the embedded protostars undergo simultaneous mass growth during the protostellar evolution.