High-mass star formation, contrary to its low-mass counterpart, remains poorly understood. This project aims to constrain the mechanisms driving the formation of OB stars and their precursors with the analysis of two main fields: Aquila, observed by the Balloon-borne Large-Aperture Submillimeter Telescope (BLAST), and the W3 Giant Molecular Cloud (GMC), provided by the Herschel Space Observatory and for which Canada has lead responsibility within the Herschel programme. The Aquila analysis, combining BLAST and interferometry data, presents a full characterization of the main high-mass star forming clumps in this field and their Ultracompact HII regions. We have found highly clustered environments, a range of evolutionary stages, and signatures of triggering within the parsec-scale clumps hosting OB stars. Our study of the W3 GMC comprises an analysis of the young stellar objects (YSO), the regions currently hosting on-going high-mass star formation, the environment, history, and compact source population. Spitzer data were used to identify, classify, and investigate the clustering properties of the YSO population. These complemented the Herschel datasets, with which we have produced column density and temperature maps, a full analysis of the probability density functions and mass distributions, and a catalog of reliable sources at ~ 36'' resolution. The clump sample was characterized and classified according to their stellar content, spectral energy distributions, the NH2/T maps, the L/M diagram, and various schemes presented in previous studies. Based on our environmental analysis and the unique intrinsic and stellar properties of the clumps hosting the clusters of high-mass stars in W3, we have proposed a new high-mass/cluster formation scenario based on external feedback from high-mass stars. This â€˜convergent constructive feedbackâ€™ mechanism could not only explain the formation of clusters with an observed decreasing age (and increasing system/source mass) toward the innermost regions, but could also ensure the availability of material during cluster formation, explain the formation of rare Trapezium-like systems, and address various outstanding issues in high-mass star formation theory. New simulations and additional observations are now required in order to constrain the details and implications of this model.