High sensitivity of the next generation telescopes, like the EVLA and ALMA, is largely due to a significant increase in observing bandwidth or collecting area or both. To fully exploit this increase in telescope sensitivity requires removal of systematic errors to high accuracy for long integration intervals and across wide bandwidths during post-processing. In general, many sources of error are directionally dependent and unless corrected for, will be a limiting factor for the imaging dynamic range with these next generation telescopes. Use of conventional techniques for imaging and calibration, which ignore frequency and direction dependence of such effects, leave residual errors significantly higher than the advertised sensitivity limits of these telescopes. In this talk, I will briefly discuss the challenges in developing algorithms to fully exploit the capabilities of the next telescopes and describe, in general, new developments in the field of imaging and calibration algorithms. I will discuss in more detail, the new algorithms for wide-band wide-field imaging (full-beam single pointing as well as mosaic imaging) and results from their application to real wide-band data. Significant increase in computing requirements and higher data volumes necessitates deployment of these algorithms on High Performance Computing (HPC) platforms. I will touch upon the work in progress in this area as well.