We introduce a primal-dual interior-point approach to address convex quadratic programming dilemmas, utilizing a novel class of efficient parametric kernel functions. Using basic analytical tools, we establish that the developed algorithm has an iteration bound of O((p+1)np+2/2(p+1)log n/e) for the large-update method, where p is a parameter with p ≥ 2. For the small-update method, we obtain the best known iteration bound, namely O(p2√n log n/e) [5, 7]. Finally, we provide numerical experiments that illustrate the algorithm’s effectiveness.