Relative perfect complexes (Q6160246)

A first notion of a relative complex was developed by \textit{P. Berthelot} (ed.) et al. [Séminaire de géométrie algébrique du Bois Marie 1966/67, SGA 6.Dirigé par P. Berthelot, A. Grothendieck et L. Illusie, Avec la collaboration de D. Ferrand, J. P. Jouanolou, O. Jussilia, S. Kleiman, M. Raynaud et J. P. Serre. Théorie des intersections et théorème de Riemann-Roch. Springer, Cham (1971; Zbl 0218.14001)], but the condition is difficult to check in practice. In the present paper the authors give an equivalent characterization of relative perfect complex with respect to a morphism \(f\colon X\rightarrow Y\) between quasi-compact and quasi-separated schemes. Namely, a complex \(\mathcal{E}\in {\boldsymbol{\mathsf{D}}}_{\mathsf{qc}} (X)\) is relatively perfect with respect to \(f\), or \(f\)-perfect for short, if \(\mathcal{E }\) is pseudo-coherent and has the further property that \(\mathcal{E}\otimes_Y^{\boldsymbol{\mathsf{L}}}\boldsymbol{\mathsf{L}}f^*M\) remains bounded for any \(\mathcal{M}\) bounded quasi-coherent complex on \(Y\). Several useful characterizations follow very naturally from this definition. The authors prove that it agrees with the classical notion in Grothendieck's seminar SGA6. The main result in the paper (Theorem 4.3 and its corollary) states that a quasi-proper morphism \(f\) has finite flat dimension if and only if the right derived direct image functor \({\boldsymbol{\mathsf{R}}}f_*\) preserves perfect complexes. Notice that a quasi-proper map between Noetherian schemes is just a proper map. As a consequence of this result the authors generalize both the semi-continuity and Grauert's Theorem in two senses. First, the Noetherian hypothesis, needed in the previous approach, is removed. Second, using and alternative definition of the derived fiber it is possible to bypass the flatness condition on the sheaf involved. In a final section the authors show how to obtain a bivariant theory out of the Grothendieck group of perfect complexes. The axioms of a bivariant theory follow in a rather straightforward way from the general properties of \(f\)-perfect complexes developed in previous sections of the paper.