Vanishing of cohomology over Cohen-Macaulay rings (Q692612)

It is said that a noetherian local ring \(R\) is an AC ring if for any \(R\)-module of finite type \(M\) there exists a nonnegative integer \(b_M\) such that for any \(R\)-module of finite type \(N\), \(\mathrm{Ext}_R^n((M,N)=0\) for \(n\gg 0\) implies \(\mathrm{Ext}_R^n((M,N)=0\) for \(n>b_M\). Local complete intersection rings are AC [\textit{C. Huneke} and \textit{D. A. Jorgensen}, Math. Scand. 93, No. 2, 161--184 (2003; Zbl 1062.13005)] but not all local Gorenstein rings are AC rings [\textit{D. A. Jorgensen} and \textit{L. M. Şega}, Adv. Math. 188, No. 2, 470--490 (2004; Zbl 1090.13009)]. In this paper, the authors prove the descent and ascent of the AC property along some classes of local homomorphisms of Cohen-Macaulay rings and use those results to produce new classes of AC rings. In particular, they prove that if \(f:R \rightarrow S\) is a local homomorphism of finite flat dimension of local Cohen-Macaulay rings and \(S\) is AC, then \(R\) is AC. The also prove the converse at least in some important cases: when \(S\) is the quotient ring of \(R\) by an ideal generated by a regular sequence, or when \(R\) and \(S\) have the same residue field, contain a coefficient field \(k\) and \(f\) is a complete intersection homomorphism of \(k\)-algebras. Reviewer's remark: though the short exact sequence used in the proof of Lemma 2.1 does not exist in general, it is not necessary: the authors also point out that the lemma can be deduced from the long exact sequence associated to the spectral sequence used there, as in C. Huneke and D. A. Jorgensen, loc.cit., Proposition 3.3.(1).