English

Boundedness of Cohomology

Commutative Algebra 2009-05-18 v1 Algebraic Geometry

Abstract

Let dNd \in \N and let \Dd\D^d denote the class of all pairs (R,M)(R,M) in which R=nN0RnR = \bigoplus_{n \in \N_0} R_n is a Noetherian homogeneous ring with Artinian base ring R0R_0 and such that MM is a finitely generated graded RR-module of dimension d\leq d. The cohomology table of a pair (R,M)\Dd(R,M) \in \D^d is defined as the family of non-negative integers dM:=(dMi(n))(i,n)N×Zd_M:= (d^i_M(n))_{(i,n) \in \N \times \Z}. We say that a subclass C\mathcal{C} of \Dd\D^d is of finite cohomology if the set {dM(R,M)\C}\{d_M \mid (R,M) \in \C\} is finite. A set S{0,...,d1}×Z\mathbb{S} \subseteq \{0,... ,d-1\}\times \Z is said to bound cohomology, if for each family (hσ)σS(h^\sigma)_{\sigma \in \mathbb{S}} of non-negative integers, the class {(R,M)\DddMi(n)h(i,n)forall(i,n)S}\{(R,M) \in \D^d\mid d^i_M(n) \leq h^{(i,n)} {for all} (i,n) \in \mathbb{S}\} is of finite cohomology. Our main result says that this is the case if and only if S\mathbb{S} contains a quasi diagonal, that is a set of the form {(i,ni)i=0,...,d1}\{(i,n_i)| i=0,..., d-1\} with integers n0>n1>...>nd1n_0> n_1 > ... > n_{d-1}. We draw a number of conclusions of this boundedness criterion.

Keywords

Cite

@article{arxiv.0905.2471,
  title  = {Boundedness of Cohomology},
  author = {Markus Brodmann and Maryam Jahangiri and Cao Huy Linh},
  journal= {arXiv preprint arXiv:0905.2471},
  year   = {2009}
}

Comments

18 pages