1: \begin{abstract}

2:  $\Name$ is

3: an extension of $\Datalog$ that allows for aggregation and recursion

4: over an arbitrary {\em commutative semiring}.

5: Like $\Datalog$, $\Name$ programs can be evaluated via the natural iterative algorithm until a fixed point is reached.

6: However unlike $\Datalog$, the natural iterative

7: evaluation of

8: some $\Name$ programs over some semirings may not  converge.

9: It is known that the commutative semirings for which the iterative evaluation of  $\Name$ programs is guaranteed to converge

10: are exactly those semirings that are stable~\cite{Khamis0PSW22}.

11: %A semiring is $p$-stable if the  number of iterations required for any one-variable recursive linear $\Name$ program to reach

12: %a fixed point is at most $p$, and a semiring is stable if there exists a $p$ for which it is $p$-stable.

13: Previously, the best known upper bound on the number of iterations

14: until convergence  over $p$-stable semirings is $\sum_{i=1}^n (p+2)^i = \Theta(p^n)$ steps, where $n$ is  (essentially)

15: the output size.

16: We establish that, in fact, the

17: natural iterative evaluation of a $\Name$ program over a $p$-stable semiring converges within a polynomial

18: number of iterations. In particular our upper bound is

19:  $O( \sigma p n^2( n^2 \lg \lambda + \lg \sigma))$ where

20:  $\sigma$ is the number of elements in the semiring present

21:  in either the input databases or the $\Name$ program, and $\lambda$ is the

22:  maximum number of terms in any product in the $\Name$ program.

23: \end{abstract}

24: