For graphs \(H_1\) and \(H_2\) , the Ramsey number \(r(H_1,H_2)\) is the smallest positive integer N such that any graph G on N vertices contains \(H_1\) as a subgraph, or its complement contains \(H_2\) as a subgraph. Let \(B_{n}^{(k)}\) denote the book graph on \(n+k\) vertices which consists of n copies of \(K_{k+1}\) all sharing a common \(K_k\) , and let \(H:=K_p(a_1,\dots ,a_{p})\) be the complete p-partite graph with parts of sizes \(a_1,\dots ,a_{p}\) . Recently, strengthening a result of Fox, He and Wigderson (Adv. Combin. 4 (2023), 21pp), Fan and Lin (J. Combin. Theory Ser. A 199 (2023), 19pp) showed that for every \(k, p, t\ge 2\) , there exists \(\delta >0\) such that the following holds for all large n. Let \(1\le a_1\le \dots \le a_{p-1}\le t\) and \(a_{p}\le \delta n\) be positive integers. If \(a_1=1\) , then \(r(H, B^{(k)}_n)\le (p-1)(n+ka_2-1)+1\) . The inequality is tight if \(n\equiv 1\pmod {a_2}\) . In this paper, we improve the above upper bounds for the cases when \(n\equiv 2\pmod {a_2}\) and \(n\equiv 3\pmod {a_2}\) . Combining the new upper bounds and constructions of the lower bounds for these cases, we are able to determine the exact values of \(r(K_p(a_1,\dots ,a_{p}), B^{(k)}_n)\) when \(p=3\) . The bound on \(1/\delta \) we obtain is not of tower-type since our proof does not rely on the regularity lemma.