Translation validation is a critical tool in program analysis: when a program $P$ is transformed into a new program $P’$, translation validation asks whether $P$ and $P’$ have the same semantics. It serves as a middle ground between compiler testing and formal verification, capable of proving that a particular run of a compiler produced correct results. However, one bottleneck holds back wider adoption of translation validation: performance. State-of-the-art tools frequently time out or require extensive manual engineering to adapt to specific use cases. In this paper, we propose a new approach to improving the scalability of translation validation by decomposing the problem along two axes: \emph{spatial} decomposition breaks $P$ and $P’$ into smaller fragments whose equivalence implies the equivalence of the full programs, while \emph{temporal} decomposition breaks the transformation from $P$ to $P’$ into intermediate steps where equivalence is easier to establish. These two decompositions are orthogonal and compose naturally, allowing their benefits to stack. Our evaluation demonstrates that this approach validates 10% of translations that existing approaches fail to validate and speeds up validation by up to $2.4\times$.