Scaling　out　network　functions　(NFs)　offers　great　flexibility　and　elasticity　to　the　management　of　service　function　chains　(SFCs).　However,　existing　solutions　overlook　the　fact　that　scaling　out　an　NF　may　influence　downstream　NFs.　Their　plans　need　several　steps　to　scale　out　overloaded　NFs,　resulting　in　unpredictable　SFC　performance　and　slow　convergence.　In　this　paper,　we　propose　FastScale,　a　novel　framework　that　makes　chain-wide　scale-out　plan　within　a　single　step.　Once　an　overloaded　NF　is　detected,　it　models　the　problem　of　scaling　out　NFs　in　consideration　of　both　overloaded　NFs　and　downstream　NFs.　It　determines　which　NFs　to　scale　and　how　many　NF　instances　should　be　used　by　each　NF　within　a　single　step　so　as　to　avoid　unnecessary　reconfiguration　and　reduce　convergence　time.　After　making　scale-out　plan,　FastScale　automatically　updates　the　thresholds　used　to　detect　overloaded　NFs　to　reduce　user　burdens.　We　have　built　a　prototype　of　FastScale.　Our　evaluation　demonstrates　that　FastScale　reduces　the　convergence　time　of　SFCs　by　up　to　68%　in　comparison　with　existing　solutions.　©　2020　IEEE.