CoMeT is a novel RowHammer mitigation mechanism that uses the Count-Min Sketch (CMS) technique to track DRAM row activations. It aims to prevent RowHammer bitflips with low area, performance, and energy costs at very low RowHammer thresholds. The key idea of CoMeT is to use hash-based counters to track DRAM rows, leveraging the CMS technique to map each row to a group of counters. When a row is activated, CoMeT increments the corresponding counters. Due to the non-unique mapping, CoMeT may overestimate but never underestimates a row's activation count. This property allows CoMeT to securely prevent RowHammer bitflips while reducing overestimations through proper hash function configuration. CoMeT consists of two main components: the Counter Table (CT) and the Recent Aggressor Table (RAT). CT tracks DRAM row activation counts with low area overhead using tagless hash-based counters and CMS. RAT tracks a small set of recently identified aggressor rows with per-row counters. CoMeT's hardware implementation minimizes false positives at very low RowHammer thresholds. Evaluations show that CoMeT prevents RowHammer bitflips with average performance overheads of 0.19% and 4.01% across 61 single-core workloads for RowHammer thresholds of 1K and 125, respectively. Compared to state-of-the-art mitigations, CoMeT requires significantly less area overhead and incurs smaller performance overheads.CoMeT is a novel RowHammer mitigation mechanism that uses the Count-Min Sketch (CMS) technique to track DRAM row activations. It aims to prevent RowHammer bitflips with low area, performance, and energy costs at very low RowHammer thresholds. The key idea of CoMeT is to use hash-based counters to track DRAM rows, leveraging the CMS technique to map each row to a group of counters. When a row is activated, CoMeT increments the corresponding counters. Due to the non-unique mapping, CoMeT may overestimate but never underestimates a row's activation count. This property allows CoMeT to securely prevent RowHammer bitflips while reducing overestimations through proper hash function configuration. CoMeT consists of two main components: the Counter Table (CT) and the Recent Aggressor Table (RAT). CT tracks DRAM row activation counts with low area overhead using tagless hash-based counters and CMS. RAT tracks a small set of recently identified aggressor rows with per-row counters. CoMeT's hardware implementation minimizes false positives at very low RowHammer thresholds. Evaluations show that CoMeT prevents RowHammer bitflips with average performance overheads of 0.19% and 4.01% across 61 single-core workloads for RowHammer thresholds of 1K and 125, respectively. Compared to state-of-the-art mitigations, CoMeT requires significantly less area overhead and incurs smaller performance overheads.