{"page":"/tools/bidder-capacity-planner","modified":"2026-10-18","license":"https://creativecommons.org/licenses/by/4.0/","columns":["Item","Input","Symbol","Default","Source or label"],"rows":[["1","Peak QPS (all exchanges, all regions)","Q_peak","100,000","Editable assumption (illustrative)"],["2","Peak-to-average ratio","k","2.0","Editable assumption"],["3","Bid request size (inbound)","R_in","4 KB","Editable assumption; the AWS Prebid test averaged 123 KB per client-to-Prebid-Server request, which bundles several impressions, so exchange-to-bidder requests are smaller (AWS, accessed 17 Oct 2026)"],["4","Bid response size","R_out","1.5 KB","Editable assumption"],["5","Bid rate (share of requests answered with a bid)","b","20%","Editable assumption"],["6","No-bid response size","R_nb","200 B","Editable assumption (HTTP 204 with headers)"],["7","Exchange timeout","tmax","120 ms","OpenRTB 2.6 example value (IAB Tech Lab); Google: \"Typically between 80ms and 1500ms\" (Google, accessed 17 Oct 2026)"],["8","Network round trip to exchange","rtt","20 ms","Editable assumption (measure per exchange and region)"],["9","Serialization and framework overhead","ovh","5 ms","Editable assumption"],["10","p99 latency target (compute only)","p99","50 ms","Editable assumption"],["11","CPU time per request","cpu_ms","2.0 ms","Editable assumption; must be measured on your bidder"],["12","Target utilization at peak","u","60%","Editable assumption (heuristic; see FAQ)"],["13","vCPUs per instance","vcpu","4","c7g.xlarge (Vantage, 17 Oct 2026)"],["14","Instance price per hour","P_h","$0.145","c7g.xlarge on demand (Vantage, 17 Oct 2026); region not stated on the page, us-east-1 assumed"],["15","Hours per month","H","730","Convention"],["16","Regions","n_reg","2","Editable assumption"],["17","Region sizing mode","mode","failover","Editable: failover sizes each region for Q_peak; share sizes each for Q_peak / n_reg"],["18","Redundancy per region","N+","1","Editable assumption (N+1)"],["19","Worker or connection limit per instance","W","256","Editable assumption"],["20","Egress price table","tiers","AWS","AWS: $0.09 / $0.085 / $0.07 / $0.05 per GB at 10 / 40 / 100 / 150+ TB, 100 GB free (CloudZero, 1 May 2026; AWS, 17 Oct 2026). GCP Premium: $0.12 / $0.11 / $0.085 per GiB at 1 / 10 / 10+ TiB (Google, effective 1 Feb 2024). Bare metal: $0 (unlimited traffic on Hetzner AX102, Hetzner, 17 Oct 2026)"],["21","Ingress price per GB","P_in","$0","Editable assumption; verify on the provider's data transfer page"],["22","Inter-region replication share of logs","rep","0%","Editable; priced at $0.02 per GB on AWS (CloudZero, 1 May 2026)"],["23","Bare-metal server price per month and cores","P_bm, c_bm","user-entered","Editable (Hetzner page shows specs but the price did not render on fetch)"]],"sourceUrls":[[],[],["https://docs.aws.amazon.com/solutions/latest/prebid-server-deployment-on-aws/configurations-for-elastic.html"],[],[],[],["https://github.com/InteractiveAdvertisingBureau/openrtb2.x/blob/main/2.6.md","https://developers.google.com/authorized-buyers/rtb/openrtb-guide"],[],[],[],[],[],["https://instances.vantage.sh/aws/ec2/c7g.xlarge"],["https://instances.vantage.sh/aws/ec2/c7g.xlarge"],[],[],[],[],[],["https://www.cloudzero.com/blog/aws-egress-costs","https://aws.amazon.com/ec2/pricing/on-demand/","https://cloud.google.com/vpc/pricing-announce","https://www.hetzner.com/dedicated-rootserver/ax102/"],[],["https://www.cloudzero.com/blog/aws-egress-costs"],[]]}