The actual infrastructure costs of running a global Edge API (part 2)

Flagsmith's recent infrastructure overhaul involved transitioning to a serverless architecture with the launch of its Global Edge API, aimed at enhancing global low latency, failover capabilities, and cost management. The API is divided into two main tasks: serving flags, which requires high uptime and low latency, and handling dashboard requests, which are less latency-sensitive. The new architecture uses AWS Global Accelerator for routing and failover, Lambda for compute, and DynamoDB global tables for data storage. The switch to serverless, including a move to arm/Graviton instances, reduced costs by 30% without performance loss. Lambda compute costs about $3 per million requests, while DynamoDB global tables cost roughly $6 due to the read-heavy nature of the platform. Global Accelerator adds minimal cost at about 20 cents per million requests, and data transfer costs are approximately $1.50 per million requests. Despite higher monthly costs, the serverless setup offers significant scalability and failover advantages, aligning with Flagsmith's priorities and reducing the complexity and risk associated with traditional database scaling.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.