Design principles
In addition to the overall Well-Architected Framework cost optimization design principles, there are three design principle for cost optimization for IoT:
-
Manage manufacturing cost tradeoffs: Business partnering criteria, hardware component selection, firmware complexity, and distribution requirements all play a role in manufacturing cost. Minimizing that cost helps determine whether a product can be brought to market successfully over multiple product generations. However, making inefficient decisions in the selection of your components and manufacturer can increase downstream costs. For example, partnering with a reputable manufacturer helps minimize downstream hardware failure and customer support cost. Selecting a dedicated crypto component can increase bill of material (BOM) cost, but reduce downstream manufacturing and provisioning complexity, since the part may already come with an onboard private key and certificate.
-
Avoid unnecessary data access, storage, and transmission: Identify and classify data collected throughout your IoT landscape and learn their corresponding business use-case. Collect the data that is required, applying filtering mechanisms at the edge to minimize sending redundant data to the cloud.
-
Process data at the edge whenever possible: Processing data at the edge can help to save costs. Process large volumes of data locally, upload only relevant insights and high-value data to the cloud.
