AWS · MLA-C02 · BETA
Cost and capacity
MLA-C02 · Lesson 21 of 23 · 2:50:01
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 21 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 4, operating, monitoring, and securing). In this lesson: inference instance families, capacity, reliability, and quotas, purchasing options and cost tools, and the cost of foundation models, embeddings, and vectors. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- The demand forecast endpoint runs on one large GPU instance, in one Availability Zone. An outage in that zone stopped the service for an hour. What should the ML engineer change?
- Three teams call the same Bedrock model. Finance wants the monthly model cost of each team, and an alert when a team goes over its limit. What should the ML engineer set up?
- A RAG assistant over two million documents is searched only a few times a day. Its vector store now costs more than its model calls. Which change lowers that cost the most?