Design Task Scheduling Infrastructure for real-time analytics
Last updated: July 20, 2025
Quick Overview
Design a low-latency task scheduling system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
Datadog
System Design
Software Engineer
Datadog
July 20, 2025Software Engineer
Onsite
System Design
Easy
49
11
2,784 solved
Design a low-latency task scheduling system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
Datadog asks this during the Onsite to assess your understanding of the full ML lifecycle. They want to see how you translate a business problem into an ML objective, design the feature pipeline, and plan for model monitoring and retraining.
What the Interviewer Expects
- Map the business problem to a concrete ML objective
- Propose reasonable features and a baseline model
- Discuss basic model evaluation metrics
- Outline a simple serving architecture
Key Topics to Cover
Feedback loops and model retraining
Model serving and latency optimization
Feature engineering and feature stores
Data collection and labeling strategy
How to Approach This
- Start by clarifying functional and non-functional requirements with the interviewer.
- Estimate the scale: QPS, storage, bandwidth. This drives your design decisions.
- Draw a high-level architecture first, then deep dive into 1-2 critical components.
- Discuss trade-offs explicitly (e.g., consistency vs availability, SQL vs NoSQL).
- Address failure scenarios, monitoring, and how the system handles 10x traffic spikes.
Possible Follow-up Questions
- How would you debug a model that works well offline but poorly online?
- What is your model retraining strategy?
- How would you handle the cold start problem?
- How would you run A/B tests on different model versions?
Practice a Similar Problem on Codemia
Solve a related problem with our interactive workspace, get AI feedback, and view detailed solutions.
Solve on CodemiaSample Answer
Requirements
- Functional Requirements:
- A task scheduling system that can accept, queue, and execute millions of tasks in real-time.
- Ability to prioritize tasks based on user-defined parameters (e.g., ...
Capacity Estimation
Assuming we want to handle 10 million tasks per minute:
- RPS Calculation:
- 10 million tasks/minute = ~166,667 tasks/second.
- Latency Target:
- Targeting a 100ms latency, we can serve 10...
Submit Your Answer
Markdown supported