ByteDance's TIGE team is looking to solve the problem of high costs during peak periods for their FusionCDN system by devising strategies to maximize free bandwidth and reduce expenses while maintaining service quality.
Requirements
- Strong background in networking, CDN systems, and traffic optimization.
- Experience with programming languages like Go, Python, or similar.
- Knowledge of algorithm development, cost modeling, and real-time systems is a plus.
Responsibilities
- study the FusionCDN system's traffic scheduling mechanisms, IPT cost model, and operational constraints.
- conduct a literature review on CDN - related optimization strategies and work with the team to define project objectives, constraints, and key performance indicators.
- gather historical traffic data to identify peak periods and patterns, and build a simplified cost model for IPT charges.
- analyzing traffic characteristics and running simulations on different traffic distribution scenarios, you'll gain insights that will inform the subsequent design work.
- design and implement an algorithm to redistribute traffic during peak periods, making the most of the free bandwidth without exceeding the limit.
- test and validate the solution in a controlled environment, evaluating its cost - savings, service quality, and scalability.
- comparing IPT costs before and after implementation and assessing metrics like latency and throughput.
Other
- Final year Ph.D or recent Ph.D graduates in Computer Science, engineering or quantitative field
- Internships at ByteDance aim to offer students industry exposure and hands-on experience.
- Please state your availability clearly in your resume (Start date, End date).
