A5047043813- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- High-Energy Particle Collisions Research
- Software-Defined Networks and 5G
- Radioactive contamination and transfer
- Nuclear physics research studies
- Stochastic processes and statistical mechanics
- Software System Performance and Reliability
- Statistical Mechanics and Entropy
- Scientific Computing and Data Management
- Radioactivity and Radon Measurements
Central China Normal University
1995
Over the last few years, at ByteDance, our compute infrastructure scale has been expanding significantly due to expedited business growth. In this journey, meet hyper-scale growth, some groups resorted managing their own stack running different scheduling systems such as Kubernetes, YARN which created two major pain points: increasing resource fragmentation across and inadequate elasticity between workloads of priorities. Isolation (and management) leads inefficient utilization prevents us...
At internet scale companies like ByteDance, data is generated and consumed at enormously high speed by many different applications. Achieving low latency on such big jobs an important problem. However, the naive approach of aggregating all required a job to single location not always feasible in geo-distributed environment. Similarly, existing approaches scheduling often try minimize WAN usage, which may come cost latency. Another crucial element ensure resource load balancing among DCs,...