Bookmarks
LazyLog: A New Shared Log Abstraction for Low-Latency Applications
Shared logs offer linearizable total order across storage shards. However, they enforce this order eagerly upon ingestion, leading to high latencies.
applicative-mental-models
The text discusses the importance of understanding program performance for effective optimization. It emphasizes that while most optimizations may not be necessary, being aware of critical performance paths is essential. The author provides latency numbers to help programmers grasp the impact of different operations on performance.
Subcategories
- applications (9)
- compression (9)
- computer_vision (8)
- deep_learning (94)
- ethics (2)
- generative_models (25)
- interpretability (17)
- natural_language_processing (24)
- optimization (7)
- recommendation (2)
- reinforcement_learning (11)
- supervised_learning (1)