Cello: Efficient computer systems optimization with predictive early termination and censored regression
Sample-efficient machine learning (SEML) has been widely applied to find optimal latency
and power tradeoffs for configurable computer systems. Instead of randomly sampling from …
and power tradeoffs for configurable computer systems. Instead of randomly sampling from …
Scope: Safe exploration for dynamic computer systems optimization
Modern computer systems need to execute under strict safety constraints (eg, a power limit),
but doing so often conflicts with their ability to deliver high performance (ie minimal latency) …
but doing so often conflicts with their ability to deliver high performance (ie minimal latency) …
Performance optimization of machine learning inference under latency and server power constraints
G Chen, X Wang - 2022 IEEE 42nd International Conference on …, 2022 - ieeexplore.ieee.org
Power capping is an important technique for high-density servers to safely oversubscribe the
power infrastructure in a data center. However, power capping is commonly accomplished …
power infrastructure in a data center. However, power capping is commonly accomplished …
CAMEO: A Causal Transfer Learning Approach for Performance Optimization of Configurable Computer Systems
Modern computer systems are highly configurable, with hundreds of configuration options
that interact, resulting in an enormous configuration space. As a result, optimizing …
that interact, resulting in an enormous configuration space. As a result, optimizing …
Generative and multi-phase learning for computer systems optimization
Machine learning and artificial intelligence are invaluable for computer systems
optimization: as computer systems expose more resources for management, ML/AI is …
optimization: as computer systems expose more resources for management, ML/AI is …
OptimML: Joint Control of Inference Latency and Server Power Consumption for ML Performance Optimization
G Chen, X Wang - ACM Transactions on Autonomous and Adaptive …, 2024 - dl.acm.org
Power capping is an important technique for high-density servers to safely oversubscribe the
power infrastructure in a data center. However, power capping is commonly accomplished …
power infrastructure in a data center. However, power capping is commonly accomplished …
Dynamic computation migration at the edge: Is there an optimal choice?
In the era of Fog computing where one can decide to compute certain time-critical tasks at
the edge of the network, designers often encounter a question whether the sensor layer …
the edge of the network, designers often encounter a question whether the sensor layer …
Distance-based sampling of software configuration spaces
C Kaltenecker, A Grebhahn… - 2019 IEEE/ACM 41st …, 2019 - ieeexplore.ieee.org
Configurable software systems provide a multitude of configuration options to adjust and
optimize their functional and non-functional properties. For instance, to find the fastest …
optimize their functional and non-functional properties. For instance, to find the fastest …
Machine learning-based prediction for dynamic architectural optimizations
R Vazquez, A Gordon-Ross… - 2019 Tenth International …, 2019 - ieeexplore.ieee.org
Embedded system complexity is rapidly evolving, becoming more desktop-system-like,
requiring more complex optimization methods to adhere to more stringent design constraints …
requiring more complex optimization methods to adhere to more stringent design constraints …
Performance optimization on big. little architectures: A memory-latency aware approach
W Wolff, B Porter - The 21st ACM SIGPLAN/SIGBED Conference on …, 2020 - dl.acm.org
The energy demands of modern mobile devices have driven a trend towards heterogeneous
multi-core systems which include various types of core tuned for performance or energy …
multi-core systems which include various types of core tuned for performance or energy …