All Tags
AWS
ai
algorithm-design
architecture
browser
cloud
cloud-efficiency
cloud-principles
cost-reduction
data-centric
data-compression
data-processing
deployment
design
documentation
edge-computing
email-sharing
energy-efficiency
energy-footprint
enterprise-optimization
green-ai
hardware
libraries
llm
locality
machine-learning
maintainability
management
measured
microservices
migration
mobile
model-optimization
model-training
multi-objective
network-traffic
parameter-tuning
performance
queries
rebuilding
scaling
services
storage-optimization
strategies
tabs
template
testing
workloads
Tactic: [Adjust CPU Load Balancing]
Tactic sort:
Awesome Tactic
Type: Architectural Tactic
Category: resource-allocation
Title
[Adjust CPU Load Balancing]
Description
Employ CPU load balancing strategies that use energy- and capacity-aware scheduling algorithms—such as Energy-Aware Scheduling (EAS) or Capacity-Aware Scheduling (CAS)—rather than relying solely on CPU time fairness. These schedulers account for both task characteristics (e.g., resource intensity, utilization) and system-level energy optimization. Development teams can also implement more granular task-to-core assignment strategies by grouping vCPUs and dynamically distributing tasks based on their computational demands. Monitoring tools are needed to track CPU utilization at a fine-grained level, enabling decisions about where and when to allocate tasks for optimal performance and sustainability
Participant
Scientific software developers
Related software artifact
Linux CPU scheduler (e.g., CFS, EAS, CAS), task allocation and runtime management systems, kernel monitoring utilities
Context
HPC software running on multi-core or heterogeneous hardware platforms (e.g., clusters with diverse threading or frequency capacities), where default fair-share CPU scheduling doesn't account for energy or utilization optimization
Software feature
Task-to-core scheduling, CPU resource distribution, execution load balancing
Tactic intent
To reduce idle time and energy waste by allocating tasks intelligently across vCPUs based on utilization patterns and task profiles
Target quality attribute
Energy efficiency
Other related quality attributes
< unknown >
Measured impact
Energy and performance efficiency
Source
Stoico, Vincenzo and Voronovs, Dmitrijs and Malavolta, Ivano and Lago, Patricia, How Does Parallelism Impact the Energy Efficiency and Performance of High-Performance Scientific Software? The Case of Haddock (February 13, 2025). (DOI: http://dx.doi.org/10.2139/ssrn.5137167)Graphical representation
