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: [Choose a Suitable DVFS Policy]

Tactic sort: Awesome Tactic

Type: Architectural Tactic

Category: resource-allocation

Tags: performance scaling

Title

[Choose a Suitable DVFS Policy]

Description

Selects and configures a Dynamic Voltage and Frequency Scaling (DVFS) policy tailored to the computational demands of different software modules. This involves using OS-level CPU governors—such as the conservative governor for gradual frequency scaling or assigning high-frequency cores only to heavy-load modules—to reduce unnecessary energy usage while preserving performance for dSemanding tasks

Participant

Scientific software developers

Related software artifact

OS kernel parameters (e.g., CPU governor settings), task schedulers, vCPU resource configuration

Context

Scientific software with modular workflows (e.g., HADDOCK) running in controlled computing environments (e.g., cloud or HPC clusters), where tasks have heterogeneous resource demands and frequency scaling policies can be adjusted per workload

Software feature

CPU frequency control, Energy management

Tactic intent

To improve energy efficiency by aligning CPU frequency behavior with the actual resource demands of software modules, avoiding over-provisioning and reducing unnecessary energy spikes

Target quality attribute

Energy efficiency

Other related quality attributes

Performance efficiency

Measured impact

Energy consumption; Execution time; CPU and memory utilization

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)