The Impact of Process Competition on Energy Consumption: Analysis and Modeling
2026-05-19T08:52:16Z•b80d93694c607d736041decdb1da41e5cc7da2e7772e4e0c5fd1081bb39a6305
BF16FP32-emulationGPUKV-cacheLLM-servingagentic-workflowscheckpointingcloud-computingcoded-computingdistributed-systemsenergy-modelingfault-tolerancegoodputheterogeneous-clustersload-balancingnode-healthobject-capability-kernelperformance-modelingrandomized-linear-algebraresource-contentionschedulingsimulationstraggler-detectiontensor-coresvideo-diffusion
What happened
Collection of recent distributed-systems and ML-systems papers covering resource contention and energy modeling, reduced-precision (BF16) emulation for FP32 GEMM on GPUs, workflow- and heterogeneity-aware scheduling for agentic LLM serving, goodput-optimized routing, unified simulation for LLM training/inference, coded and approximate distributed computing, tiered checkpointing for LLM training, straggler detection and node health, load balancing for video diffusion training, and a minimal kernel design for multi-party object coordination. Common themes are performance/efficiency across large,
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_dc
- Record identifier
- b80d93694c607d736041decdb1da41e5cc7da2e7772e4e0c5fd1081bb39a6305
- Enrichment time
- 2026-05-19T08:52:16Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.