Using Bandwidth Throttling to Quantify Application Sensitivity to Heterogeneous Memory

Foyer, Clément; Goglin, Brice

doi:10.1109/mchpc54807.2021.00008

Cited by 3 publications

(2 citation statements)

References 11 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For graph workloads, we use six applications from the Ligra [14] graph processing framework on an RMAT scale 24 input graph. Additionally, to measure memory bandwidth we use STREAM triad, and to measure memory latency we use a pointerchasing benchmark [15]. For comprehensive coverage of HPC workloads, we select seven applications from the "Seven Dwarfs" [12] of major numerical algorithms.…”

Section: B Workloadsmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Emerging CXL-enabled Memory Pooling for HPC Systems

Wahlgren¹,

Gokhale²

2022

Preprint

View full text Add to dashboard Cite

Current HPC systems provide memory resources that are statically configured and tightly coupled with compute nodes. However, workloads on HPC systems are evolving. Diverse workloads lead to a need for configurable memory resources to achieve high performance and utilization. In this study, we evaluate a memory subsystem design leveraging CXL-enabled memory pooling. Two promising use cases of composable memory subsystems are studied -fine-grained capacity provisioning and scalable bandwidth provisioning. We developed an emulator to explore the performance impact of various memory compositions. We also provide a profiler to identify the memory usage patterns in applications and their optimization opportunities. Seven scientific and six graph applications are evaluated on various emulated memory configurations. Three out of seven scientific applications had less than 10% performance impact when the pooled memory backed 75% of their memory footprint. The results also show that a dynamically configured highbandwidth system can effectively support bandwidth-intensive unstructured mesh-based applications like OpenFOAM. Finally, we identify interference through shared memory pools as a practical challenge for adoption on HPC systems.

show abstract

Section: B Workloadsmentioning

confidence: 99%

“…Alternatively, another memory system emulation technique is the use of the Linux resctrl interface to configure memory bandwidth throttling in x86 CPUs [15]. The authors study memory bandwidth sensitivity in a set of HPC workloads.…”

Section: Related Workmentioning

confidence: 99%