MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization

@inproceedings{MemScope_RTSS25,
  title     = {{MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization}},
  author    = {Ghaemi, Golsana and Franco, Gabriel and Taram, Kazem and Mancuso, Renato},
  booktitle = {46th IEEE Real-Time Systems Symposium (RTSS 2025)},
  year      = 2025,
  month     = dec,
  address   = {Boston, USA},
  url       = {https://cs-people.bu.edu/rmancuso/files/papers/MemScope_RTSS25.pdf}
}

CopyDownload .bib

TY  - CONF
AU  - Ghaemi, Golsana
AU  - Franco, Gabriel
AU  - Taram, Kazem
AU  - Mancuso, Renato
TI  - MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization
T2  - 46th IEEE Real-Time Systems Symposium (RTSS 2025)
PY  - 2025
DA  - 2025/12//
CY  - Boston, USA
AB  - This paper presents an open-source kernel-level heterogeneous memory characterization framework (M EM S COPE ) for embedded systems. M EM S COPE enables precise characterization of the temporal behavior of available memory modules under configurable contention stress scenarios. M EM S COPE leverages kernel-level control over physical memory allocation, cache maintenance, CPU state, interrupts, and I/O device activity to accurately benchmark heterogeneous memory subsystems. This gives us the privilege to directly map pieces of contiguous physical memory and instantiate allocators, allowing us to finely control cores to create and eliminate interference. Additionally, we can minimize noise and interruptions, guaranteeing more consistent and precise results compared to equivalent user-space solutions. Running our Framework on a Xilinx Zynq UltraScale+ ZCU102 CPU-FPGA platform demonstrates its capability to precisely benchmark bandwidth and latency across various memory types, including PL-side DRAM and BRAM, in a multi-core system.
ER  - 

CopyDownload .ris

%0 Conference Paper
%A Ghaemi, Golsana
%A Franco, Gabriel
%A Taram, Kazem
%A Mancuso, Renato
%T MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization
%B 46th IEEE Real-Time Systems Symposium (RTSS 2025)
%D 2025
%C Boston, USA
%X This paper presents an open-source kernel-level heterogeneous memory characterization framework (M EM S COPE ) for embedded systems. M EM S COPE enables precise characterization of the temporal behavior of available memory modules under configurable contention stress scenarios. M EM S COPE leverages kernel-level control over physical memory allocation, cache maintenance, CPU state, interrupts, and I/O device activity to accurately benchmark heterogeneous memory subsystems. This gives us the privilege to directly map pieces of contiguous physical memory and instantiate allocators, allowing us to finely control cores to create and eliminate interference. Additionally, we can minimize noise and interruptions, guaranteeing more consistent and precise results compared to equivalent user-space solutions. Running our Framework on a Xilinx Zynq UltraScale+ ZCU102 CPU-FPGA platform demonstrates its capability to precisely benchmark bandwidth and latency across various memory types, including PL-side DRAM and BRAM, in a multi-core system.

CopyDownload .enw

[
    {
        "id": "MemScope_RTSS25",
        "type": "paper-conference",
        "title": "MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization",
        "author": [
            {
                "family": "Ghaemi",
                "given": "Golsana"
            },
            {
                "family": "Franco",
                "given": "Gabriel"
            },
            {
                "family": "Taram",
                "given": "Kazem"
            },
            {
                "family": "Mancuso",
                "given": "Renato"
            }
        ],
        "container-title": "46th IEEE Real-Time Systems Symposium (RTSS 2025)",
        "issued": {
            "date-parts": [
                [
                    2025,
                    12
                ]
            ]
        },
        "publisher-place": "Boston, USA",
        "abstract": "This paper presents an open-source kernel-level heterogeneous memory characterization framework (M EM S COPE ) for embedded systems. M EM S COPE enables precise characterization of the temporal behavior of available memory modules under configurable contention stress scenarios. M EM S COPE leverages kernel-level control over physical memory allocation, cache maintenance, CPU state, interrupts, and I/O device activity to accurately benchmark heterogeneous memory subsystems. This gives us the privilege to directly map pieces of contiguous physical memory and instantiate allocators, allowing us to finely control cores to create and eliminate interference. Additionally, we can minimize noise and interruptions, guaranteeing more consistent and precise results compared to equivalent user-space solutions. Running our Framework on a Xilinx Zynq UltraScale+ ZCU102 CPU-FPGA platform demonstrates its capability to precisely benchmark bandwidth and latency across various memory types, including PL-side DRAM and BRAM, in a multi-core system."
    }
]

CopyDownload .json

G. Ghaemi, G. Franco, K. Taram, and R. Mancuso, “MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization,” in 46th IEEE Real-Time Systems Symposium (RTSS 2025), Dec. 2025.

CopyDownload .txt

Ghaemi, G., Franco, G., Taram, K., & Mancuso, R. (2025). MemScope: Open-Source Kernel-Level Framework for Heterogeneous Memory Characterization. In 46th IEEE Real-Time Systems Symposium (RTSS 2025).

CopyDownload .txt