Enabling a reliable STT-MRAM main memory simulation

Conference object English OPEN
Asifuzzaman, Kazi ; Sánchez-Verdejo, Rommel ; Radojković, Petar (2017)
  • Publisher: Association for Computing Machinery
  • Related identifiers: doi: 10.1145/3132402.3132416
  • Subject: Supercomputadors | STT-MRAM | Main memory | High performance computing | Processors, High performance | High-performance computing | :Enginyeria elèctrica [Àrees temàtiques de la UPC]
    acm: Hardware_MEMORYSTRUCTURES

STT-MRAM is a promising new memory technology with very desirable set of properties such as non-volatility, byte-addressability and high endurance. It has the potential to become the universal memory that could be incorporated to all levels of memory hierarchy. Although STT-MRAM technology got significant attention of various major memory manufacturers, to this day, academic research of STT-MRAM main memory remains marginal. This is mainly due to the unavailability of publicly available detailed timing parameters which are required to perform a cycle accurate main memory simulation. Our study presents a detailed analysis of STT-MRAM main memory timing and propose an approach to perform a reliable system level simulation of the memory technology. We seamlessly incorporate STT-MRAM timing parameters into DRAMSim2 memory simulator and use it as a part of the simulation infrastructure of the high-performance computing (HPC) systems. Our results suggests that, STT-MRAM main memory would provide performance comparable to DRAM, while opening up various opportunities for HPC system improvements. Most importantly, our study enables researchers to conduct reliable system level research on STT-MRAM main memory, and to explore the opportunities that this technology has to offer. This work was supported by BSC, Spanish Government through Programa Severo Ochoa (SEV-2015-0493), by the Spanish Ministry of Science and Technology through TIN2015-65316-P project and by the Generalitat de Catalunya (contracts 2014-SGR-1051 and 2014-SGR-1272). This work has also received funding from the European Union's Horizon 2020 research and innovation programme under ExaNoDe project (grant agreement No 671578). The authors wish to thank Terry Hulett, Duncan Bennett and Ben Cooke from Everspin Technologies Inc., for their technical support. Peer Reviewed
Share - Bookmark