Demand for in-memory data processing of large-scale data is expanding, and expectations
for storage-class memories (SCMs) are increasing accordingly. SCM achieves low standby
power and higher density compared to DRAM. However, SCM is relatively slower than
DRAM and requires more dynamic power. Therefore, it is necessary to improve speeds and
reduce power usage by SCM by performing memory hierarchical control such as power-
efficient prefetch control according to application memory access characteristics. However …

• Emergence of new applications such as Big Data analytics, deep learning for artificial
intelligence (AI) and CPS/IoT have increased demands for in-memory data processing of
large-scale data.• In-memory processing used to require large-capacity DRAM main
memory, but DRAM-only approach is hitting the limit due to DRAM's capacity scaling issue
and significant background power.• With new type byte-addressable high-speed non-volatile
memories, or storage-class memories (SCMs), we can explore high performance, low …