Flikker: saving DRAM refresh-power through critical data partitioning

By Song Liu, K. Pattabiraman, Thomas Moscibroda, B. Zorn. ASPLOS XVI, 2011

472 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Approximate storage in solid-state memories (2013, background): “storage [14, 24, 38], modern non-volatile memory technologies also exhibit properties that make them candidates for storing data approximately.”
Evaluating STT-RAM as an energy-efficient main memory alternative (2013, background): “In [37], non-critical data is refreshed at lower rates.”
Drammer: Deterministic Rowhammer Attacks on Mobile Platforms (2016, background): “Flikker [29] proposes to allocate critical data in memory regions with higher refresh rates than non-critical data.”
RAIDR: Retention-aware intelligent DRAM refresh (2012, background): “Hardware-software cooperative techniques have been proposed to decrease refresh rate and allow retention errors only in unused [11, 50] or non-critical [26] regions of memory, but these substantially complicate the operating system while still requiring signiﬁcant hardware support.”
A Survey of Techniques for Approximate Computing (2016, methodology): “Liu et al. [2012] use a SW-based ACT for saving refresh power in DRAM memories.”