Improving the error behavior of DRAM by exploiting its Z-channel property

In this paper, we present a new communication theoretic channel model for Dynamic Random Access Memory (DRAM) retention errors, that relies on the fully asymmetric retention error behavior of DRAM cells. This new model shows that the traditional approach is over pessimistic and we confirm this with real measurements of DDR3 and DDR4 DRAM devices. Together with an exploitation of the vendor specific true- and anti-cell structure, a low complexity bit-flipping approach is presented, that can largely increase DRAM's reliability with minimum overhead.

[1]  Song Liu,et al.  Flikker: saving DRAM refresh-power through critical data partitioning , 2011, ASPLOS XVI.

[2]  J. Lucas,et al.  Sparkk : Quality-Scalable Approximate Storage in DRAM , 2014 .

[3]  Kinam Kim,et al.  A New Investigation of Data Retention Time in Truly Nanoscaled DRAMs , 2009, IEEE Electron Device Letters.

[4]  Ariel J. Feldman,et al.  Lest we remember: cold-boot attacks on encryption keys , 2008, CACM.

[5]  Hiren D. Patel,et al.  Reverse-engineering embedded memory controllers through latency-based analysis , 2015, 21st IEEE Real-Time and Embedded Technology and Applications Symposium.

[6]  Mauro Olivieri,et al.  Introducing approximate memory support in Linux Kernel , 2017, 2017 13th Conference on Ph.D. Research in Microelectronics and Electronics (PRIME).

[7]  Kenneth G. Paterson,et al.  A Coding-Theoretic Approach to Recovering Noisy RSA Keys , 2012, IACR Cryptol. ePrint Arch..

[8]  L.G. Tallini,et al.  On the capacity and codes for the Z-channel , 2002, Proceedings IEEE International Symposium on Information Theory,.

[9]  Arnab Raha,et al.  Quality-aware data allocation in approximate DRAM* , 2015, 2015 International Conference on Compilers, Architecture and Synthesis for Embedded Systems (CASES).

[10]  Richard W. Hamming,et al.  Error detecting and error correcting codes , 1950 .

[11]  Arnab Raha,et al.  Quality Configurable Approximate DRAM , 2017, IEEE Transactions on Computers.

[12]  Jay M. Berger.  A Note on Error Detection Codes for Asymmetric Channels , 1961, Inf. Control..

[13]  B. Narasimham,et al.  A multi-bit error detection scheme for DRAM using partial sums with parallel counters , 2008, 2008 IEEE International Reliability Physics Symposium.

[14]  Hyun-Soo Park,et al.  23.4 An extremely low-standby-power 3.733Gb/s/pin 2Gb LPDDR4 SDRAM for wearable devices , 2017, 2017 IEEE International Solid-State Circuits Conference (ISSCC).

[15]  Norbert Wehn,et al.  Reverse Engineering of DRAMs: Row Hammer with Crosshair , 2016, MEMSYS.

[16]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[17]  Norbert Wehn,et al.  Efficient reliability management in SoCs - an approximate DRAM perspective , 2016, 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC).

[18]  Stefan Mangard,et al.  DRAMA: Exploiting DRAM Addressing for Cross-CPU Attacks , 2015, USENIX Security Symposium.

[19]  Jonghyuk Kim,et al.  23.2 A 5Gb/s/pin 8Gb LPDDR4X SDRAM with power-isolated LVSTL and split-die architecture with 2-die ZQ calibration scheme , 2017, 2017 IEEE International Solid-State Circuits Conference (ISSCC).

[20]  Luca G. Tallini.  Bounds on the capacity of the unidirectional channels , 2005, IEEE Transactions on Computers.

[21]  Liviu Miclea,et al.  Unidirectional error detection, localization and correction for DRAMs: Application to on-line DRAM repair strategies , 2011, 2011 IEEE 17th International On-Line Testing Symposium.