TY - GEN
T1 - Crystal
T2 - 43rd International Conference on Parallel Processing, ICPP 2014
AU - Knyaginin, Dmitry
AU - Gaydadjiev, Georgi N.
AU - Stenström, Per
PY - 2014/11/13
Y1 - 2014/11/13
N2 - Non-Volatile Memory (NVM) technologies can be used to reduce system-level execution time, energy, or cost but they add a new design dimension. Finding the best amounts of DRAM and NVM in hybrid main memory systems is a nontrivial design-time issue, the best solution to which depends on many factors. Such resource partitioning between DRAM and NVM can be framed as an optimization problem where the minimum of a target metric is sought, trends matter more than absolute values, and thus the precision of detailed modeling is overkill. Here we present Crystal, an analytic approach to early and rapid design-time resource partitioning of hybrid main memories. Crystal provides first-order estimates of system-level execution time and energy, sufficient to enable exhaustive search of the best amount and type of NVM for given workloads and partitioning goals. Crystal thus helps system designers to quickly find the most promising hybrid configurations for detailed evaluation. E.g., Crystal shows how for specific workloads higher system-level performance and energy efficiency can be achieved by employing an NVM with the speed and energy consumption of NAND Flash instead of a much faster and more energy efficient NVM like phase-change memory.
AB - Non-Volatile Memory (NVM) technologies can be used to reduce system-level execution time, energy, or cost but they add a new design dimension. Finding the best amounts of DRAM and NVM in hybrid main memory systems is a nontrivial design-time issue, the best solution to which depends on many factors. Such resource partitioning between DRAM and NVM can be framed as an optimization problem where the minimum of a target metric is sought, trends matter more than absolute values, and thus the precision of detailed modeling is overkill. Here we present Crystal, an analytic approach to early and rapid design-time resource partitioning of hybrid main memories. Crystal provides first-order estimates of system-level execution time and energy, sufficient to enable exhaustive search of the best amount and type of NVM for given workloads and partitioning goals. Crystal thus helps system designers to quickly find the most promising hybrid configurations for detailed evaluation. E.g., Crystal shows how for specific workloads higher system-level performance and energy efficiency can be achieved by employing an NVM with the speed and energy consumption of NAND Flash instead of a much faster and more energy efficient NVM like phase-change memory.
UR - http://www.scopus.com/inward/record.url?scp=84932636642&partnerID=8YFLogxK
U2 - 10.1109/ICPP.2014.18
DO - 10.1109/ICPP.2014.18
M3 - Conference contribution
AN - SCOPUS:84932636642
T3 - Proceedings of the International Conference on Parallel Processing
SP - 90
EP - 100
BT - Proceedings - 43rd International Conference on Parallel Processing, ICPP 2014
PB - Institute of Electrical and Electronics Engineers (IEEE)
Y2 - 9 September 2014 through 12 September 2014
ER -