… With support for fastbulkbitwise operations, we show that Buddy significantly shifts the trade-off in favor of bit vectors. To demonstrate this, we compare the performance of union, …
S Angizi, D Fan - 2019 IEEE/ACM International Conference on …, 2019 - ieeexplore.ieee.org
… We estimate the energy that DRAM chip consumes to perform the four bulkbit-wise … 5× faster than that of Ambit solution and 49× faster than GPU. This is mainly because of fast and …
V Seshadri, O Mutlu - arXiv preprint arXiv:1905.09822, 2019 - arxiv.org
… bulkbitwise operations completely inside main memory. Ambit exploits the internal organization and analog operation of DRAM-… With support for fastbulkbitwise operations, we show …
… We believe that the Ambit’s support for fast and ecient bulkbitwise operations can enable better design of other applications to take advantage of such operations, which would result in …
S Angizi, D Fan - arXiv preprint arXiv:1904.05782, 2019 - arxiv.org
… a fast (< 100ns) in-memory copy operation within DRAM sub-arrays, rather than using ∼ 1µs conventional operation in Von-Neumann computing systems, RowClone-Fast Parallel …
V Seshadri, O Mutlu - arXiv preprint arXiv:1610.09603, 2016 - arxiv.org
… that exploits DRAM technology to perform bulk copy and … work that uses DRAM to perform bulkbitwise AND and OR … a 4KB page of data 12.0x faster and with 74.4x less energy …
X Xin, Y Zhang, J Yang - 2020 IEEE International Symposium …, 2020 - ieeexplore.ieee.org
… • We present a lightweight mechanism to implement bulkbitwise operation in DRAM, … offer fast analytics in specific applications for databases. Traditionally, the bulkbitwise operation …
F Zhang, S Angizi, D Fan - 2021 58th ACM/IEEE Design …, 2021 - ieeexplore.ieee.org
… (2) A new in-DRAM computing circuit and architecture, termed as Max-PIM, is then proposed to support complete bulkbit-wise Boolean logic and optimized for our proposed min/max-in-…
… as the underlying bulk-bitwise technology as it can operate on both bitlines and wordlines [36], enabling a richer functionality compared to other (eg, DRAM-based) bulk-bitwise PIM. …