Unveiling the Real Performance of LPDDR5 Memories
Lukas Steiner, Matthias Jung, Michael Huonker, Norbert Wehn

TL;DR
This paper critically evaluates the actual bandwidth performance of LPDDR5 memories compared to LPDDR4, revealing that higher peak bandwidth does not always translate to real-world gains and some configurations may be suboptimal.
Contribution
It provides a comprehensive analysis of real bandwidth utilization of LPDDR5, highlighting conditions where upgrades may not yield expected performance improvements.
Findings
LPDDR5 does not always outperform LPDDR4 in real bandwidth.
Certain LPDDR5 configurations are unsuitable for specific workloads.
Peak bandwidth claims may not reflect actual performance.
Abstract
LPDDR5 is the latest low-power DRAM standard and expected to be used in various application fields. The vendors have published promising peak bandwidths up to 50 % higher than those of the predecessor LPDDR4. In this paper we evaluate the best-case and worst-case real bandwidth utilization of different LPDDR5 configurations and compare the results to corresponding LPDDR4 configurations. We also show that an upgrade from LPDDR4 to LPDDR5 does not always bring a bandwidth advantage and that some LPDDR5 configurations should be avoided for specific workloads.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsLow-power high-performance VLSI design · Parallel Computing and Optimization Techniques · Semiconductor materials and devices
