by Mark D. Hill on Jan 12, 2026 | Tags: Accelerators, Memory, Modelling
TL;DR: Latency-tolerant architectures, e.g., GPUs, increasingly use memory/storage hierarchies, e.g., for KV Caches to speed Large-Language Model AI inference. To aid codesign of such workloads and architectures, we develop the simple PipeOrgan analytic model for...
Read more...
by Zixuan Wang, Suyash Mahar, Luyi Li, Jangseon Park, Jinpyo Kim, Theodore Michailidis, Yue Pan, Mingyao Shen, Tajana Rosing, Dean Tullsen, Steven Swanson, Jishen Zhao on Dec 1, 2025 | Tags: AlphaFold, CXL, Fabric, Heterogeneous Systems, Memory, Profiling
This is the second article in the series, following our first blog in Dec 2023: Tuning the Symphony of Heterogeneous Memory Systems Modern applications are increasingly memory hungry. Applications like Large-Language Models (LLM), in-memory databases, and data...
Read more...
by Geraldo Oliveira, Juan Gómez-Luna, Onur Mutlu on Jan 22, 2025 | Tags: Emerging Technology, Memory, Processing-in-Memory
Processing-in-Memory (PIM) is a computing paradigm that aims to overcome the data movement bottleneck (i.e., wasted execution cycles and energy, resulting from the back-and-forth data movement between memory units and compute units) by making memory (and storage)...
Read more...
by Hyeran Jeon, Dong Li, Jie Ren on May 31, 2023 | Tags: Architecture, Conference, CXL, Heterogeneous and Composable Memory, hpca, Memory
Introduction Memory systems are evolving into heterogeneous and composable architectures. Heterogeneous and Composable Memory (HCM) offers a feasible solution for terabyte- or petabyte-scale systems, addressing the performance and efficiency demands of emerging...
Read more...
by Richard L. Sites on Dec 19, 2022 | Tags: Architecture, Data Centers, Memory, Performance
When I worked at Google, fleet-wide profiling revealed that 25-35% of all CPU time was spent just moving bytes around: memcpy, strcmp, copying between user and kernel buffers in network and disk I/O, hidden copy-on-write in soft page faults, checksumming, compressing,...
Read more...