Processing in memory: the tipping point
Overview & Synopsis
The tipping point for adoption of PIM is imminent for three main reasons: • Firstly, PIM avoids the von Neumann bottleneck, a fundamental limitation to the effective use of computing systems for a large range of important data-centric applications. • Secondly, it matches the community's requirement for efficient application acceleration by reducing the amount of expensive data transfers, and demonstrates orders of magnitude improvements in performance and energy. • Thirdly, PIM has reached a high technological readiness confirmed by various industrial prototypes and products. Wide acceptance of PIM, however, depends on our ability to create an ecosystem in which a number of PIM approaches can be designed and evaluated, leading to the selection of potential winners and adoption by system architects and end users.
hal.science)
← / → for pages • + / - for zoom • 0 to fit width
Powered by Mozilla PDF.js v3.11
Document Information
| Document Title | Processing in memory: the tipping point |
|---|---|
| Author / Organization | Petar Radojković, Paul Carpenter, Pouya Esmaili-Dokht, Rémy Cimadomo, Henri-Pierre Charles, Sebastian Abu, Paolo Amato |
| Subject & Discipline | Multidisciplinary Academic Open Research |
| Categories | Science & Mathematics |
| Publication Year | 2021 |
| License / Rights | Open Access / Creative Commons |
| Resource Verification | Listed PDF (Verification Pending) |
| Original Host Website |
HAL Open Science
(hal.science)
|
| File Format | application/pdf (Portable Document Format) |
| Language | EN |
| Catalog ID | DOC-43133 |
| Original Source Link | Visit Source Page on hal.science » |