Computing Nearer to Data

Coughlin, Tom; Tonti, W.

doi:10.1109/mc.2022.3171354

Cited by 5 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The current research is mainly for power planning considering wind power, scenery bundling, and other forms of new energy power plants connected to the grid [2,4,6] , such as research [3] for scenery marsh distributed power planning, and research [5] considering scenery joint consumption for power planning. The inclusion of concentrated solar power into the power source to be planned can well make up for the defects of new energy output and improve the output characteristics of thermal power units.…”

Section: Introductionmentioning

confidence: 99%

Joint Planning of CSP and New Energy Based on Bi-Level Programming Model

Chen

Zuo

Yang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

As a flexible unit that can respond quickly, CSP stations can be included in the planned power source of power supply planning, which can effectively improve the consumption of volatile new energy and improve resource utilization. This paper mainly studies the optimal planning of power grids containing CSP stations and fluctuating new energy sources and proposes joint power planning based on the bi-level programming model. The results of calculation examples show that the proposed bi-level programming model considering CSP stations is better in terms of economy, efficiency, and practicality and is of great importance to the improvement in the cleanliness of system operation and energy utilization.

show abstract

Section: Introductionmentioning

confidence: 99%

Joint Planning of CSP and New Energy Based on Bi-Level Programming Model

Chen

Zuo

Yang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

On Hardware Security and Trust for Chiplet-Based 2.5D and 3D ICs: Challenges and Innovations

Suzano,

Abouzeid,

Di Natale

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Semiconductor Architectures Enable Compute in Memory

Handy¹,

Coughlin²

2023

Computer

View full text Add to dashboard Cite

Compute in memory (CIM) was a strong theme at IEEE's International Solid-State Circuits Conference (ISSCC) in February. 1 A total of 19 CIM papers were chosen for presentation at this very selective event, which carefully picks only those research efforts that share the most innovative thinking. The developments in semiconductor technologies, such as CIM, will have a great impact on future computing technology and the software that manages that technology. COMBINING MEMORY AND PROCESSINGAll but two of the CIM chips highlighted in this article have been designed to push data-intensive inference tasks like image recognition to the edge. While this will lead to phenomenal reductions in the communication bandwidth between the edge device and the central server, conventional computing architectures would vastly increase endpoint energy consumption, and this would be impractical for most applications (and even impossible for many mobile devices). Neural network inference engines, particularly those that combine memory and processing, provide a compelling alternative since they perform basic recognition tasks while consuming very little energy.An earlier article, in the May issue of Computer, 2 explained the need for CIM but didn't go into concrete examples. This article highlights papers that show the difficulty of architecting a quality CIM chip for edge inference, where most of these devices are expected to be used, and gives some remarkable ideas of how to address them.

show abstract

Computing Nearer to Data

Cited by 5 publications

References 13 publications

Joint Planning of CSP and New Energy Based on Bi-Level Programming Model

Joint Planning of CSP and New Energy Based on Bi-Level Programming Model

On Hardware Security and Trust for Chiplet-Based 2.5D and 3D ICs: Challenges and Innovations

Semiconductor Architectures Enable Compute in Memory

Contact Info

Product

Resources

About