skip to main content


This content will become publicly available on June 14, 2024

Title: Resilient mobile distributed computing framework: a coded computing and named data networking approach
Award ID(s):
1910348
NSF-PAR ID:
10466673
Author(s) / Creator(s):
; ; ; ; ;
Publisher / Repository:
SPIE
Date Published:
Format(s):
Medium: X
Location:
Orlando, United States
Sponsoring Org:
National Science Foundation
More Like this
  1. In this paper, a flexible resource sharing paradigm is introduced, to enable the allocation of users’ computing tasks in a social cloud computing system offering both Virtual Machines (VMs) and Serverless Computing (SC) functions. VMs are treated as a safe computing resource, while SC due to the uncertainty introduced by its shared nature, is treated as a common pool resource, being susceptible to potential over-exploitation. These computing options are differentiated based on the potential satisfaction perceived by the user, as well as their corresponding pricing, while taking into account the social interactions among the users. Considering the inherent uncertainty of the considered computing environment, Prospect Theory and the theory of the Tragedy of the Commons are adopted to properly reflect the users’ behavioral characteristics, i.e., gain-seeking or loss-averse behavior, as well as to formulate appropriate prospect-theoretic utility functions, embodying the social-aware and risk-aware user’s perceived satisfaction. A distributed maximization problem of each user’s expected prospect-theoretic utility is formulated as a non-cooperative game among the users and the corresponding Pure Nash Equilibrium (PNE), i.e., optimal computing jobs offloading to the VMs and the SC, is determined, while a distributed low-complexity algorithm that converges to the PNE is introduced. The performance and key principles of the proposed framework are demonstrated through modeling and simulation. 
    more » « less
  2. This article presents C3SRAM, an in-memory-computing SRAM macro. The macro is an SRAM module with the circuits embedded in bitcells and peripherals to perform hardware acceleration for neural networks with binarized weights and activations. The macro utilizes analog-mixed-signal (AMS) capacitive-coupling computing to evaluate the main computations of binary neural networks, binary-multiply-and-accumulate operations. Without the need to access the stored weights by individual row, the macro asserts all its rows simultaneously and forms an analog voltage at the read bitline node through capacitive voltage division. With one analog-to-digital converter (ADC) per column, the macro realizes fully parallel vector–matrix multiplication in a single cycle. The network type that the macro supports and the computing mechanism it utilizes are determined by the robustness and error tolerance necessary in AMS computing. The C3SRAM macro is prototyped in a 65-nm CMOS. It demonstrates an energy efficiency of 672 TOPS/W and a speed of 1638 GOPS (20.2 TOPS/mm 2 ), achieving 3975 × better energy–delay product than the conventional digital baseline performing the same operation. The macro achieves 98.3% accuracy for MNIST and 85.5% for CIFAR-10, which is among the best in-memory computing works in terms of energy efficiency and inference accuracy tradeoff. 
    more » « less
  3. The number of female students in computing fields remain low despite the millions of dollars spent on research for attracting more female students. In order to attract more female students to these male dominated fields, we first have to understand in which educational years we are losing female students. For the purpose of this study, we utilized the data from [title blinded for anonymity] an NSF funded study. Approximately, 1650 students from three large public universities in Florida participated in this survey. The survey contained 39 questions around identity, field of study, as well as fields they wanted to pursue during middle school, high school, and college. The responses gathered through the Qualtrics survey system were analyzed in R by the research team. The research questions that guided this study were: (1) To what extent are female students interested in computing related fields at middle school, the beginning of high school, and the beginning of college? (2) How have these occupational pursuits changed over time? Do they differ for gender? The results of the study indicated a majority of female students that were attracted to the computing fields during middle school remained in those fields during high school and college years. However, there was no significant flow from other majors to the computing fields observed during the different educational years. 
    more » « less