MoDNN: Local distributed mobile computing system for Deep Neural Network
The DOI matches its Crossref record; 19 of the 20 references it cites were found in a registry or an index.
Abstract
Although Deep Neural Networks (DNN) are ubiquitously utilized in many applications, it is generally difficult to deploy DNNs on resource-constrained devices, e.g., mobile platforms. Some existing attempts mainly focus on client-server computing paradigm or DNN model compression, which require either infrastructure supports or special training phases, respectively. In this work, we propose MoDNN - a local distributed mobile computing system for DNN applications. MoDNN can partition already trained DNN models onto several mobile devices to accelerate DNN computations by alleviating device-level computing cost and memory usage. TWo model partition schemes are also designed to minimize non-parallel data delivery time, including both wakeup time and transmission time. Experimental results show that when the number of worker nodes increases from 2 to 4, MoDNN can accelerate the DNN computation by 2.17-4.28 X. Besides the parallel execution, the performance speedup also partially comes from the reduction of the data delivery time, e.g., 30.02% w.r.t. conventional 2D-grids partition.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“MoDNN, proposed in [177], executes DL in a local distributed mobile computing system over a Wireless Local Area Network (WLAN).Each layer of DNNs is partitioned into slices to increase parallelism and to reduce memory footprint, and these slices are executed layerby-layer.By the execution parallelism among multiple end devices, the DL computation can be significantly accelerated.”
From Convergence of Edge Computing and Deep Learning: A Comprehensive Survey (2020) · cites this paper as [177]
In the citing paper
“The experiment shows that with two to four worker nodes, MoDNN accelerates the DNN model inference by 2.17-4.28×.”
From Edge Intelligence: Paving the Last Mile of Artificial Intelligence With Edge Computing (2019) · cites this paper as [85]
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 4×
“A representative example is V GG [1], which demonstrates state-of-the-art performance in ImageNet Large Scale Visual Recognition Challenge 2014 (ILSVRC14).”
BackgroundMethodInfluentialGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 1×
“1 depicts our measurement results of the computing time and the memory usage of different network layers from three popular DNN models running on smartphones – Lenet [9], Inception - BN [10], and V GG [1].”
BackgroundBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“1 depicts our measurement results of the computing time and the memory usage of different network layers from three popular DNN models running on smartphones – Lenet [9], Inception - BN [10], and V GG [1].”
BackgroundMapReduce: simplified data processing on large clusters
Dean, J., & Ghemawat, S. (2008) · Communications of the ACM · doi:10.1145/1327452.1327492 · Cited by 18,476 · cited 1×
“MapReduce is a programming model for simplifying parallel data processing in distributed systems [7].”
BackgroundBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S. (2024) · arXiv (Cornell University) · doi:10.57702/o9raffed · Cited by 14,214
A tutorial on spectral clustering
von Luxburg, U. (2007) · Statistics and Computing · doi:10.1007/s11222-007-9033-z · Cited by 10,349
Some simplified NP-complete graph problems
Garey, M. R., Johnson, D. S., & Stockmeyer, L. (1976) · Theoretical Computer Science · doi:10.1016/0304-3975(76)90059-1 · Cited by 2,059 · cited 1×
“If we consider the weight matrix as an undirected graph, generating k clusters with minimal connections between them is a NP hard problem [14].”
BackgroundMXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
Chen, T., Li, M., Li, Y., Lin, M., Wang, N., Wang, M., Xiao, T., Xu, B., Zhang, C., & Zhang, Z. (2015) · arXiv · doi:10.48550/arxiv.1512.01274 · Cited by 1,819 · cited 2×
“We modified and recompiled the MXNet libraries so that it can support Android systems with ARM architecture through JAVA Native Interface (JNI) [17].”
MethodCo-clustering documents and words using bipartite spectral graph partitioning
Dhillon, I. S. (2001) · Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/502512.502550 · Cited by 1,697 · cited 1×
“Therefore, spectral co-clustering algorithm is introduced to address this drawback by normalizing the original connection matrix A to A norm and performing Singular Value Decomposition (SVD) on A norm [16].”
MethodScaling distributed machine learning with the parameter server
Li, M., Andersen, D. G., Park, J. W., Smola, A. J., Ahmed, A. A., Josifovski, V., Long, J. E., Shekita, E. J., & Su, B.-Y. (2014) · Operating Systems Design and Implementation · doi:10.5555/2685048.2685095 · Cited by 1,021
Research works printed but not linked: 2
- han, 2015, Deep compression: Compressing deep neural network with pruning, trained quantization and huffman coding, Computing ResearchRepository, 2
- 0, Wifi Alliance Wifi Direct Specifications
Cited by
Items 201–300 of 327
Kurniawan, I. F., Asyhari, A. T., He, F., & Liu, Y. (2021) · Computer Communications · doi:10.1016/j.comcom.2021.07.020 · Cited by 4
DENNI: Distributed Neural Network Inference on Severely Resource Constrained Edge Devices
Sahu, R., Toepfer, R., Sinclair, M. D., & Duwe, H. (2021) · IEEE International Performance, Computing, and Communications Conference · doi:10.1109/ipccc51483.2021.9679448 · Cited by 8
Li, N., Iosifidis, A., & Zhang, Q. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838547 · Cited by 4
“Two examples of segment-based spatial partitioning, kernelsize based segmentation [7], [8] and computing-power based segmentation [1], [9], are proposed to partition an input tensor flexibly depending on the computation resource of each ES and the number of available ESs.”
BackgroundMethodInfluentialOdetola, T. A., Adeyemo, A. A., Khalid, F., & Hasan, S. R. (2023) · Microprocessors and Microsystems · doi:10.1016/j.micpro.2023.104827 · Cited by 6
Creating Robust Deep Neural Networks with Coded Distributed Computing for IoT
Hadidi, R., Cao, J., Asgari, B., & Kim, H. (2023) · IEEE International Conference on Edge Computing and Communications (EDGE) · doi:10.1109/edge60047.2023.00029 · Cited by 5
“As a result, pushing the frontier of DNNs computations to the edge is receiving a tremendous amount of interest both from academia [1]–[9] and from the industry with commercial edge-tailored hardware accelerators such as NVIDIA Jetson Nano, edge TPU, and Intel Movidius.”
BackgroundMistNet: A superior edge-cloud privacy-preserving training framework with one-shot communication
Guo, W., Cui, J., Li, X., Qu, L., Li, H., Hu, A., & Cai, T. (2023) · Internet of Things · doi:10.1016/j.iot.2023.100975 · Cited by 4
AI Workload Allocation Methods for Edge-Cloud Computing: A Review
Rafea, S. A., & Jasim, A. D. (2023) · Al-Iraqia Journal of Scientific Engineering Research · doi:10.58564/ijser.2.4.2023.125 · Cited by 6
Zha, Z., Yang, Y., Xia, Y., Wang, Z., Luo, B., Li, K., Ye, C., Xu, B., & Peng, K. (2024) · Applied Sciences · doi:10.3390/app14198656 · Cited by 4
“Mao et al. [21] proposed MoDNN—”
Liu, F., Wenbo, Z., Wang, Y., Changzhi, D., & Jiang, L. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.03903 · Cited by 4
“For instance, it takes 16 seconds on a mobile to complete an image recognition using VGG16 [25], which is intolerable for most applications [17].”
BackgroundScalable Smartphone Cluster for Deep LearningPreprint
Na, B., Jang, J., Park, S., Kim, S., Kim, J., Jeong, M. S., Kim, K. C., Heo, S., Kim, Y., & Yoon, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.12172 · Cited by 8
Adaptive model switching of collaborative inference for multi-CNN streams in UAV swarm
Li, Y., Qu, Y., Dong, C., Qin, Z., Zhang, L., & Wu, Q. (2025) · Chinese Journal of Aeronautics · doi:10.1016/j.cja.2025.103564 · Cited by 5
Dynamic Distribution of Edge Intelligence at the Node Level for Internet of Things
Mohammed, H., Odetola, T. A., Guo, N., & Hasan, S. R. (2021) · IEEE International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas47672.2021.9531765 · Cited by 5
“al [8] proposes a local distributed computing approach that utilizes parallel computation for the convolution and fully connected layers of the CNN model.”
BackgroundMethodFluid Dynamic DNNs for Reliable and Adaptive Distributed Inference on Edge Devices
Xun, L., Hu, M., Zhao, H., Singh, A. K., Hare, J. S., & Merrett, G. V. (2024) · Design, Automation, and Test in Europe · doi:10.23919/date58400.2024.10546704 · Cited by 4
A cosine similarity-based token subsampling method for vision transformer in cloud computing
Li, Q., Kaneko, H., & Meng, L. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-10718-w · Cited by 3
Quality/Latency-Aware Real-time Scheduling of Distributed Streaming IoT Applications
Barijough, K. M., Zhao, Z., & Gerstlauer, A. (2019) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3358209 · Cited by 4
“In the machine learning community, motivated by the increasing popularity of deep learning applications and limitations of embedded devices, there have been multiple proposals for specifically distributing inference of neural networks across network hosts [14, 20, 31].”
BackgroundNeural Network Partitioning for Fast Distributed Inference
Viramontes, R., & Davoodi, A. (2023) · International Symposium on Quality Electronic Design (ISQED) · doi:10.1109/isqed57927.2023.10129343 · Cited by 4
“MoDNN [9] is another intra-layer approach that demonstrates latency improvements by partitioning convolution and fully connected layers over multiple devices.”
BackgroundCollaborative non-chain DNN inference with multi-device based on layer parallel
Zhang, Q., Sun, S., Luo, J., Liu, M., Li, Z., Yang, H., & Wang, Y. (2023) · Digital Communications and Networks · doi:10.1016/j.dcan.2023.11.004 · Cited by 4
Securing Pseudo-Model Parallelism-Based Collaborative DNN Inference for Edge Devices
Adeyemo, A. A., Patel, P., Hasan, S. R., Rahaman, M. A., & Homsi, S. (2024) · IEEE Access · doi:10.1109/access.2024.3477293 · Cited by 6
“Several examples of sequential collaboration exist in the literatures [17], [18], [19], [20], [21], [22], and [23].”
BackgroundMultimedia Data Analysis With Edge Computing
Chen, S. (2021) · IEEE Multimedia · doi:10.1109/mmul.2021.3124292 · Cited by 3
A Novel Adaptive Computation Offloading Strategy for Collaborative DNN Inference over Edge Devices
Zhang, Y., Liu, X., Jia, X., Yuan, D., & Li, X. (2022) · IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA) · doi:10.1109/ispa-bdcloud-socialcom-sustaincom57177.2022.00055 · Cited by 4
“The other representative offloading strategies for comparison purposes are as follows: Neurosurgeon [12], DNNoff [16], MoDNN [20], and DeepWear [7].”
MethodMAFAT: Memory-Aware Fusing and Tiling of Neural Networks for Accelerated Edge Inference
Farley, J., & Gerstlauer, A. (2023) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-34214-1_7 · Cited by 5
“Meanwhile, distributed solutions such as [8] and [12] rely on partitioning convolutions into separate tasks and running them on separate devices, but they require additional communication and a network of devices.”
BackgroundMethodInfluentialEdgeSP: Scalable Multi-device Parallel DNN Inference on Heterogeneous Edge Clusters
Gao, Z., Sun, S., Zhang, Y., Mo, Z., & Zhao, C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-030-95388-1_21 · Cited by 10
Khaydarova, R., Fishchenko, V., Mouromtsev, D. I., Shmatkov, V., & Lapaev, M. V. (2020) · Conference of Open Innovations Association (FRUCT) · doi:10.23919/fruct48808.2020.9087457 · Cited by 4
“Existing works on deployment of CNN in SBS such as MoDNN [25] and DeepThings [24] are still under research and there is still no open-source version and supported solutions.”
BackgroundMethodHow Secure is Distributed Convolutional Neural Network on IoT Edge Devices?Preprint
Mohammed, H., Odetola, T. A., & Hasan, S. R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.09276 · Cited by 4
“Several partitioning schemes have been adopted in literature for optimum partitioning based on the computation, memory, and communication (bandwidth) capabilities of the nodes in the network [1], [5], [6], [11].”
BackgroundMethodInfluentialMemory-aware fusing and tiling of neural networks for accelerated edge inferencePreprint
Farley, J. (2021) · arXiv (Cornell University) · doi:10.26153/tsw/42343 · Cited by 4
DeepZoning: Re-accelerate CNN Inference with Zoning Graph for Heterogeneous Edge Cluster
Wang, J., Ma, R., Yang, X., Qi, Q., Zhuang, Z., Wang, J., Liao, J., & Guo, S. (2024) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3701995 · Cited by 2
“Cooperative Inference: Cooperative Inference: Researchers have explored the DNN inference completely at the edge with multiple mobile devices [12, 24, 42, 46, 47].”
InfluentialRACI: A Resource-Aware Cooperative Inference Framework on Heterogeneous Edge Devices
Wang, Z. F., Ren, A., Liu, D., Fang, H., Shi, J., Tan, Y., & Chen, X. (2024) · doi:10.1145/3676536.3676831 · Cited by 2
Self-citationCrowdNet: Adaptive Collaborative Inference for Dynamic Mobile Intelligent Service
Chen, G., Li, W., Fang, Y., Zhang, Y., & Lu, S. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3588100 · Cited by 2
“Ontheotherhand,theD2Dsynergy[28],[29],[30],[31],[32], [33], [34] facilitates direct communication and collaboration amongmobiledevicesformodelinference.”
Edge-PRUNE: A Dataflow-Based Framework for Distributed Signal Processing and Machine Learning
Boutellier, J. J., Tan, B., Nurmi, J., & Bhattacharyya, S. S. (2025) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2025.3598453 · Cited by 2
“In contrast to endpoint-server collaborative inference, several works [6], [29], [32] have proposed the distribution of DNN inference across multiple endpoint devices.”
Ismail, M., & Le, L. B. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11161364 · Cited by 3
“The first is to adopt the per DNN layer partition and distribution while the second is to use the per-unit partitioning technique where the task partitioning is applied to the rows or columns of a feature map [10], [11].”
Modifier guided resilient CNN inference enables fault-tolerant edge collaboration for IoT
Jamshidi, O., Abbasi, M., Ramazani, A., Shahraki, A. S., & Taherkordi, A. (2025) · Scientific Reports · doi:10.1038/s41598-025-28454-z · Cited by 3
“Models like BranchyNet 25 and MoDNN 26 extend this by constructing split models over layers or branches but still rely on the availability of each subcomponent and typically require cloud fallback when devices fail or quit early.”
Wang, Z., Zhao, Y., Wang, Y., Li, Y., Liu, Z., Qiu, C., Wang, X., & Hu, Q. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-25158-0_17 · Cited by 2
Persival: Using Delayed Remote Updates in a Distributed Mobile Simulation
Kässinger, J., Rosin, D., Dürr, F., Mehler, B., Hubatscheck, T., & Rothermel, K. (2023) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn58024.2023.10230159 · Cited by 2
“partition a single CNN and distribute it to multiple mobile devices [9].”
BackgroundDIDS: A distributed inference framework with dynamic scheduling capability
Yan, Y., Hu, Y., Cai, Q., Yang, W., & Li, K. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.07.037 · Cited by 3
Collaborative Inference for Deep Neural Networks in Edge Environments
Liu, M., Gu, Y., Dong, S., Wei, L., Liu, K., Yan, Y., Song, Y., Cheng, H., Tang, L., & Zhang, S. (2024) · KSII Transactions on Internet and Information Systems · doi:10.3837/tiis.2024.07.003 · Cited by 2
“If executed on a Nexus 5 smartphone, the task would take approximately 16 seconds [5].”
InfluentialDistributed convolutional neural network model on resource-constrained cluster
Khaydarova, R., Mouromtsev, D. I., Lapaev, M. V., & Fishchenko, V. (2020) · Scientific and technical journal of information technologies mechanics and optics · doi:10.17586/2226-1494-2020-20-5-739-746 · Cited by 2
Experimental test-bed for Computation Offloading for Cooperative Inference on Edge Devices
Bovee, N., Piccolo, S., Ho, S.-S., & Wang, N. (2023) · ACM/IEEE Symposium on Edge Computing (SEC) · doi:10.1145/3583740.3626814 · Cited by 4
Elastic Execution of Multi-Tenant DNNs on Heterogeneous Edge MPSoCs
Heidari, S., Ghasemi, M., Kim, Y. G., Wu, C.-J., & Vrudhula, S. B. K. (2024) · IEEE/ACM Symposium on Edge Computing (SEC) · doi:10.1109/sec62691.2024.00029 · Cited by 2
“Frameworks such as MoDNN [27] and DeepThings [45] partition the computation of DNNs between resource-constrained mobile devices using input data partitioning.”
Collaborative Inference Acceleration with Non-Penetrative Tensor Partitioning
Liu, Z., Xu, C., Lv, Z., Liu, Z., & Zhao, S. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10888994 · Cited by 2
“MoDNN [23] investigates the problem of additional communication overhead among IoT devices caused by image partition along multiple dimensions.”
Conflict-Resilient Incremental Offloading of Deep Neural Networks to the Edge of Smart Environment
Chen, Z., Xu, Z., Wan, J., Tian, J., Liu, L., & Zhang, Y. (2021) · Mathematical Problems in Engineering · doi:10.1155/2021/9985006 · Cited by 2
“Modnn [27] uses multiple edge devices to perform DNN inference and accelerate task inference by reducing the computational cost and memory usage of a single device.”
MethodLegion: Tailoring Grouped Neural Execution Considering Heterogeneity on Multiple Edge Devices
Choi, K., Lee, S., Kang, B. W., & Park, Y. (2021) · IEEE International Conference on Computer Design: VLSI in Computers and Processors (ICCD) · doi:10.1109/iccd53106.2021.00067 · Cited by 4
Kilcioglu, E., Stupia, I., & Vandendorpe, L. (2023) · IEEE Access · doi:10.1109/access.2023.3314381 · Cited by 2
“Since the energy consumption of convolutional layers dominates the energy consumption of other layers [34], [46], we mainly focus on the convolutional layer energy consumption in this paper.”
BackgroundMethodKilcioglu, E., Stupia, I., & Vandendorpe, L. (2023) · IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) · doi:10.1109/pimrc56721.2023.10293767 · Cited by 3
“The first collaborative computing scenario in this regard is MoDNN [8] by Mao et al. , who propose a distributed mobile device computing system to target latency minimization.”
BackgroundDistributed Edge Machine Learning Pipeline Scheduling with Reverse Auctions
Imes, C., King, D., & Walters, J. P. (2023) · International Conference on Fog and Mobile Edge Computing (FMEC) · doi:10.1109/fmec59375.2023.10306169 · Cited by 2
“MoDNN distributes DNN inference on resource-constrained mobile devices [19].”
BackgroundReducing Inference Latency with Concurrent Architectures for Image Recognition at Edge
Hadidi, R., Cao, J., Ryoo, M. S., & Kim, H. (2023) · IEEE International Conference on Edge Computing and Communications (EDGE) · doi:10.1109/edge60047.2023.00046 · Cited by 2
“hem complementary to further reduce inference latency. Concurrency & Distribution: With increasingly larger architectures and widespread usage of deep learning, distribution have gained attention [8, 11, 20,32,42]. Most of the techniques either exploit data or model parallelism [8,26]. Data parallelism only increa…”
BackgroundEfficient Deployment of Deep Neural Networks on Hardware Devices for Edge AI
Bouzidi, H. (2024) · theses.fr (ABES) · doi:10.70675/a25bbd6cz8a5bz4d93za232z46df005062c4 · Cited by 8
Cluster-based Object Detection System with Scalable Performance for Autonomous Driving
Kim, H., Lee, Y., Shin, J., & Kim, J.-C. (2023) · International Conference on Control, Automation and Systems (ICCAS) · doi:10.23919/iccas59377.2023.10316963 · Cited by 1
Hybrid Digital/Analog Memristor-based Computing Architecture for Sparse Deep Learning Acceleration
Zheng, Q., Li, S., Wang, Y., Li, Z., Chen, Y., & Li, H. H. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558703 · Cited by 1
“Bio-inspired model such as deep learning has achieved great interests with wide range of applications in diverse areas including: computer vision [1], [2], natural language processing [3], recommendation system [4] and hardware security [5]–[7] in recent years.”
BackgroundSelf-citationDeepDecompose: A Distributed inference framework for CNN on GPU-equipped Edge Clusters
Li, L., Li, Y., & Tan, H. (2024) · International Conference on Computing, Networks and Internet of Things (CNIOT) · doi:10.1145/3670105.3670200 · Cited by 2
“Mao et al. [8] proposed MoDNN, a system for DNN computation on mobile platforms.”
Enhancing Transmission Efficiency in Vision Transformer Edge Computing via Token Clustering
Li, Q., & Meng, L. (2024) · International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI) · doi:10.1109/iiki65561.2024.00019 · Cited by 1
Yao, Y., Wang, Z., Luo, P., Cheng, Y., Hou, J., & Li, X. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3607483 · Cited by 4
“Due to the significant disparity in computational power between edge nodes and end devices, existing strategies often assign minimal tasks to the end device, resulting inalargevolumeof intermediatedata—or eventheoriginal data-uploaded [11], [12], [13], [14], [20], [21].”
DNN Partitioning for GPU–CPU Heterogeneous Devices Based on Imitation Learning
Li, L., Liu, T., Zhao, M., Cong, P., Wu, Z., & Zhou, J. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3624053 · Cited by 1
“On one hand, cloud–edge collaborative methods [8], [9], [10], [12], [13], [14] o ffl oad DNN computations between edge devices and remote servers.”
ADaPS: Adaptive Data Partitioning to Parallelize CNN Inference on Resource-Constrained Hardware
Cuadrat, J. M., & Egger, B. (2025) · ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES) · doi:10.1145/3735452.3735532 · Cited by 1
“Search Type Heterogeneous Devices Partitioning Directions Communication Layer Fusion MoDNN [11] Layer by layer No H or W Centralized No DeepThings [25] Fixed partition No H or W Centralized No DeeperThings [20] Fixed partition No C Centralized Yes CoEdge [23] Layer by layer Yes H or W Centralized…”
InfluentialSplit CNN Inference on Networked Microcontrollers
Lu, J., Suresh, S., Liu, H., Hong, Q., & Wang, Q. (2026) · IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM) · doi:10.1109/wowmom69805.2026.00030 · Cited by 2
“1: Model partitioning strategy in COTS [16] (left) and MoDNN [17] (right).”
Optimal Partitioning of Distributed Neural Networks for Various Communication Environments
Jeong, J., & Yang, H. (2021) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic51459.2021.9415248 · Cited by 1
“1(a) shows an approach proposed by Mao et al. [4] where the KD (Knowledge Distillation) technique [5] was utilized to generate a small (or student) network from a high performance (teacher) network.”
MethodResearch on Cloud-edge Joint Task Inference Algorithm in Edge Intelligence
Zheng, Y. (2021) · 電腦學刊 · doi:10.53106/199115992021083204017 · Cited by 2
“Model partitions can be divided into two types: partitions between servers and devices [2] and partitions between devices [3].”
BackgroundNazir, A., Mir, R. N., & Qureshi, S. (2021) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.6593 · Cited by 1
An Efficient Scheme for Data Pattern Matching in IoT Networks
Ali, A. A., & Saraereh, O. A. (2022) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.025994 · Cited by 1
Deep Learning Distribution Model Using Osmotic Computing
Gurgel, L., Souza, A., Cacho, N., & Lopes, F. (2023) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc257844.2023.10293453 · Cited by 1
“• Edge distributed Running the model by distributing it among edge devices [15], [16].”
BackgroundMethodRen, J., Yu, Z., Xing, T., Cui, H., Chen, Y., & Guo, B. (2023) · IEEE Smart World Congress (SWC) · doi:10.1109/swc57546.2023.10449264 · Cited by 1
“On the one hand, a previous work [23] has suggested that it is not advisable to treat the extraction of different parts of a feature map as individual tasks, as this leads to excessive communication costs.”
BackgroundSelf-citationYannibelli, V., Hirsch, M., Toloza, J. M., Majchrzak, T. A., Grønli, T., Zunino, A., & Mateos, C. (2024) · IEEE Access · doi:10.1109/access.2024.3469641 · Cited by 1
“In [10], [11], and [23], like in our work, performed experiments include smartphones as battery-powered nodes, however the impact of the load balancing proposals on batteries were not assessed.”
Environment-Aware Dynamic Pruning for Pipelined Edge Inference
O’Quinn, A., Snedeker, C., Zhang, S., & Kline, J. (2025) · IEEE International Conference on Edge Computing and Communications (EDGE) · doi:10.1109/edge67623.2025.00018 · Cited by 2
CoCoI: Distributed Coded Inference System for Straggler Mitigation
Liu, X., Huang, C. H., & Tang, M. (2025) · International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt) · doi:10.23919/wiopt66569.2025.11123371 · Cited by 1
“Some considered edge-edge collaboration [8]–[10].”
An Edge Intelligence Framework for Resource Constrained Community Area Network
Oderhohwo, O., Mohammed, H., Odetola, T. A., Guo, T. N., Hasan, S. R., & Dogbe, F. (2020) · IEEE International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas48704.2020.9184597 · Cited by 1
“In [8] and [19], a distributed mobile system is designed for large scale CNNs using a partitioning scheme for parallel processing.”
MethodQUDOS: quorum-based cloud-edge distributed DNNs for security enhanced industry 4.0
Wallis, K., Reich, C., Varghese, B., & Schindelhauer, C. (2021) · IEEE/ACM International Conference on Utility and Cloud Computing (UCC) · doi:10.1145/3468737.3494094 · Cited by 2
“They are: (i) Profiling and estimation based [11, 26], (ii) linear programming based [7, 13], (iii) DNN structural modification based [17, 30], (iv) Approximation based [9], and (v) Benchmarking based [15, 27].”
MethodHeterogeneous platform-aware workload feature recognition for edge intelligence
Li, D., Li, Y., Hu, H., Zhang, T., & Jiang, C. (2022) · Physical Communication · doi:10.1016/j.phycom.2022.101620 · Cited by 2
Inference Acceleration with Adaptive Distributed DNN Partition over Dynamic Video Stream
Cao, J., Li, B., Fan, M., & Liu, H. (2022) · Algorithms · doi:10.3390/a15070244 · Cited by 1
Toward Efficient and Adaptive Design of Video Detection System with Deep Neural Networks
Mao, J., Yang, Q. X., Li, A., Nixon, K. W., Li, H. H., & Chen, Y. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3484946 · Cited by 1
“Although DNNs can achieve state-of-the-art accuracy when running on high-performance platforms [14–16], the low availability of computing resources on embedded platforms limits accuracy and/or viability of the DNNs in those contexts [17–19].”
BackgroundSelf-citationEdge-Cloud Collaboration for Human Activity Recognition on Multiple Subjects
Xiao, W., Xie, L., Ning, J., Fu, Z., Zhao, M., Lin, Z., & Lin, Q. (2022) · IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM) · doi:10.1109/wowmom54355.2022.00030 · Cited by 2
Hybrid Partitioning for Embedded and Distributed CNNs Inference on Edge Devices
Kaboubi, N., Letondeur, L., Coupaye, T., Desprez, F., & Trystram, D. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-28180-8_12 · Cited by 1
Collaborative Pipeline Using Opportunistic Mobile Resources via D2D for Computation-Intensive TasksPreprint
Guo, T. N., Mohammed, H., & Hasan, S. R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.09301 · Cited by 1
“Existing processing capabilities of NEDs are too limited to achieve real-time and dynamic Artificial intelligence (AI) inference [?], [29]–[31].”
BackgroundReducing Inference Latency with Concurrent Architectures for Image RecognitionPreprint
Hadidi, R., Cao, J., Ryoo, M. S., & Kim, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.07092 · Cited by 1
“hem complementary to further reduce inference latency. Concurrency & Distribution: With increasingly larger architectures and widespread usage of deep learning, distribution have gained attention [8, 11, 20,32,42]. Most of the techniques either exploit data or model parallelism [8,26]. Data parallelism only increa…”
BackgroundQin, Z., Xu, Z., Dong, Q., Chen, Y., & Chen, X. (2017) · International Conference on Computer Aided Design · doi:10.1109/iccad.2017.8203864
“When deployed on a Google smartphone, it takes ∼2ms for a single image classification, which is far from practical use like real-time video detection [8].”
MethodSelf-citationHMC-Based Accelerator Design For Compressed Deep Neural Networks
Min, C. (2020) · D-Scholarship@Pitt (University of Pittsburgh)
Creating Robust Deep Neural Networks With Coded Distributed Computing for IoT SystemsPreprint
Hadidi, R., Cao, J., & Kim, H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.04447
Image Depth Analysis: From Deep Learning to Parallel Cluster Computing
Ding, L., & Du, W. (2021) · International Conference on Inventive Research in Computing Applications (ICIRCA) · doi:10.1109/icirca51532.2021.9544606
Imperio: Cyber Foraging Framework for On-Device Inference based Mobile Applications
Wijayasekara, V. A., & Vekneswaran, P. (2021) · International Conference on Information and Automation for Sustainability (ICIAfS) · doi:10.1109/iciafs52090.2021.9605965 · Cited by 1
Machine Learning-based Optimal Framework for Internet of Things Networks
Alsafasfeh, M. H., Arida, Z. A., & Saraereh, O. A. (2022) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.024093
“Reference [24] used multiple IoT devices to perform the DNN inference for the first time, and achieved task inference acceleration by reducing the computational cost and memory usage of a single device.”
MethodJointCS: Joint Search for Deep Model Compression and Segmentation on Heterogeneous IoT Devices
Li, X., Guo, B., Liu, S., Qiu, C., Liang, Y., & Yu, Z. (2021) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads53394.2021.00059
A Black-Box Graph Partitioner for Generalized Deep Neural Network Parallelization
Cuadrat, J. M., Park, D., & Egger, B. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-29315-3_12 · Cited by 2
“MoDNN [7] divides the network and sends the data via Wi-Fi to different devices.”
Brief Announcement: Accelerate CNN Inference with Zoning Graph at Dynamic Granularity
Ma, R., Yang, X., Qi, Q., Wang, J., Zhuang, Z., Wang, J., & Wang, X. (2023) · ACM Symposium on Parallelism in Algorithms and Architectures (SPAA) · doi:10.1145/3558481.3591311
“We select three highly modularized network architectures GoogLeNet [5], InceptionV3 [4], and ResNeXt [6] for evaluation and compare the performances with the MoDNN [3], AOFL [8], CoEdge[7], local deployment.”
MethodSynchronization-Aware NAS for an Efficient Collaborative Inference on Mobile Platforms
Kang, B. W., Wohn, J., Lee, S., Park, S., Noh, Y., & Park, Y. (2023) · ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems (LCTES) · doi:10.1145/3589610.3596284
“Channel partitioning [23, 34, 41] divides weights into disjoint sets without redundancy; however, it requires data synchronization at every layer owing to the layer dependency.”
BackgroundMethodLü, X., & Zhuo, C. (2023) · China Semiconductor Technology International Conference (CSTIC) · doi:10.1109/cstic58779.2023.10219320
“Additionally, the system's performance is compared to that of MoDNN to assess its efficiency.”
BackgroundMethodInfluentialIntra-layer model division for Edge Machine Learning Computational Needs
Hamed, D., ElHayani, M., & Soubra, H. (2023) · Novel Intelligent and Leading Emerging Sciences Conference (NILES) · doi:10.1109/niles59815.2023.10296725
Deep spiking neural networks with integrate and fire neuron using steep switching device
Woo, S. Y., Pak, S., & Lee, S. (2024) · Solid-State Electronics · doi:10.1016/j.sse.2024.108860 · Cited by 1
DeepSlicing: Collaborative and Adaptive CNN Inference With Low Latency
Zhang, S., Chen, Y., Zhang, S., & Chen, Z. (2023) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3693-0230-9.ch006
Self-citationZhang, M., Shi, H., & Ma, R. (2024) · Sustainability · doi:10.3390/su16041599 · Cited by 1
Toward Timely and Reliable DNN Inference in Vehicular Edge Intelligence
Liu, K., Dai, P., Lee, V. C. S., Ng, J. K., & Son, S. H. (2023) · Springer eBooks · doi:10.1007/978-981-99-9647-6_10
Interference-Aware Optimization of Energy-Efficient Relay-Assisted Collaborative CNN Execution
Kilcioglu, E., Stupia, I., & Vandendorpe, L. (2023) · IEEE Virtual Conference on Communications (VCC) · doi:10.1109/vcc60689.2023.10474727
“This paper emphasizes the energy consumption of convolutional layers, as they predominantly contribute to energy usage over other layers [6], [11].”
BackgroundFederated Edge Computing Strategy for Fault Tolerance in Distributed Deep Neural NetworksPreprint
Jamshidi, O., Abbasi, M., Ramazani, A., Khosravi, M. R., & Taherkordi, A. (2024) · Research Square · doi:10.21203/rs.3.rs-4168386/v1
Shen, J., Tziritas, N., & Theodoropoulos, G. K. (2024) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-57808-3_3 · Cited by 1
“Auto-Split [1] ✓ ✓ two-way partitioning with module-wise quantization PTEENet [7] ✓ adaptive early-exiting MoDNN [11] ✓ Linear module weight parititioning k -way DAG Partitioning [9] ✓ k -way partitioning Edgent [8] ✓ ✓ two-way partitioning and adaptive early-exiting BranchyGNN[15] ✓ ✓ ✓ two-way…”
BackgroundMethodBelej, O., Nestor, N., Artyshchuk, I., Pytel, K., Spas, N., & Polietaiev, M. (2024) · International Conference on Perspective Technologies and Methods in MEMS Design · doi:10.1109/memstech63437.2024.10620061
“In [1], the authors proposed MoDNN, a locally distributed mobile computing framework for DNN applications.”
BackgroundPeccia, F. N., Viehl, A., & Bringmann, O. (2024) · Design, Automation, and Test in Europe · doi:10.23919/date58400.2024.10546603
“In Section V, ablation studies of the different parameters of the algorithm are analysed and DIAPASON is compared against the commonly used baseline MoDNN [5] and the state-of-the-art approach CoopAI [6] using numerical simulations.”
BackgroundMethodInfluentialCreating Distributed Artificial Neural Networks Based on Orthogonal Transformations
Vershkov, N., Babenko, M. G., Lutsenko, V. V., & Kuchukova, N. N. (2024) · Proceedings of the Institute for System Programming of RAS · doi:10.15514/ispras-2024-36(4)-5
“Furthermore, modular artificial neural networks (ANNs) have been developed on multiple computational nodes [7], [8], creating distributed computing structures.”
A Dynamic Distributed Scheduler for DNN Inference on the Edge
Hu, F., & Mishra, S. (2024) · International Symposium on Parallel Computing and Distributed Systems (PCDS) · doi:10.1109/pcds61776.2024.10743972
“Various strategies have been devised to distribute a pre-trained DNN across multiple mobile devices, aiming to speed up DNN inference on these devices [13]–[16].”
BackgroundPreetha Evangeline, D., Sahu, A. R., & Anandhakumar, P. (2024) · International Symposium on Networks, Computers and Communications (ISNCC) · doi:10.1109/isncc62547.2024.10759060
Collaborative Inference for Adaptive DNN Pipeline-Aware Alignment
Ren, J., Liu, L., Bi, R., Li, S., Li, S., & Lv, X. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-0055-7_25
Advances in Partitioning Techniques: A Prospective towards Artificial Intelligence
Guggari, S., Umadevi, V., & Kadappa, V. (2025) · Chapman & Hall/CRC eBooks · doi:10.1201/9781003471905
Создание искусственных нейронных сетей для работы в распределенных вычислительных средах
Vershkov, N., Kuchukova, N. N., & Lutsenko, V. V. (2025) · International Workshop on Information, Computation, and Control Systems for Distributed Environments (ICCS-DE) · doi:10.47350/iccs-de.2025.07
Meng, H., Hu, Y., Liu, K., & Guan, J. (2025) · Future Generation Computer Systems · doi:10.1016/j.future.2025.108084
Metadata
- Authors
- Mao, J., Chen, X., Nixon, K. W., Krieger, C., & Chen, Y.
- Affiliations
- University of Pittsburgh (US)
- George Mason University (US)
- University of Maryland, College Park (US)
- Venue
- Design, Automation & Test in Europe Conference & Exhibition (DATE), 2017
- Publisher
- IEEE
- Published
- 2017-03
- Pages
- 1396-1401
- OpenAlex
- W2612193523
- Cited by
- 331
- References
- 20
- Citations by year
- 2026: 21 · 2025: 35 · 2024: 47 · 2023: 62 · 2022: 43 · 2021: 51 · 2020: 38 · 2019: 17 · 2018: 11 · 2017: 6
- FWCI
- 15.69
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Design, Automation, and Test in Europe (publishedVersion)
- Funders
- National Science Foundation
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- deep neural network, mobile computing, distributed computing, model partitioning, resource-constrained devices, computational acceleration, parallel execution, memory reduction, worker nodes
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.