Dipanjan Sengupta

Bio: Dipanjan Sengupta is an academic researcher from Intel. The author has contributed to research in topics: Graph (abstract data type) & Programming paradigm. The author has an hindex of 9, co-authored 20 publications receiving 258 citations. Previous affiliations of Dipanjan Sengupta include Georgia Institute of Technology & Adobe Systems.

GraphReduce: processing large-scale graphs on accelerator-based systems

Abstract: Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device's internal memory capacity. GraphReduce adopts a combination of edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and device. GraphReduce-based programming is performed via device functions that include gatherMap, gatherReduce, apply, and scatter, implemented by programmers for the graph algorithms they wish to realize. Extensive experimental evaluations for a wide variety of graph inputs and algorithms demonstrate that GraphReduce significantly outperforms other competing out-of-memory approaches.

GraphIn: An Online High Performance Incremental Graph Processing Framework

Abstract: The massive explosion in social networks has led to a significant growth in graph analytics and specifically in dynamic, time-varying graphs. Most prior work processes dynamic graphs by first storing the updates and then repeatedly running static graph analytics on saved snapshots. To handle the extreme scale and fast evolution of real-world graphs, we propose a dynamic graph analytics framework, GraphIn, that incrementally processes graphs on-the-fly using fixed-sized batches of updates. As part of GraphIn, we propose a novel programming model called I-GAS based on gather-apply-scatter programming paradigm that allows for implementing a large set of incremental graph processing algorithms seamlessly across multiple CPU cores. We further propose a property-based, dual-path execution model to choose between incremental or static computation. Our experiments show that for a variety of graph inputs and algorithms, GraphIn achieves upi?źto 9.3 million updates/sec and over 400$$\times $$ speedup when compared to static graph recomputation.

Deep Graph Similarity Learning for Brain Data Analysis

Abstract: We propose an end-to-end graph similarity learning framework called Higher-order Siamese GCN for multi-subject fMRI data analysis. The proposed framework learns the brain network representations via a supervised metric-based approach with siamese neural networks using two graph convolutional networks as the twin networks. Our proposed framework performs higher-order convolutions by incorporating higher-order proximity in graph convolutional networks to characterize and learn the community structure in brain connectivity networks. To the best of our knowledge, this is the first community-preserving graph similarity learning framework for multi-subject brain network analysis. Experimental results on four real fMRI datasets demonstrate the potential use cases of the proposed framework for multi-subject brain analysis in health and neuropsychiatric disorders. Our proposed approach achieves an average AUC gain of $75$% compared to PCA, an average AUC gain of $65.5$% compared to Spectral Embedding, and an average AUC gain of $24.3$% compared to S-GCN across the four datasets, indicating promising applications in clinical investigation and brain disease diagnosis.

Scheduling multi-tenant cloud workloads on accelerator-based systems

Abstract: Accelerator-based systems are making rapid inroads into becoming platforms of choice for high end cloud services. There is a need therefore, to move from the current model in which high performance applications explicitly and programmatically select the GPU devices on which to run, to a dynamic model where GPUs are treated as first class schedulable entities. The Strings scheduler realizes this vision by decomposing the GPU scheduling problem into a combination of load balancing and per-device scheduling. (i) Device-level scheduling efficiently uses all of a GPU's hardware resources, including its computational and data movement engines, and (ii) load balancing goes beyond obtaining high throughput, to ensure fairness through prioritizing GPU requests that have attained least service. With its methods, Strings achieves improvements in system throughput and fairness of up to 8.70x and 13%, respectively, compared to the CUDA runtime.

Multi-tenancy on GPGPU-based servers

Abstract: While GPUs have become prominent both in high performance computing and in online or cloud services, they still appear as explicitly selected 'devices' rather than as first class schedulable entities that can be efficiently shared by diverse server applications. To combat the consequent likely under-utilization of GPUs when used in modern server or cloud settings, we propose 'Rain', a system level abstraction for GPU "hyperthreading" that makes it possible to efficiently utilize GPUs without compromising fairness among multiple tenant applications. Rain uses a multi-level GPU scheduler that decomposes the scheduling problem into a combination of load balancing and per-device scheduling. Implemented by overriding applications' standard GPU selection calls, Rain operates without the need for application modification, making possible GPU scheduling methods that include prioritizing certain jobs, guaranteeing fair shares of GPU resources, and/or favoring jobs with least attained GPU services. GPU multi-tenancy via Rain is evaluated with server workloads using a wide variety of CUDA SDK and Rodinia suite benchmarks, on a multi-GPU, multi-core machine typifying future high end server machines. Averaged over ten applications, GPU multi-tenancy on a smaller scale server platform results in application speedups of up to 1.73x compared to their traditional implementation with NVIDIA's CUDA runtime. Averaged over 25 pairs of short and long running applications, on an emulated larger scale server machine, multi-tenancy results in system throughput improvements of up to 6.71x, and in 43% and 29.3% improvements in fairness compared to using the CUDA runtime and a naive fair-share scheduler.

Intelligent video surveillance: a review through deep learning techniques for crowd analysis

Abstract: Big data applications are consuming most of the space in industry and research area. Among the widespread examples of big data, the role of video streams from CCTV cameras is equally important as other sources like social media data, sensor data, agriculture data, medical data and data evolved from space research. Surveillance videos have a major contribution in unstructured big data. CCTV cameras are implemented in all places where security having much importance. Manual surveillance seems tedious and time consuming. Security can be defined in different terms in different contexts like theft identification, violence detection, chances of explosion etc. In crowded public places the term security covers almost all type of abnormal events. Among them violence detection is difficult to handle since it involves group activity. The anomalous or abnormal activity analysis in a crowd video scene is very difficult due to several real world constraints. The paper includes a deep rooted survey which starts from object recognition, action recognition, crowd analysis and finally violence detection in a crowd environment. Majority of the papers reviewed in this survey are based on deep learning technique. Various deep learning methods are compared in terms of their algorithms and models. The main focus of this survey is application of deep learning techniques in detecting the exact count, involved persons and the happened activity in a large crowd at all climate conditions. Paper discusses the underlying deep learning implementation technology involved in various crowd video analysis methods. Real time processing, an important issue which is yet to be explored more in this field is also considered. Not many methods are there in handling all these issues simultaneously. The issues recognized in existing methods are identified and summarized. Also future direction is given to reduce the obstacles identified. The survey provides a bibliographic summary of papers from ScienceDirect, IEEE Xplore and ACM digital library.

Mosaic: Processing a Trillion-Edge Graph on a Single Machine

Abstract: Processing a one trillion-edge graph has recently been demonstrated by distributed graph engines running on clusters of tens to hundreds of nodes. In this paper, we employ a single heterogeneous machine with fast storage media (e.g., NVMe SSD) and massively parallel coprocessors (e.g., Xeon Phi) to reach similar dimensions. By fully exploiting the heterogeneous devices, we design a new graph processing engine, named Mosaic, for a single machine. We propose a new locality-optimizing, space-efficient graph representation---Hilbert-ordered tiles, and a hybrid execution model that enables vertex-centric operations in fast host processors and edge-centric operations in massively parallel coprocessors.Our evaluation shows that for smaller graphs, Mosaic consistently outperforms other state-of-the-art out-of-core engines by 3.2-58.6x and shows comparable performance to distributed graph engines. Furthermore, Mosaic can complete one iteration of the Pagerank algorithm on a trillion-edge graph in 21 minutes, outperforming a distributed disk-based engine by 9.2×.

NeuGraph: Parallel Deep Neural Network Computation on Large Graphs

Abstract: Recent deep learning models have moved beyond low dimensional regular grids such as image, video, and speech, to high-dimensional graph-structured data, such as social networks, e-commerce user-item graphs, and knowledge graphs. This evolution has led to large graph-based neural network models that go beyond what existing deep learning frameworks or graph computing systems are designed for. We present NeuGraph, a new framework that bridges the graph and dataflow models to support efficient and scalable parallel neural network computation on graphs. NeuGraph introduces graph computation optimizations into the management of data partitioning, scheduling, and parallelism in dataflow-based deep learning frameworks. Our evaluation shows that, on small graphs that can fit in a single GPU, NeuGraph outperforms state-of-the-art implementations by a significant margin, while scaling to large real-world graphs that none of the existing frameworks can handle directly with GPUs. (Please stay tuned for further updates.)

Dynamic Contrast-Enhanced Magnetic Resonance Imaging in Oncology

Abstract: An Overview of Contrast-Enhanced MRI in Oncology.- Contrast Agents for Magnetic Resonance Imaging.- The Role of Blood Pool Contrast Agents in the Study of Tumor Pathophysiology.- Dynamic Susceptibility Contrast-Enhanced MRI in Oncology.- Quantification of Contrast-Enhancement in Dynamic Studies.- Pharmacokinetic Modelling of Contrast Kinetics in Dynamic MR Studies.- Imaging Techniques: Imaging Techniques for Dynamic Contrast-Enhanced Imaging. Consensus Recommendations for Imaging in Angiogenesis.- Clinical Applications: Dynamic Magnetic Resonance Imaging in Cerebral Malignancy. Dynamic Magnetic Resonance Imaging in Breast Tumours. Dynamic Contrast Imaging in Cervical Carcinoma. Dynamic Contrast-Enhanced Imaging in the Prostate. Contrast-Enhanced MR Imaging in Musculoskeletal Tumors. Contrast-Enhanced MR Imaging in the Liver.- Applications: The Use of Dynamic Contrast-Enhanced MRI Multicentre Trials. Applications of Dynamic Magnet Resonance in Drug Development.- Subject Index.