Explore the vast range of titles available.

With the increasing capacity and cost-efficiency of DRAM in multi-core environments, in-memory databases have emerged as fundamental solutions for delivering high performance. The index structure is a crucial component of the in-memory database, which, leveraging fast access to DRAM, plays an important role in the performance improvement and scalability of in-memory databases. A skiplist is one of the most widely used in-memory index structures and it has been adopted by popular databases. However, skiplists suffer from poor performance due to their structural limitations. In this work, we propose ESL, a high-performance and scalable skiplist. ESL efficiently enhances the performance of traverse operations by optimizing index levels for the CPU cache. With CPU cache-optimized index levels, we synergistically leverage a combination of exponential and linear searches. In addition, ESL reduces synchronization overhead by updating the index levels asynchronously, while tolerating inconsistencies. In our YCSB evaluation, ESL improves throughput by up to 2.8× over other skiplists in high-level evaluations. ESL also shows lower tail latency than other skiplists by up to 35×. Also, ESL consistently shows higher throughput in our real-world workload evaluation.

Nonlinear physical systems hold great promise for energy‐efficient and low‐hardware‐cost information processing. However, their computational capabilities remain constrained by the complexity and tunability of system nonlinearity. Here we report a dual‐ferroelectric gate‐tunable memristor with a dipole coupling effect, achieving enlarged hysteresis, rich temporal dynamics, and nonvolatile heterosynaptic plasticity. By harnessing the dynamic nonlinearity of the dual‐ferroelectric memristor, multimodal reservoir computing with an in‐material fusion strategy has been achieved, which is demonstrated with a multimodal object recognition task. By exploring the static nonlinearity of the dual‐ferroelectric memristor, nonlinear in‐memory computing is realized with gate‐tunable nonlinear functions, which successfully accelerates the Euclidean distance computation in the K‐means clustering task. This work achieves strong coupling between the intrinsic physical dynamics and computational functionalities, offering new opportunities for more efficient hardware‐accelerated systems. A dual‐ferroelectric gate‐tunable memristor with dipole coupling is designed, which demonstrates enlarged hysteresis, rich temporal dynamics, and nonvolatile heterosynaptic plasticity. It enables multimodal reservoir computing via dynamic nonlinearity for object recognition and nonlinear in‐memory computing via static nonlinearity for K‐means clustering acceleration, paving the way for efficient hardware computing systems.

The rapid proliferation of video data from various sources underscore the pressing need for effective Content-based Video Retrieval (CBVR) systems. Traditional retrieval methodologies are increasingly inadequate for managing the complexities and scale of video big data, which necessitates the development of advanced distributed computing frameworks. This study identifies and addresses critical challenges in CBVR , specifically the implementation of lambda architecture for the retrieval of both streaming and batch video data, the enhancement of in-memory analytics for video data structures, and the efficient indexing of heterogeneous video features. We propose , a novel scale-out system which integrates state-of-the-art big data technologies with deep learning algorithms. The system architecture is inspired by lambda principles and is designed to facilitate both near real-time and offline video indexing and retrieval. Key contributions of this research include: (1) the formulation of a lambda-style architecture tailored for video big data, (2) the development of an in-memory processing framework that provides a high-level abstraction for video analytics, (3) the introduction of a unified distributed indexer, termed Distributed Encoded Deep Feature Indexer (DEFI), capable of indexing multi-type features from both streaming and batch video datasets, and (4) a comprehensive bottleneck analysis of the proposed system. Performance evaluations utilizing three benchmark datasets demonstrate the system’s effectiveness, revealing insights into performance bottlenecks related to storage, video stream acquisition, processing, and indexing. This research provides a foundational framework for scalable and efficient video analytics, significantly advancing the state-of-the-art in cloud-based CBVR systems.

Healthcare data in images, texts and other unstructured formats have continued to grow exponentially while generating storage concerns. Even though there are other complexities, volume complexity is a major challenge for Disk-Resident technique in storage optimization. Hence, this research aimed to empirically compare the efficiency of Disk-Resident and In-Memory single structure database technique (as opposed to multiple structure In-Memory database), using descriptive and inferential big data analytical approaches. The essence was to discover a more cost-effective storage option for healthcare big data. Data from Nigerian Health Insurance Scheme (NHIS) alongside sample patients’ history from Made-in-Nigeria Primary Healthcare Information System (MINPHIS) which included patients’ investigation, patients’ bio-data and patients’ diagnoses were the primary data for this research. An implementation of both Disk-Resident and single structure In-Memory resident data storage was carried out on these big data sources. After storage, each quantity of data items stored for different data items in Disk-Resident was then compared with that of single structure In-Memory resident system using size of items as comparison criteria and different analyses made.The results obtained showed that single structure In-Memory technique conserved up to 90.57% of memory spaces with respect to the traditional (common) Disk-Resident technique for text data items. This shows that with this In-Memory technique, an improved performance in terms of storage was obtained.

Currently, Kronecker Matrix–Matrix Multiplication play a crucial role in many advanced applications across science and engineering, such as Quantum Computing (Tensor Representation of Quantum States, Quantum Gate Construction), Machine Learning and Data Science (Kernel Methods, Tensor Decompositions), and Signal and Image Processing (Multi-dimensional Filtering, Compression Algorithms). However, the implementation of the Kronecker Matrix–Matrix Multiplication increasingly relies on systems with enhanced computational capabilities. Specifically, current implementations expend large amounts of external memory and requires a large number of processing units to perform this operation. As is commonly acknowledged, cutting-edge high-performance computing schemes still faces limitations in terms of energy and performance due to the bottleneck in data transfer between processing units and memory. To mitigate this limitation, memory processing units (MPUs) enable direct computation on in-memory data, reducing latency and eliminating the need for data transfer. On the other hand, spiking neural P systems, with their inherent parallelism and distributed processing capabilities, are therefore well-suited as foundational components for implementing such memory architectures efficiently. From the mathematical point of view, we present for the first time a neural, synaptic, and dendritic model to support the Kronecker Matrix–Matrix multiplication. To this end, the proposed spiking neural P system with their cutting-edge variants, such as anti-spikes, communication on request, synaptic weights, and dendritic–axonal delays, facilitates the creation of neural memory cells and spike-based routers. Hence, these elements potentially allow the design of novel processing memory architectures that markedly enhance data transfer efficiency between computational units and memory.

Two-dimensional van der Waals (2D vdW) ferromagnets possess outstanding scalability, controllable ferromagnetism, and out-of-plane anisotropy, enabling the compact spintronics-based non-volatile in-memory computing (nv-IMC) that promises to tackle the memory wall bottleneck issue. Here, by employing the intriguing room-temperature ferromagnetic characteristics of emerging 2D Fe3GeTe2 with the dissimilar electronic structure of the two spin-conducting channels, we report on a new type of non-volatile spin-orbit torque (SOT) magnetic tunnel junction (MTJ) device based on Fe3GeTe2/MgO/Fe3GeTe2 heterostructure, which demonstrates the uni-polar and high-speed field-free magnetization switching by adjusting the ratio of field-like torque to damping-like torque coefficient in the free layer. Compared to the conventional 2T1M structure, the developed 3-transistor-2-MTJ (3T2M) cell is implemented with the complementary data storage feature and the enhanced sensing margin of 201.4% (from 271.7 mV to 547.2 mV) and 276% (from 188.2 mV to 520 mV) for reading “1” and “0”, respectively. Moreover, superior to the traditional CoFeB-based MTJ memory cell counterpart, the 3T2M crossbar array architecture can be executed for AND/NAND, OR/NOR Boolean logic operation with a fast latency of 24 ps and ultra-low power consumption of 2.47 fJ/bit. Such device to architecture design with elaborated micro-magnetic and circuit-level simulation results shows great potential for realizing high-performance 2D material-based compact SOT magnetic random-access memory, facilitating new applications of highly reliable and energy-efficient nv-IMC.

With the emergence of GPS-equipped portable devices and Online Social Networks, geo-tagged textual data have been highly produced on a continuous basis, which can provide important information for various applications, such as marketing, disaster response, and so on. Therefore, processing continuous spatial-keyword queries over streaming data is a hot topic for the research community nowadays. However, applying such queries to moving objects is computationally expensive due to the frequent updates of objects’ information that will continuously change the queries’ answers. Few research works focus on processing spatial-keyword queries over moving objects, so this problem demands more exploration by research. This paper proposes Lagic; a cloud-based solution scheme to process continuous spatial-keyword range queries over moving objects. Lagic is the first model that provides an exact solution to the problem and minimizes the overhead on users’ devices. A parallelized in-memory indexing structure is proposed to ensure the efficiency and scalability of Lagic. Short-term Safe Regions and a new approach for Buffer Regions are presented to reduce the number of required computations to update queries’ answer sets in an incremental manner. Evaluations show that Lagic can reduce the total processing time to seven folds less than a baseline model. It also provides better computational scalability and efficiency. Furthermore, Lagic shows stability in continuous running time against variations of queries’ and objects’ attributes.

Advanced data structures is a core course in Computer Science which most graduate program in Computer Science, Computer Science and Engineering, and other allied engineering disciplines, offer during the first year or first semester of the curriculum. The objective of this course is to enable students to have the much-needed foundation for advanced technical skill, leading to better problem-solving in their respective disciplines. Although the course is running in almost all the technical universities for decades, major changes in the syllabus have been observed due to the recent paradigm shift of computation which is more focused on huge data and internet-based technologies. Majority of the institute has been redefined their course content of advanced data structure to fit the current need and course material heavily relies on research papers because of nonavailability of the redefined text book advanced data structure. To the best of our knowledge well-known textbook on advanced data structure provides only partial coverage of the syllabus. The book offers comprehensive coverage of the most essential topics, including: Part I details advancements on basic data structures, viz., cuckoo hashing, skip list, tango tree and Fibonacci heaps and index files. Part II details data structures of different evolving data domains like special data structures, temporal data structures, external memory data structures, distributed and streaming data structures. Part III elucidates the applications of these data structures on different areas of computer science viz, network, www, DBMS, cryptography, graphics to name a few. The concepts and techniques behind each data structure and their applications have been explained. Every chapter includes a variety of Illustrative Problems pertaining to the data structure(s) detailed, a summary of the technical content of the chapter and a list of Review Questions, to reinforce the comprehension of the concepts. The book could be used both as an introductory or an advanced-level textbook for the advanced undergraduate, graduate and research programmes which offer advanced data structures as a core or an elective course. While the book is primarily meant to serve as a course material for use in the classroom, it could be used as a starting point for the beginner researcher of a specific domain. Dr. Suman Saha had spent the last 14 years developing as a scientist in the recent research areas of Data and information science covering information retrieval, web mining, decision theory, social network analysis and big data technologies. He started his research in the field of web mining as a senior research scientist in the “Center for Soft Computing Research: A National Facility”, Indian Statistical Institute, Kolkata, India for a duration of almost five years. After that his research continued as Assistant Professor in the dept. of computer science, Jaypee University of Information Technology, Himachal, India in addition to the teaching and other departmental responsibilities for last eight years. He obtained his PhD from Jaypee University of Information Technology preceded by M.Tech in computer science, from Indian Statistical Institute and M.Sc. in Mathematics, from University of Calcutta. His thesis title is “Community Detection in Complex Network: Metric Space, Nearest Neighbor Search, Low-Rank Approximation and Optimality” During his last eight years stay at Jaypee University of Information Technology as assistant professor he had taught various courses like advanced web mining, cloud computing, advanced algorithm, fundamentals of algorithm, advanced data structure and many others. He is supervising 2 PhD students and guided around 15 master thesis as well as around 50 bachelor thesis. Dr. Shailendra Shukla has completed “MS-(Information Security)” from “Indian Institute of Information Technology Allahabad”, and then completed PhD from “Indian Institute of Technology Patna” in computer science. His doctorial work is based on “On Boundary Detection and Localization in Wireless Sensor Networks”. In this work he proposed a collection of networking algorithms which addresses the security problems like routing in Internet of Things, localization, boundary node detection (surveillance), virtual coordinate assignment (Geography routing/localizations), cyber physical systems, monitoring and surveillance. He has published articles in various publication houses like in Elsevier, Springer, IEEE. Currently he is working as an assistant professor at Jaypee University Waknaghat. He is supervising 2 PhD students and guided 5 master student. I Part One: Theoretical Advancements. Introduction. O(1) Search by Hashing. O(log(n)) ordered search (Trees & Lists). Find set, find min & find word. II Part Two: Evolving Paradigms. Evolving paradigms of data structures. Spatial Data Structures. Temporal Data Structures. External Memory Data Structures. Distributed Data Structure. Synopsis Data Structures. III Part Three: Recent Applications. Introduction to applications. Application to Cryptography. Application to IR and WWW. Application to Data science. Application to Network and IOT. Application to System. Application to Database. Application to Image and Graphics. IV Bibliography and Index. Bibliography. First index. Second index.

Large‐scale machine‐learning (ML) algorithms require extensive memory interactions. Managing or reducing data movement can significantly increase the speed and efficiency of many ML tasks. Towards this end, the authors devise an energy efficient in‐memory computing (IMC) kernel for linear classification and design an initial prototype. The authors achieve a power savings of over 6.4 times than a conventional discrete system while improving reliability by 54.67%. The authors employ a split‐data‐aware technique to manage process, voltage, and temperature variations and to achieve fair trade‐offs between energy efficiency, area requirements, and accuracy. The authors utilise a trimodal architecture with a hierarchical tree structure to further decrease power consumption. The authors also explore alternatives to the hierarchical tree structure with a significantly reduced number of linear regression blocks, while maintaining a competitive classification accuracy. Overall, the scheme provides a fast, energy efficient, and competitively accurate binary classification kernel.

Driven by the increasing requirements of high-performance computing applications, supercomputers are prone to containing more and more computing nodes. Applications running on such a large-scale computing system are likely to spawn millions of parallel processes, which usually generate a burst of I/O requests, introducing a great challenge into the metadata management of underlying parallel file systems. The traditional method used to overcome such a challenge is adopting multiple metadata servers in the scale-out manner, which will inevitably confront with serious network and consistence problems. This work instead pursues to enhance the metadata performance in the scale-up manner. Specifically, we propose to improve the performance of each individual metadata server by employing GPU to handle metadata requests in parallel. Our proposal designs a novel metadata server architecture, which employs CPU to interact with file system clients, while offloading the computing tasks about metadata into GPU. To take full advantages of the parallelism existing in GPU, we redesign the in-memory data structure for the name space of file systems. The new data structure can perfectly fit to the memory architecture of GPU, and thus helps to exploit the large number of parallel threads within GPU to serve the bursty metadata requests concurrently. We implement a prototype based on BeeGFS and conduct extensive experiments to evaluate our proposal, and the experimental results demonstrate that our GPU-based solution outperforms the CPU-based scheme by more than 50% under typical metadata operations. The superiority is strengthened further on high concurrent scenarios, e.g., the high-performance computing systems supporting millions of parallel threads.