Spatiotemporal Locality Research Articles

Dynamic trajectory partition optimization method based on historical trajectory data

Partitioning dynamic trajectory data can improve the efficiency and accuracy of trajectory data processing, provide a foundation for trajectory data mining and analysis. However, with the continuous growth of trajectory data scales and the urgent demand for trajectory query efficiency and accuracy, partitioning methods have become crucial. The partitioning method of dynamic trajectory data faces significant challenges in terms of spatiotemporal trajectory locality, partition load balancing, and partition time. To address these challenges, we propose a method based on historical trajectory pre-partitioning, which can store data more effectively in distributed systems. We partition similar historical trajectory data to achieve preliminary partitioning of the data. In addition, we also construct a cost model to ensure that the workload of each partition is close to consistency. Extensive experiments have demonstrated the excellent partitioning efficiency and query efficiency achieved by our design compared to other partitioning methods.

Local propagation of visual stimuli in focus of attention

Fast reactions to changes in the surrounding visual environment require efficient attention mechanisms to reallocate computational resources to the most relevant locations in the visual field. In this paper, we present a biologically-plausible computational model of focus of attention that exhibits spatiotemporal locality and that is very well-suited for parallel and distributed implementations. Attention emerges as a wave propagation process originated by visual stimuli corresponding to details and motion information. The resulting field obeys the principle of “inhibition of return” so as not to get stuck in potential holes. The proposed model is obtained as a hyperbolic regularization of the Poisson equation to which it reduces in the limit of high speed of propagation. According to the MultiMatch algorithm for scanpaths comparison, the proposed model achieves very competitive results when considering dynamical input stimuli.

NCache: A Machine-Learning Cache Management Scheme for Computational SSDs

Inside a solid-state disk (SSD), cache stores frequently accessed data to shorten user-I/O response time and reduce the number of read/write operations in flash memory, thereby improving SSD performance and lifetime. Most existing cache schemes anchor in the spatiotemporal locality of I/O requests in workloads. In the face of a long-time workload, high performance and hit rate often get lost in these caching schemes. Flash memory-aware caching schemes trade hit ratio to prolong SSD lifetime. In this paper, we advocate for a machine learNing-based caching scheme named NCache to optimize both hit ratio and SSD performance. In NCache, we construct a machine learning (i.e., ML) model to predict whether data are re-accessed before being evicted from the cache. The cache replacement scheme preferentially evicts data that would not be accessed in the cache. The cache space is conserved for valid data that are likely to be repeatedly accessed. A pipelined scheme is implemented to accelerate the ML model, alleviating the time-cost of NCache. A double-linked list boosts the data addressing and cache replacement process. NCache is orthogonal to the existing caching schemes within the flash translation layer. The results validate NCache under a handful of real-world enterprise traces. Taking prn_0 as an example, NCache reduces the response time of LRU, CFLRU, GCaR_LRU, GCaR_CFLRU, and LCR by up to 15% with an average of 6.4%. The erase count is slashed by 16% at the maximum. Importantly, NCache is adroit at optimizing write amplification by up to 15.9%.

GIN: Graph-Based Interaction-Aware Constraint Policy Optimization for Autonomous Driving

Applying reinforcement learning to autonomous driving entails particular challenges, primarily due to dynamically changing traffic flows. To address such challenges, it is necessary to quickly determine response strategies to the changing intentions of surrounding vehicles. This letter proposes a new policy optimization method for safe driving using graph-based interaction-aware constraints. In this framework, the motion prediction and control modules are trained simultaneously while sharing a latent representation that contains a social context. To reflect social interactions, we illustrate the movements of agents in graph form and filter the features with the graph convolution networks. This helps preserve the spatiotemporal locality of adjacent nodes. Furthermore, we create feedback loops to combine these two modules effectively. As a result, this approach encourages the learned controller to be safe from dynamic risks and renders the motion prediction robust to abnormal movements. In the experiment, we set up a navigation scenario comprising various situations with CARLA, an urban driving simulator. The experiments show state-of-the-art performance on navigation strategy and motion prediction compared to the baselines.

DeResolver: A Decentralized Conflict Resolution Framework with Autonomous Negotiation for Smart City Services

As various smart services are increasingly deployed in modern cities, many unexpected conflicts arise due to various physical world couplings. Existing solutions for conflict resolution often rely on centralized control to enforce predetermined and fixed priorities of different services, which is challenging due to the inconsistent and private objectives of the services. Also, the centralized solutions miss opportunities to more effectively resolve conflicts according to their spatiotemporal locality of the conflicts. To address this issue, we design a decentralized negotiation and conflict resolution framework named DeResolver, which allows services to resolve conflicts by communicating and negotiating with each other to reach a Pareto-optimal agreement autonomously and efficiently. Our design features a two-step self-supervised learning-based algorithm to predict acceptable proposals and their rankings of each opponent through the negotiation. Our design is evaluated with a smart city case study of three services: intelligent traffic light control, pedestrian service, and environmental control. In this case study, a data-driven evaluation is conducted using a large dataset consisting of the GPS locations of 246 surveillance cameras and an automatic traffic monitoring system with more than 3 million records per day to extract real-world vehicle routes. The evaluation results show that our solution achieves much more balanced results, i.e., only increasing the average waiting time of vehicles, the measurement metric of intelligent traffic light control service, by 6.8% while reducing the weighted sum of air pollutant emission, measured for environment control service, by 12.1%, and the pedestrian waiting time, the measurement metric of pedestrian service, by 33.1%, compared to priority-based solution.

A data grouping model based on cache transaction for unstructured data storage systems

Cache prefetching technology has become the mainstream data access optimization strategy in the data centers. However, the rapidly increasing of unstructured data generates massive pairwise access relationships, which can result in a heavy computational burden for the existing prefetching model and lead to severe degradation in the performance of data access. We propose cache-transaction-based data grouping model (CTDGM) to solve the problems described above by optimizing the feature representation method and grouping efficiency. First, we provide the definition of the cache transaction and propose the method for extracting the cache transaction feature (CTF). Second, we design a data chunking algorithm based on CTF and spatiotemporal locality to optimize the relationship calculation efficiency. Third, we propose CTDGM by constructing a relation graph that groups data into independent groups according to the strength of the data access relation. Based on the results of the experiment, compared with the state-of-the-art methods, our algorithm achieves an average increase in the cache hit rate of 12% on the MSR dataset with small cache size (0.001% of all the data), which in turn reduces the number of data I/O accesses by 50% when the cache size is less than 0.008% of all the data.

Estimation of brain response to multimodal stimuli by index of spatiotemporal locality by magnetoencephalography

AbstractBrain‐computer interface (BCI) enables people who cannot move their own body freely to manipulate machines and helps their communication and life. Recent brain‐computer interface (BCI) uses multimodal stimuli to increase bit rate of the system, so it is important to reveal when and which part of the brain part is activated to discriminate target stimuli. Although magnetoencephalography (MEG) measures brain activity with high spatial and temporal resolution, there are a few indexes to estimate spatiotemporal locality of response. We propose an index to estimate spatiotemporal locality of response to multimodal stimuli of an oddball paradigm from magnetoencephalography (MEG) signals. The validity of proposed index is demonstrated by simulation. Discrimination task was conducted with different ratio of target/nontarget stimuli with visual and auditory location. Response to visual target stimuli was strong and made no spatiotemporal difference, so visual dominance was implied.

Estimation of Brain Response to Multimodal Stimuli by Index of Spatiotemporal Locality by Magnetoencephalography

Estimation of Brain Response to Multimodal Stimuli by Index of Spatiotemporal Locality by Magnetoencephalography

Analysis on spatio-temporal locality and degradation of the rogue wave in an optical fiber

In this research, we give an analysis on the conditions and causes of breaking rogue wave’s spatio-temporal locality and degradation of high-order rogue wave in an optical fiber by numerical simulation and analytic theory. Our simulation results show that both of increasing input power and third-order dispersion can undermine the spatio-temporal locality of the rogue wave. And they also cause the degradation of the high-order rogue wave as well. Then we find that the rogue wave clusters theory has an appropriate explanation for these phenomena. We consider that the essence of these phenomena is the interaction of rogue wave clusters on the spatio-temporal plane, in addition, give a detailed analysis of this by rogue wave triplet model. And the theoretical results are in good agreement with the simulation results in fiber.

Replication Strategy for Spatiotemporal Data Based on Distributed Caching System.

The replica strategy in distributed cache can effectively reduce user access delay and improve system performance. However, developing a replica strategy suitable for varied application scenarios is still quite challenging, owing to differences in user access behavior and preferences. In this paper, a replication strategy for spatiotemporal data (RSSD) based on a distributed caching system is proposed. By taking advantage of the spatiotemporal locality and correlation of user access, RSSD mines high popularity and associated files from historical user access information, and then generates replicas and selects appropriate cache node for placement. Experimental results show that the RSSD algorithm is simple and efficient, and succeeds in significantly reducing user access delay.

Fat Caches for Scale-Out Servers

Emerging scale-out servers are characterized by massive memory footprints and bandwidth requirements. On-chip stacked DRAM caches have been proposed to provide the required bandwidth for manycore servers through caching of secondary data working sets. However, the disparity between provided capacity and working set sizes precludes their effective deployment in servers, calling for high-capacity cache architectures. High-capacity caches--enabled by the emergence of high-bandwidth memory technologies--exhibit high spatiotemporal locality due to coarse-grained access patterns and long cache residency periods stemming from skewed dataset access distributions. The observed spatiotemporal behavior favors a page-based organization that naturally exploits spatial locality while minimizing tag storage requirements and enabling a practical in-SRAM tag array architecture. By storing tags in SRAM, caches avoid the complexity of in-DRAM metadata found in state-of-the-art DRAM caches.

Dynamic replication to reduce access latency based on fuzzy logic system

In a distributed environment, limited available bandwidth resources lead a high data access latency. Replication is a popular method that can upgrade the access performance and increase the data availability. However, unreasonable replication would cause over-consumption of system resources and finally a further deterioration on data access latency. So, In this paper, a theoretical model of access latency optimization with replication is presented firstly, which complement the blank space, and then a well-designed dynamic replication strategy is proposed, which consists of three algorithms: replica selection algorithm, replica layout algorithm and replica replacement algorithm. Replica selection algorithm selects the optimal replica with a hierarchical time cost based on the derivation of the theoretical model. Replica layout algorithm selects the optimal node for placing the replica based on the spatio-temporal locality of data access. Replica replacement algorithm, in which the fuzzy logic system is introduced originally, deletes replica when the available storage space is insufficient. FLSDR is tested by OptorSim and experimental results show that FLSDR achieves better performance in comparison with other algorithms in terms of mean job execution time, computing resource usage, number of data scheduling between clusters and number of replicas.

Can We Group Storage? Statistical Techniques to Identify Predictive Groupings in Storage System Accesses

Storing large amounts of data for different users has become the new normal in a modern distributed cloud storage environment. Storing data successfully requires a balance of availability, reliability, cost, and performance. Typically, systems design for this balance with minimal information about the data that will pass through them. We propose a series of methods to derive groupings from data that have predictive value, informing layout decisions for data on disk. Unlike previous grouping work, we focus on dynamically identifying groupings in data that can be gathered from active systems in real time with minimal impact using spatiotemporal locality. We outline several techniques we have developed and discuss how we select particular techniques for particular workloads and application domains. Our statistical and machine-learning-based grouping algorithms answer questions such as “What can a grouping be based on?” and “Is a given grouping meaningful for a given application?” We design our models to be flexible and require minimal domain information so that our results are as broadly applicable as possible. We intend for this work to provide a launchpad for future specialized system design using groupings in combination with caching policies and architectural distinctions such as tiered storage to create the next generation of scalable storage systems.

DISCOURSES OF ROUTINE AND AFFECTS IN THE TEXT OF REGIONAL SPACE (on the example of Volgograd region)

The article discusses social space, which is the region of the text that is defined by a set of messages sent to settlements, streets, squares, monuments, buildings, and a system of signs that reflect possible interpretations of the actors, due to the peculiarities of the subjective state, language and experience. Using the methods of discursive and semiotic interpretation, we can distinguish two components of a heterogeneous, multi-code text — space of routine and space of affects. Space of routine presupposes a rational perception of space, rational response, it is a formation of traditions, attitudes, norms of conduct governing the behavior of social actors in a specific spatio-temporal locality. Changes in the perception of the usual attributes of the regional landscape can in a certain sense and individual context trigger affective reactions. The followihg factors form the space of the affects: sensual perception of space, emotional stimulation and intrapsychic representation, the sense of injury/cure, dangers/safety, negativity/ pleasure, shame/pride, or a combination of them. Dualism in the perception of space, its dialectical nature, congruence, flexibility, variability, the conflict of local and individual meanings, existing practices, patterns, mentalities and cultures, situations, the media are a condition of existence, and labeling of the regional text, updating its discursiveness and research.

Distributed cache replacement method for geospatial data using spatiotemporal locality-based sequence

Specific features of tile access patterns can be applied in a cache replacement strategy to a limited distributed high-speed cache for the cloud-based networked geographic information services (NGISs), aiming to adapt to changes in the access distribution of hotspots. By taking advantage of the spatiotemporal locality, the sequential features in tile access patterns, and the cache reading performance in the burst mode, this article proposes a tile sequence replacement method, which involves structuring a Least Recently Used (LRU) stack into three portions for the different functions in cache replacement and deriving an expression for the temporal locality and popularity of the relevant tile to facilitate the replacement process. Based on the spatial characteristics of both the tiles and the cache burst mode with regard to reading data, the proposed method generates multiple tile sequences to reflect spatiotemporal locality in tile access patterns. Then, we measure the caching value by a technique based on a weighted-based method. This technique draws on the recent access popularity and low caching costs of tile sequences, with the aim of balancing the temporal and spatial localities in tile access. It ranks tile sequences in a replacement queue to adapt to the changes in accessed hotspots while reducing the replacement frequency. Experimental results show that the proposed method effectively improves the hit rate and utilization rate for a limited distributed cache while achieving satisfactory response performance and high throughput for users in an NGIS. Therefore, it can be adapted to handle numerous data access requests in NGISs in a cloud-based environment.

RRB vector: a practical general purpose immutable sequence

State-of-the-art immutable collections have wildly differing performance characteristics across their operations, often forcing programmers to choose different collection implementations for each task. Thus, changes to the program can invalidate the choice of collections, making code evolution costly. It would be desirable to have a collection that performs well for a broad range of operations. To this end, we present the RRB-Vector, an immutable sequence collection that offers good performance across a large number of sequential and parallel operations. The underlying innovations are: (1) the Relaxed-Radix-Balanced (RRB) tree structure, which allows efficient structural reorganization, and (2) an optimization that exploits spatio-temporal locality on the RRB data structure in order to offset the cost of traversing the tree. In our benchmarks, the RRB-Vector speedup for parallel operations is lower bounded by 7x when executing on 4 CPUs of 8 cores each. The performance for discrete operations, such as appending on either end, or updating and removing elements, is consistently good and compares favorably to the most important immutable sequence collections in the literature and in use today. The memory footprint of RRB-Vector is on par with arrays and an order of magnitude less than competing collections.

Optimizing Energy Efficiency for Minimum Latency Broadcast in Low-Duty-Cycle Sensor Networks

Multihop broadcasting in low-duty-cycle Wireless Sensor Networks (WSNs) is a very challenging problem, since every node has its own working schedule. Existing solutions usually use unicast instead of broadcast to forward packets from a node to its neighbors according to their working schedules, which is, however, not energy efficient. In this article, we propose to exploit the broadcast nature of wireless media to further save energy for low-duty-cycle networks, by adopting a novel broadcasting communication model. The key idea is to let some early wake-up nodes postpone their wake-up slots to overhear broadcasting messages from its neighbors. This model utilizes the spatiotemporal locality of broadcast to reduce the total energy consumption, which can be essentially characterized by the total number of broadcasting message transmissions. Based on such model, we aim at minimizing the total number of broadcasting message transmissions of a broadcast for low-duty-cycle WSNs, subject to the constraint that the broadcasting latency is optimal. We prove that it is NP-hard to find the optimal solution, and design an approximation algorithm that can achieve a polylogarithmic approximation ratio. Extensive simulation results show that our algorithm outperforms the traditional solutions in terms of energy efficiency.

Replacement Method based on Access Spatiotemporal Locality in a Heterogeneous Distributed Cluster-based Caching System for WebGIS

Replacement Method based on Access Spatiotemporal Locality in a Heterogeneous Distributed Cluster-based Caching System for WebGIS

Semantic locality and context-based prefetching using reinforcement learning

Most modern memory prefetchers rely on spatio-temporal locality to predict the memory addresses likely to be accessed by a program in the near future. Emerging workloads, however, make increasing use of irregular data structures, and thus exhibit a lower degree of spatial locality. This makes them less amenable to spatio-temporal prefetchers. In this paper, we introduce the concept of Semantic Locality, which uses inherent program semantics to characterize access relations. We show how, in principle, semantic locality can capture the relationship between data elements in a manner agnostic to the actual data layout, and we argue that semantic locality transcends spatio-temporal concerns. We further introduce the context-based memory prefetcher, which approximates semantic locality using reinforcement learning. The prefetcher identifies access patterns by applying reinforcement learning methods over machine and code attributes, that provide hints on memory access semantics. We test our prefetcher on a variety of benchmarks that employ both regular and irregular patterns. For the SPEC 2006 suite, it delivers speedups as high as 2.8X (20% on average) over a baseline with no prefetching, and outperforms leading spatio-temporal prefetchers. Finally, we show that the context-based prefetcher makes it possible for naive, pointer-based implementations of irregular algorithms to achieve performance comparable to that of spatially optimized code.

Replacement Method based on Access Spatiotemporal Locality in a Heterogeneous Distributed Cluster-based Caching System for WebGIS

Replacement Method based on Access Spatiotemporal Locality in a Heterogeneous Distributed Cluster-based Caching System for WebGIS