The rise of edge computing has shifted computing resources closer to end-users, benefiting numerous delay-sensitive, computation-intensive applications. To speed up computation, distributedcomputing is a promising te...
详细信息
ISBN:
(纸本)9781728190549
The rise of edge computing has shifted computing resources closer to end-users, benefiting numerous delay-sensitive, computation-intensive applications. To speed up computation, distributedcomputing is a promising technique that allows parallel execution of computation tasks across multiple compute nodes. However, current research predominantly revolves around the master-worker paradigm, limiting resource sharing within one-hop neighborhoods. This limitation can render distributedcomputing ineffective in scenarios with limited nearby resources or constrained/dynamic connectivity. In this paper, we address this limitation by introducing a new distributedcomputing strategy that extends resource sharing beyond one-hop neighborhoods through exploring layered network structures and multi-hop routing. Our approach involves transforming the network graph into a sink tree and solving a joint optimization problem formulated based on the layered tree structure for task allocation and scheduling. Simulation results demonstrate a significant improvement over the traditional distributedcomputing and computation offloading strategies.
With the rapid development of information technology, the ethical use of data and users' privacy has become a big concern. In European Union, the General Data Protection Regulation (GDPR) has been enforced since 2...
详细信息
ISBN:
(纸本)9798350339864
With the rapid development of information technology, the ethical use of data and users' privacy has become a big concern. In European Union, the General Data Protection Regulation (GDPR) has been enforced since 2018, aiming to protect a person's privacy. However, the growth of the data economy might be hindered due to the lack of a trusted, secure and privacy-aware tool. Motivated by this observation, this work presents DataVaults - a secure, distributed and privacy-preserving personal data management platform. The main goal is to mitigate various privacy concerns through allowing an individual to maintain the ownership, handle and share the data based on their willingness. The platform enables a flexible data sharing method with fair compensation schemes. In particular, with the DataVaults platform, an individual can protect the sharing of personal data and fairly define how value can be captured, created, released and cashed out for the benefit of all the stakeholders involved (companies or not).
Modern software systems in every application domain are increasingly built as distributedsystems. Business applications are structured as cooperating microservices, IoT devices communicate with cloud-based services o...
详细信息
ISBN:
(纸本)9798350366266;9798350366259
Modern software systems in every application domain are increasingly built as distributedsystems. Business applications are structured as cooperating microservices, IoT devices communicate with cloud-based services over a network, and Web sites store data in globally dispersed data centers to support fast access in to localities in which their users reside. Behind all these systems lurk distributedcomputing infrastructures that architects and engineers must exploit to satisfy application service level agreements. To be successful, it is essential that architects understand the inherent complexity of distributedsystems. In this half day tutorial, I'll guide the attendees through the fundamental characteristics that distributedsystems exhibit. Each characteristic will be related to the software architecture quality attributes that they directly impact. The topics covered include communications reliability and latencies, message delivery semantics, state management, idempotence, data safety, consistency, time, distributed consensus, cascading failures and failover and recovery. I'll introduce each concept using an example distributed system and multiple 'props' to illustrate concepts. Once I've explained a concept using the example, I'll move on to show how the concept manifests itself in a software system and its effects on quality attributes requirements and inherent trade-offs. The tutorial will be suitable for graduate students, engineers and architects who have no or minimal exposure to distributedsystems concepts. The presentation format will be suitable for a mix of both in person and remote participants. It will combine interactive sessions with short technical explanations and examples to illustrate each distributedsystems concept.
The increasing popularity of applications like the Metaverse has led to the exploration of new, more effective ways of communication. Semantic communication, which focuses on the meaning behind transmitted information...
详细信息
ISBN:
(纸本)9798350339864
The increasing popularity of applications like the Metaverse has led to the exploration of new, more effective ways of communication. Semantic communication, which focuses on the meaning behind transmitted information, represents a departure from traditional communication paradigms. As mobile devices become increasingly prevalent, it is important to explore the potential of edge computing to aid the semantic encoding/decoding process, which requires significant computing power and storage capabilities. However, establishing knowledge bases (KBs) for domain-oriented communication can be time-consuming. To address this challenge, this paper proposes a semantic caching model in edge computing system that caches domain-specialized general models and user-specific individual models. This approach has the potential to reduce the time and resources required to establish individual KBs while accurately capturing the semantics behind users' messages, ultimately leading to more efficient and accessible semantic communication.
distributed tracing serves as a fundamental element in the monitoring of cloud-based and datacenter systems. It provides visibility into the full lifecycle of a request or operation across multiple services, which is ...
详细信息
ISBN:
(纸本)9798350368543;9798350368536
distributed tracing serves as a fundamental element in the monitoring of cloud-based and datacenter systems. It provides visibility into the full lifecycle of a request or operation across multiple services, which is essential for understanding system dependencies and performance bottlenecks. To mitigate computational and storage overheads, most tracing frameworks adopt a uniform sampling strategy, which inevitably captures overlapping and redundant information. More advanced methods employ learning-based approaches to bias the sampling toward more informative traces. However, existing methods fall short of considering the high-dimensional and dynamic nature of trace data, which is essential for the production deployment of trace sampling. To address these practical challenges, in this paper we present TRACEMESH, a scalable and streaming sampler for distributed traces. TRACEMESH employs Locality-Sensitivity Hashing (LSH) to improve sampling efficiency by projecting traces into a low-dimensional space while preserving their similarity. In this process, TRACEMESH accommodates previously unseen trace features in a unified and streamlined way. Subsequently, TRACEMESH samples traces through evolving clustering, which dynamically adjusts the sampling decision to avoid over-sampling of recurring traces. The proposed method is evaluated with trace data collected from both open-source microservice benchmarks and production service systems. Experimental results demonstrate that TRACEMESH outperforms state-of-the-art methods by a significant margin in both sampling accuracy and efficiency.
In this paper, we put forward a distributed, Transformer-assisted deep reinforcement learning scheme for latency-sensitive, mobility-aware and queue-aware task offloading in mobile edge computingsystems. The proposed...
详细信息
ISBN:
(纸本)9781728190549
In this paper, we put forward a distributed, Transformer-assisted deep reinforcement learning scheme for latency-sensitive, mobility-aware and queue-aware task offloading in mobile edge computingsystems. The proposed scheme adopts an attention-based transformer and deep reinforcement learning for minimizing the average cost and the task processing latency. Since the proposed scheme is distributed, there is no single point of failure in the system. Simulation results show that the proposed scheme could significantly outperform a number of baseline schemes in the literature.
Decentralization initiatives like Solid and *** enable data owners to control who has access to their data and to stimulate innovation by creating both application and data markets. Once data owners share their data w...
详细信息
ISBN:
(纸本)9798350328127
Decentralization initiatives like Solid and *** enable data owners to control who has access to their data and to stimulate innovation by creating both application and data markets. Once data owners share their data with others, though, it is no longer possible for them to control how their data are used. To address this issue, we propose a usage control architecture to monitor compliance with usage control policies. To this end, our solution relies on blockchain and trusted execution environments. We demonstrate the potential of the architecture by describing the various workflows needed to realize a motivating use case scenario for data markets. Additionally, we discuss the merits of the approach from privacy, security, integrateability, and affordability perspectives.
This research aims to optimize the processing of medical data by developing and implementing an efficient distributedcomputing platform by leveraging machine learning and edge computing. By doing so, we seek to strik...
详细信息
ISBN:
(纸本)9798350304367;9798350304374
This research aims to optimize the processing of medical data by developing and implementing an efficient distributedcomputing platform by leveraging machine learning and edge computing. By doing so, we seek to strike a balance between the computational requirements of machine learning models and the need to process medical data locally in many mobile medical imaging scenarios, thus addressing the challenges posed by volume, privacy, and security.
In this paper, we propose and analyze HierAdMo, a three-tier adaptive momentum accelerated client-edge-cloud Federated Learning (FL) algorithm. HierAdMo combines the momentum acceleration on both worker and edge level...
详细信息
ISBN:
(纸本)9798350339864
In this paper, we propose and analyze HierAdMo, a three-tier adaptive momentum accelerated client-edge-cloud Federated Learning (FL) algorithm. HierAdMo combines the momentum acceleration on both worker and edge levels. However, simply combining these two levels of momenta may lead to disagreement between them, negatively influencing convergence performance. To this end, we embed an online adaptive method that scales down the momentum when disagreement occurs. We provide mathematical proof for the convergence of HierAdMo for non-i.i.d. data and the tighter convergence upper bound compared with a version of HierAdMo without adaptation (HierAdMo-R). Finally, extensive experiments based on real-world datasets are conducted, verifying that HierAdMo outperforms existing mainstream benchmarks and achieves the optimal or near-optimal convergence performance compared with HierAdMo-R under a wide range of settings.
As datacenters become the new computing platform, integrating server-centric distributedsystems into modern network hardware is gaining interest under the diminishing Moore's law. Researchers want to build reusab...
详细信息
ISBN:
(纸本)9798350339864
As datacenters become the new computing platform, integrating server-centric distributedsystems into modern network hardware is gaining interest under the diminishing Moore's law. Researchers want to build reusable primitives that can take advantage of modern network hardware and offload common system components of a broad range of applications. In this paper, we present Mostly Reliable Totally Ordered Multicast, a reusable network primitive that can embed reliable group communication into the network. MRTOM is a network-centric approach that handles message replication, ordering, and reliable delivery using a network fast path, freeing server CPUs for application logic. MRTOM can be implemented in the programmable switches and edge interfaces (e.g., SmartNICs), significantly reducing network traffic compared to the existing approaches and improving job finish time amid packet loss. With MRTOM, we were able to accelerate multiple high-performance applications whose fast path can be totally offloaded into the network. For example, a Paxos application, MRTOM-Paxos, achieves > 1,100,000 transactions/secs and 23 mu s minimum latency. A replicated key-value store, MRTOM-KV, also shows significant latency reduction with eBPF/XDP in the Linux Kernel, which is further improved by offloading into SmartNICs.
暂无评论