A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
Filters
Efficient Parallel Reinforcement Learning Framework using the Reactor Model
[article]
2024
arXiv
pre-print
Preserved Fulltext
Other Versions
This allows the scheduler to eliminate work needed for synchronization, such as acquiring and releasing locks for each actor or sending and processing coordination-related messages. ...
Parallel Reinforcement Learning (RL) frameworks are essential for mapping RL workloads to multiple computational resources, allowing for faster generation of samples, estimation of values, and policy improvement ...
This makes the model highly adaptable and scalable, suitable for applications ranging from simple concurrent programs to complex distributed systems. ...
arXiv:2312.04704v2
fatcat:nth42qnuefdczdy2vqg4opzz4m
A.M.B.R.O.S.I.A
2020
Proceedings of the VLDB Endowment
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We also introduce novel programming language constructs for resiliently handling non-determinism. ...
We coin the term "virtual resiliency", analogous to virtual memory, for the platform feature which allows failure oblivious code to run in a failure resilient manner. ...
AMBROSIA DESIGN 3.1 Ambrosia's Approach and Architecture Ambrosia's approach for implementing virtual resiliency is an evolution of past approaches for creating deterministic robust distributed systems ...
doi:10.14778/3377369.3377370
fatcat:fwsthl5wozef5hxncc73pnxqz4
Embedded Policing and Policy Enforcement based Security in the era of Digital-Physical Convergence for Next-Generation Vehicular Electronics
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Much of this has been enabled by technological advancements in embedded system architectures, which provided opportunities for vehicle manufacturers to implement intelligent vehicle services and consolidate ...
This paper presents a novel embedded policing and policy enforcement platform architecture and the accompanied security modelling approach for next-generation in-vehicle domain controllers. ...
secure system lock-down. • Initiation system reset. ...
arXiv:2004.10672v1
fatcat:hda7w26qnzdspp5bmlyd6ku5ze
2020 Index IEEE Transactions on Circuits and Systems I: Regular Papers Vol. 67
2020
IEEE Transactions on Circuits and Systems Part 1: Regular Papers
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
., +, TCSI March 2020 852-865 Multiprocessing systems Kernel-Based Resource Allocation for Improving GPU Throughput While Minimizing the Activity Divergence of SMs. ...
Zhao, J., +, TCSI Oct. 2020 3498-3511 Adaptive systems Adaptive Fuzzy Output Feedback Event-Triggered Control for a Class of Switched Nonlinear Systems With Sensor Failures. ...
., +, 48-59 A PVT-Resilient, Highly-Linear Fifth-Order Ring-Oscillator-Based Filter. ...
doi:10.1109/tcsi.2021.3055003
fatcat:kbmst5td2bbvtl7vpbj3knnkri
Adaptive Optimizations for Stream-based Workflows
2020
2020 IEEE/ACM Workflows in Support of Large-Scale Science (WORKS)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
a bridge to existing parallel programming frameworks like MPI or Python multiprocessing. ...
Therefore, our aim is to improve the performance of dispel4py workflows by testing different workload strategies that automatically adapt to workflows. ...
But, for the Dynamic Scheduling technique, we will need to modify it first to adapt it to distributed-memory architectures. ...
doi:10.1109/works51914.2020.00010
fatcat:jaivbc4qnbbkznernlttfaz2xi
The CLAM approach to multithreaded communication on shared-memory multiprocessors: design and experiments
1998
IEEE Transactions on Parallel and Distributed Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
motivation and design decisions behind the CLAM system, along with two simple but effective scheduling algorithms for the simultaneous support of multiple, threads-based user-space protocols. ...
The Connectionless, Lightweight, and Multiway (CLAM) communications environment provides efficient and scalable multi-protocol support for distributed applications using mullimodal data. 1bis paper presents ...
In viewing the SANR algorithm's time-slice adaptability in the multiprocess case, as shown in Figure 20 (b), we observe that adaptability is hampered by the increased randomness in the times between receive ...
doi:10.1109/71.655241
fatcat:t5rl7arunrbxrcr4xnv5vu6764
A Linux-based tool for hardware bring up, Linux development, and manufacturing
2005
IBM Systems Journal
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2007; you can also visit the original URL.
The file type is application/pdf.
BML, which has fast boot times and can be modified quickly, is used in fault detection during chip manufacturing, POWER5 chip verification, system-board verification, and benchmarking for performance. ...
operating system. ...
The POWER5 hypervisor provides additional virtualization capabilities compared to those for POWER4 systems, and in particular a high degree of resiliency to runtime errors. ...
doi:10.1147/sj.442.0319
fatcat:yqxlktdbqzasvf3w6klsksqvty
Adaptive Optimizations for Stream-based Workflows
2020
Zenodo
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Therefore, ouraim is to improve the performance of dispel4py workflows bytesting different workload strategies that automatically adapt toworkflows. ...
For achieving this objective, we have implementedthree new techniques, calledNaive Assignment,StagingandDynamic Scheduling. ...
But, for the Dynamic Scheduling technique, we will need to modify it first to adapt it to distributed-memory architectures. ...
doi:10.5281/zenodo.4396295
fatcat:7auo2a4yxnht3bki4uwzm7uphu
Towards Flexible, Fault Tolerant Hardware Service Wrappers for the Digital Manufacturing on a Shoestring Project
2020
IFAC-PapersOnLine
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
We explain the existing Shoestring demonstrator architecture, and discuss how a 'crash-only' microservices architecture would improve fault tolerance and adaptability of the system. ...
These form part of a common architecture for data acquisition in the Digital Manufacturing on a Shoestring project. ...
This should enable SMEs to rapidly develop IT solutions for condition based maintenance and computer maintenance information systems, whilst not tying them into a closed, inflexible or inappropriately ...
doi:10.1016/j.ifacol.2020.11.065
fatcat:3lolxokwgvekbbucj5xwwv4gce
KV-Cache: A Scalable High-Performance Web-Object Cache for Manycore
2013
2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2014; you can also visit the original URL.
The file type is application/pdf.
KV-Cache's highly optimized architecture benefits from true "absolute" zero-copy by performing direct memory access (DMA) for each transmit and receive path, and eliminating any software memory copying ...
Our solution, based on TU Dresden's Fiasco.OC microkernel operating system, offers scalability and performance that significantly exceeds that of its Linuxbased counterpart. ...
The Genode architecture ensures tight admission control of memory, CPU and other resources in the system as the basis for limiting failure and attack propagation. ...
doi:10.1109/ucc.2013.34
dblp:conf/ucc/WaddingtonCKS13
fatcat:inpww6lr4nhkbcgzrwmvmmospm
2021 Index IEEE Transactions on Circuits and Systems I: Regular Papers Vol. 68
2021
IEEE Transactions on Circuits and Systems Part 1: Regular Papers
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
The Author Index contains the primary entry for each item, listed under the first author's name. ...
., +, TCSI Aug. 2021 3520-3533 Adaptive systems Adaptive Fuzzy Fast Finite-Time Dynamic Surface Tracking Control for Nonlinear Systems. ...
., +, TCSI Oct. 2021 4013-4024 Design Flow for Hybrid CMOS/Memristor Systems-Part I: Modeling and Verification Steps. ...
doi:10.1109/tcsi.2021.3134605
fatcat:txqhqj7nvnh6pp5dqloynq5jku
Satisfying Secure Load Balancing Expectations in the Cloud
2018
Americas Conference on Information Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
The end user expectation is for seamless and immediate services to support all of their information needs. ...
This paper provides a design solution to the tension between the end user and cloud service supplier expectations for secure load balancing by designing a trust evaluation framework in which the competing ...
The Lock-free multiprocessing solution avoids the use of shared memory in contrast to other multiprocessing load balancing solutions which exploit shared memory and lock it to maintain a user session. ...
dblp:conf/amcis/CusackG18
fatcat:juywxpt6n5axdfc2dtk6p2m2yy
Building a High-performance Message-passing System for MEMORY CHANNEL Clusters
1996
Digital technical journal of Digital Equipment Corporation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2014; you can also visit the original URL.
The file type is application/pdf.
The method most frequently used was to build parallel systems out of clusters of commodity workstations or servers that could be used as a virtual supercomputer. 1 The motivation for this work was the ...
is directly reflected in the physical memory of another system. ...
Acknowledgments The authors would like to acknowledge the following people for their contributions to this project: Gavan Duffy, whose testing made the TruCluster MEMORY CHANNEL Software a much more robust ...
dblp:journals/dtj/LawtonBDRR96
fatcat:n6uusimcbnct3o4dl3dnl2s4mu
Cybersecurity of Industrial Cyber-Physical Systems: A Review
2022
ACM Computing Surveys
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
In particular, multi-dimensional adaptive attack taxonomy is presented and utilized for evaluating real-life ICPS cyber incidents. ...
Hence, finding a solution for the problems mentioned in these reports is relatively hard. We bridge this gap by defining and reviewing ICPSs from a cybersecurity perspective. ...
ACKNOWLEDGMENTS This work is partially supported by EPSRC PETRAS (EP/S035362/1) and GCHQ National Resilience Fellowship. ...
doi:10.1145/3510410
fatcat:fa73nli6zzfxjkedcosbj3j4li
A Teradata content-based multimedia object manager for massively parallel architectures
1996
Proceedings of the 1996 ACM SIGMOD international conference on Management of data - SIGMOD '96
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2012; you can also visit the original URL.
The file type is application/pdf.
The Teradata Multimedia Object Manager is a generalpurpose content analysis multimedia server designed for symmetric multiprocessing and massively parallel processing environments. ...
In addition, Marilyn Folster of AT&T Network Systems and George Thiruvathukal of R.R. Donnelly & Sons' MetroMail division provided valuable comments on the paper. ...
Symmetric multiprocessing (SMP) and massively parallel processing (MPP) hardware architectures provide such capabilities. ...
doi:10.1145/233269.233321
dblp:conf/sigmod/OConnellISWABCCLPWW96
fatcat:kyrcsa6lzvhwtbnmehw5j2b4ia
« Previous
Showing results 1 — 15 out of 142 results