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.
Filters
NURA
2022
Abstract Proceedings of the 2022 ACM SIGMETRICS/IFIP PERFORMANCE Joint International Conference on Measurement and Modeling of Computer Systems
Preserved Fulltext
Some pieces of prior work (e.g. spatial multitasking) have limited opportunity to improve resource utilization, while others, e.g. simultaneous multi-kernel, provide fine-grained resource sharing at the ...
In terms of fairness, NURA has almost similar results to spatial multitasking, while it outperforms simultaneous multi-kernel by 76%, on average. ...
There are two known approaches of multitasking in GPUs: (1) spatial multi-tasking [3] , and (2) simultaneous multi-kernel (SMK) execution [2] . ...
doi:10.1145/3489048.3522656
fatcat:xcmtppre3rer3etvsjnrvzjtei
Dynamic Resource Management for Efficient Utilization of Multitasking GPUs
2017
SIGARCH Computer Architecture News
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 2017; you can also visit the original URL.
The file type is application/pdf.
Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions GPU resource at a streaming multiprocessor (SM) granularity, or simultaneous multikernel (SMK), which ...
In this paper, we propose GPU Maestro that performs dynamic resource management for efficient utilization of multitasking GPUs. ...
Acknowledgments We would like to thank the anonymous reviewers as well as the fellow members of CCCP research group for their valuable comments and feedbacks. ...
doi:10.1145/3093337.3037707
fatcat:7vikinfjtbbmnperrnxretampm
Dynamic Resource Management for Efficient Utilization of Multitasking GPUs
2017
ACM SIGOPS Operating Systems Review
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 2017; you can also visit the original URL.
The file type is application/pdf.
Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions GPU resource at a streaming multiprocessor (SM) granularity, or simultaneous multikernel (SMK), which ...
In this paper, we propose GPU Maestro that performs dynamic resource management for efficient utilization of multitasking GPUs. ...
Acknowledgments We would like to thank the anonymous reviewers as well as the fellow members of CCCP research group for their valuable comments and feedbacks. ...
doi:10.1145/3093315.3037707
fatcat:5xjasiupnrcctp6oarqwrtbukm
Dynamic Resource Management for Efficient Utilization of Multitasking GPUs
2017
Proceedings of the Twenty-Second International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS '17
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 2017; you can also visit the original URL.
The file type is application/pdf.
Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions GPU resource at a streaming multiprocessor (SM) granularity, or simultaneous multikernel (SMK), which ...
In this paper, we propose GPU Maestro that performs dynamic resource management for efficient utilization of multitasking GPUs. ...
Acknowledgments We would like to thank the anonymous reviewers as well as the fellow members of CCCP research group for their valuable comments and feedbacks. ...
doi:10.1145/3037697.3037707
dblp:conf/asplos/ParkPM17
fatcat:flmnbk4x3je4loearqmn2uzwkm
Dynamic Resource Management for Efficient Utilization of Multitasking GPUs
2017
SIGPLAN notices
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 2017; you can also visit the original URL.
The file type is application/pdf.
Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions GPU resource at a streaming multiprocessor (SM) granularity, or simultaneous multikernel (SMK), which ...
In this paper, we propose GPU Maestro that performs dynamic resource management for efficient utilization of multitasking GPUs. ...
Acknowledgments We would like to thank the anonymous reviewers as well as the fellow members of CCCP research group for their valuable comments and feedbacks. ...
doi:10.1145/3093336.3037707
fatcat:xkqmepvakbh23gfyrnkjot6uge
Characterizing Fine-Grained Resource Utilization for Multitasking GPGPU in Cloud Systems
2021
IEEE Access
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.
In this article, we show that efficient resource sharing in GPGPU is possible without run-time profiling if resource usage characteristics of workloads are analyzed down to a fine-grained unit level. ...
To determine the co-location of workloads, previous studies have shown that run-time performance profiling and dynamic relocation of workloads is necessary due to interference between workloads. ...
., device memory and shared memory. The coarse-grained utilization columns in the tables are calculated by the maximum value of the fine-grained utilizations within the same resource classifications. ...
doi:10.1109/access.2021.3132492
fatcat:y5yucr4qbbeevcmwkfpydzrqbe
A Survey of Multi-Tenant Deep Learning Inference on GPU
[article]
2022
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 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
With such strong computing scaling of GPUs, multi-tenant deep learning inference by co-locating multiple DL models onto the same GPU becomes widely deployed to improve resource utilization, enhance serving ...
This survey aims to summarize and categorize the emerging challenges and optimization opportunities for multi-tenant DL inference on GPU. ...
However, as we introduced before, achieving fine-grained resource partitioning is non-achievable until recently GPU vendors release a series of resource sharing and partitioning support like multistreams ...
arXiv:2203.09040v3
fatcat:utvpoyvvajfhfghgpf45nxnbne
On the Performance and Isolation of Asymmetric Microkernel Design for Lightweight Manycores
2019
2019 IX Brazilian Symposium on Computing Systems Engineering (SBESC)
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.
Multikernel operating systems (OSs) were introduced to match the architectural characteristics of lightweight manycores. ...
While several multikernel OS designs are possible, in this work we argue on one that is structured in asymmetric microkernel instances. ...
Notwithstanding, due to a lack of context information, this is not enough for the kernel to either run a fine-grain data prefetch algorithm or a coherency protocol. ...
doi:10.1109/sbesc49506.2019.9046080
dblp:conf/sbesc/PennaSLCBFM19
fatcat:oe576cky2jfp3h4hynrszytmbq
AMOEBA: A Coarse Grained Reconfigurable Architecture for Dynamic GPU Scaling
[article]
2019
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.
A GPU consists of several StreamingMulti-processors (SMs) that collectively determine how shared resources are partitioned and accessed. ...
However, neither scaling up nor scaling out can meet the scalability requirement of all applications running on a given GPU system, which inevitably results in performance degradation and resource under-utilization ...
Dhar et al. proposed fine grained and coarse grained reconfigurations of SMs in GPUs in order to reduce the underutilization of resources and power consumption [15] . ...
arXiv:1911.03364v1
fatcat:tcmbakgikjhtdl2khxzei3zf74
Concurrent query processing in a GPU-based database system
2019
PLoS ONE
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.
The unrivaled computing capabilities of modern GPUs meet the demand of processing massive amounts of data seen in many application domains. ...
Comparing to earlier studies of enabling concurrent tasks support on GPU such as MultiQx-GPU, we use a different approach that is to control the launching parameters of multiple GPU kernels as provided ...
fine-grained context switches. ...
doi:10.1371/journal.pone.0214720
pmid:30990851
pmcid:PMC6467383
fatcat:4u2hmql235c4fkxajva5mcx6m4
AtomNAS: Fine-Grained End-to-End Neural Architecture Search
[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.
Other Versions
We propose a fine-grained search space comprised of atomic blocks, a minimal search unit that is much smaller than the ones used in recent NAS algorithms. ...
Instead of a search-and-retrain two-stage paradigm, our method simultaneously searches and trains the target architecture. ...
This perspective enables a much larger and more fine-grained search space. ...
arXiv:1912.09640v2
fatcat:qyjyg33dkvc53pglyppjrnee44
Classification-Driven Search for Effective SM Partitioning in Multitasking GPUs
2018
Proceedings of the 2018 International Conference on Supercomputing - ICS '18
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.
Spatial multitasking in which independent applications co-execute on different sets of SMs is a promising solution to share GPU resources. ...
Graphics processing units (GPUs) feature an increasing number of streaming multiprocessors (SMs) with each successive generation. ...
We thank CalcUA for letting us use the NVIDIA P100 GPU. This work is supported by the European Research Council (ERC) Ad- ...
doi:10.1145/3205289.3205311
dblp:conf/ics/ZhaoWE18
fatcat:b4hzqrztwrabbjcydmgka3kqcq
A State-of-the-Art Survey on Real-Time Issues in Embedded Systems Virtualization
2012
Journal of Software Engineering and Applications
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.
A State-of-the-Art Survey on Real-Time Issues in Embedded Systems Virtualization 278 that are not specific to any virtualization approach, but we believe they are of sufficient importance to dedicate separate ...
fine-grained control, e.g., an important task in the GPOS can be assigned a higher priority than a less important task in the RTOS. ...
[75] presented a redesign of SPUMONE as a multikernel architecture. (Multikernel means that a separate copy of the kernel or VMM runs on each core on a multicore processor.) ...
doi:10.4236/jsea.2012.54033
fatcat:iiqszwe3brhjxl4ycyf6ynbp2u
Algorithms for Preemptive Co-scheduling of Kernels on GPUs
2020
2020 IEEE 27th International Conference on High Performance Computing, Data, and Analytics (HiPC)
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.
Currently, the decision on the simultaneous execution of kernels is performed by the hardware, which can lead to unreasonable use of resources. ...
In this work, we tackle the problem of co-scheduling for GPUs in high competition scenarios. ...
[23] proposed Simultaneous Multikernel (SMK) that allows fine-grain sharing within each SM, kernels with complimentary resource usage are co-scheduled in the same SM to achieve resource fairness and ...
doi:10.1109/hipc50609.2020.00033
fatcat:piwx3puyyfdhpbdtxkk7hpzllq
Elastic Multi-resource Fairness: Balancing Fairness and Efficiency in Coupled CPU-GPU Architectures
2016
SC16: International Conference for High Performance Computing, Networking, Storage and Analysis
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 2016; you can also visit the original URL.
The file type is application/pdf.
We show that EMRF satisfies fairness properties of sharing incentive, envy-freeness and pareto efficiency. ...
Heterogeneous computing poses new challenging issues on the fair allocation of computational resources among users due to the availability of different kinds of computing devices (e.g., CPU and GPU). ...
Shuhao Zhang's work is partially funded by the Economic Development Board and the National Research Foundation of Singapore. ...
doi:10.1109/sc.2016.74
dblp:conf/sc/TangHZN16
fatcat:l4hcko577favfa35c7jyxudvka
« Previous
Showing results 1 — 15 out of 51 results