Filters








15,234 Hits in 6.0 sec

Towards Multi-Resource Fair Allocation with Placement Constraints

Wei Wang, Baochun Li, Ben Liang, Jun Li
2016 Proceedings of the 2016 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Science - SIGMETRICS '16  
It remains unclear how multi-resource fair sharing is defined and achieved in the presence of placement constraints.  ...  Multi-resource fair schedulers have been widely implemented in compute clusters to provide service isolation guarantees.  ...  It remains unclear how multi-resource fair allocation should be defined and achieved for jobs with placement constraints.  ... 
doi:10.1145/2896377.2901493 dblp:conf/sigmetrics/WangLLL16 fatcat:n3kboaagbnea5d24pl34atlfte

Towards Multi-Resource Fair Allocation with Placement Constraints

Wei Wang, Baochun Li, Ben Liang, Jun Li
2016 Performance Evaluation Review  
It remains unclear how multi-resource fair sharing is defined and achieved in the presence of placement constraints.  ...  Multi-resource fair schedulers have been widely implemented in compute clusters to provide service isolation guarantees.  ...  It remains unclear how multi-resource fair allocation should be defined and achieved for jobs with placement constraints.  ... 
doi:10.1145/2964791.2901493 fatcat:id7blyg7szh3tiqkd4ktnpn2aq

Per-Server Dominant-Share Fairness (PS-DSF): A Multi-Resource Fair Allocation Mechanism for Heterogeneous Servers [article]

Jalal Khamse-Ashari, Ioannis Lambadaris, George Kesidis, Bhuvan Urgaonkar, Yiqiang Zhao
2017 arXiv   pre-print
We identify important shortcomings in existing multi-resource fair allocation mechanisms - Dominant Resource Fairness (DRF) and its follow up work - when used in such environments.  ...  We develop a new fair allocation mechanism called Per-Server Dominant-Share Fairness (PS-DSF) which we show offers all desirable sharing properties that DRF is able to offer in the case of a single "resource  ...  CONCLUSION In summary, we studied the problem of multi-resource fair allocation for heterogeneous servers while respecting placement constraints.  ... 
arXiv:1611.00404v2 fatcat:4xy35jdgmzfhpd24ruoqj4gxja

An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers [article]

Jalal Khamse-Ashari, Ioannis Lambadaris, George Kesidis, Bhuvan Urgaonkar, Yiqiang Zhao
2017 arXiv   pre-print
In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness (DRF) and its follow-up work.  ...  Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs.  ...  There are limited works in the literature investigating multi-resource fair allocation in the presence of user placement constraints [10] , [11] .  ... 
arXiv:1712.10114v1 fatcat:gu6ws7schzawvh6ohh5o3jjltu

Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads [article]

Deepak Narayanan, Keshav Santhanam, Fiodar Kazhamiaka, Amar Phanishayee, Matei Zaharia
2020 arXiv   pre-print
Existing schedulers for clusters of accelerators, which are used to arbitrate these expensive training resources across many users, have shown how to optimize for various multi-job, multi-user objectives  ...  , like fairness and makespan.  ...  Hierarchical [61] Multi-level policy: FIFO, fairness, etc. since these multi-resource jobs occupy a larger share of the cluster per unit time.  ... 
arXiv:2008.09213v1 fatcat:mkij4aqxa5b3xhdxewng5trkum

Wide-area analytics with multiple resources

Chien-Chun Hung, Ganesh Ananthanarayanan, Leana Golubchik, Minlan Yu, Mingyang Zhang
2018 Proceedings of the Thirteenth EuroSys Conference on - EuroSys '18  
We propose Tetrium, a system for multi-resource allocation in geodistributed clusters, that jointly considers both compute and network resources for task placement and job scheduling.  ...  Tetrium significantly reduces job response time, while incorporating several other performance goals with simple control knobs.  ...  Related Work Multi-resource Scheduling: Previous works [26, 27] achieve fairness by allocating only compute resources (CPU and memory).  ... 
doi:10.1145/3190508.3190528 dblp:conf/eurosys/HungAGYZ18 fatcat:pczthtnkxfbb5fu6qsecm5mrci

Towards distributed, fair, deadline-driven resource allocation for cloudlets

Stratos Dimopoulos, Chandra Krintz, Rich Wolski
2019 Proceedings of the 4th Workshop on Middleware for Edge Clouds & Cloudlets - MECC '19  
In this paper we present our vision for a two-level, distributed resource allocator that preserves fairness and satisfies deadlines of low latency workloads in a multi-cloudlet environment with offloading  ...  We analyze the opportunities and challenges that offloading and the multi-cloud environment impose and we suggest the changes required to a fair-preserving and deadline-driven resource allocator originally  ...  BACKGROUND AND RELATED WORK Justice [5] is an allocator designed for the resource-constrained settings of a cloudlet with the dual goal of preserving fairness and satisfying job deadlines for multi-analytics  ... 
doi:10.1145/3366614.3368102 fatcat:r7oxdgwhjrht5hqwr4iepy52jm

A Resource Allocation Framework for Network Slicing

Mathieu Leconte, Georgios S. Paschos, Panayotis Mertikopoulos, Ulas C. Kozat
2018 IEEE INFOCOM 2018 - IEEE Conference on Computer Communications  
With tractability in mind, we propose a novel optimization framework which allows fine-grained resource allocation for slices both in terms of network bandwidth and cloud processing.  ...  Furthermore, by tuning a slice-specific parameter, system designers can trade off traffic-fairness with computingfairness to provide a mixed fairness strategy.  ...  Thus, to provision resources for network slices, considerable effort has been dedicated towards the problem of VNF placement and routing [2] [3] [4] [5] [6] [7] [8] . This problem and its M.  ... 
doi:10.1109/infocom.2018.8486303 dblp:conf/infocom/LecontePMK18 fatcat:v4jb7re2kbck7iikqfojy7ycqa

Deep Learning Workload Scheduling in GPU Datacenters: Taxonomy, Challenges and Vision [article]

Wei Gao, Qinghao Hu, Zhisheng Ye, Peng Sun, Xiaolin Wang, Yingwei Luo, Tianwei Zhang, Yonggang Wen
2022 arXiv   pre-print
More detailed summary with the surveyed paper and code links can be found at our project website: https://github.com/S-Lab-System-Group/Awesome-DL-Scheduling-Papers  ...  An efficient scheduler design for such GPU datacenter is crucially important to reduce the operational cost and improve resource utilization.  ...  It formulates the resource allocation as an MILP problem with the resource utilization fairness as the optimization objective.  ... 
arXiv:2205.11913v3 fatcat:fnbinueyijb4nc75fpzd6hzjgq

Choosy

Ali Ghodsi, Matei Zaharia, Scott Shenker, Ion Stoica
2013 Proceedings of the 8th ACM European Conference on Computer Systems - EuroSys '13  
Max-Min Fairness is a flexible resource allocation mechanism used in most datacenter schedulers.  ...  We propose Constrained Max-Min Fairness (CMMF), an extension to max-min fairness that supports placement constraints, and show that it is the only policy satisfying an important property that incentivizes  ...  While there has been much work on max-min fairness for datacenter schedulers, little focus has been on max-min fairness with placement constraints.  ... 
doi:10.1145/2465351.2465387 dblp:conf/eurosys/GhodsiZSS13 fatcat:3pw4ciaaezellnp6xca6nvfy34

Multi-resource packing for cluster schedulers

Robert Grandl, Ganesh Ananthanarayanan, Srikanth Kandula, Sriram Rao, Aditya Akella
2014 Proceedings of the 2014 ACM conference on SIGCOMM - SIGCOMM '14  
We observe that fair allocations do not o er the best performance and the above heuristics are compatible with a large class of fairness policies; hence, we show how to simultaneously achieve good performance  ...  We present Tetris, a multi-resource cluster scheduler that packs tasks to machines based on their requirements of all resource types.  ...  Dominant Resource Fairness (DRF) [ ] is a recent multi-resource fairness algorithm that is available with YARN. DRF considers memory and CPU requirements.  ... 
doi:10.1145/2619239.2626334 dblp:conf/sigcomm/GrandlAKRA14 fatcat:qulrhczvufggrhgrcjm65qibem

Multi-resource packing for cluster schedulers

Robert Grandl, Ganesh Ananthanarayanan, Srikanth Kandula, Sriram Rao, Aditya Akella
2014 Computer communication review  
We observe that fair allocations do not o er the best performance and the above heuristics are compatible with a large class of fairness policies; hence, we show how to simultaneously achieve good performance  ...  We present Tetris, a multi-resource cluster scheduler that packs tasks to machines based on their requirements of all resource types.  ...  Dominant Resource Fairness (DRF) [ ] is a recent multi-resource fairness algorithm that is available with YARN. DRF considers memory and CPU requirements.  ... 
doi:10.1145/2740070.2626334 fatcat:og5n7krtojea7eb67zofmxgpv4

VNF and Container Placement: Recent Advances and Future Trends [article]

Wissal Attaoui, Essaid Sabir, Halima Elbiaze, Mohsen Guizani
2022 arXiv   pre-print
Hence, there is a need for placement methods that can scale with the issue's complexity and find appropriate solutions in a reasonable duration.  ...  For each matter, various solutions are proposed through different surveys and research papers in which each one addresses the placement problem in a specific manner by suggesting single objective or multi-objective  ...  They adopt a heuristic algorithm named Dominant Resource Fairness to run this scheduler.  ... 
arXiv:2204.00178v1 fatcat:giwlibsaknbkdnuvkao6nhkdnq

Multi-Criteria Virtual Machine Placement in Cloud Computing Environments: A literature Review [article]

Wissal Attaoui, Essaid Sabir
2018 arXiv   pre-print
The VNF placement in 5G network is also discussed to highlight the convergence toward optimal usage of mobile services by including NFV/Software-Defined-Network technologies.  ...  This article discusses the different problems that may disrupt the placement of VMs and Virtual Network Functions (VNFs), and classifies the existing solutions into five major objective functions based  ...  The raised problem lies on a complex resource allocation algorithm named (MERA) "Multi-dimensional Energy-efficient Resource Allocation".  ... 
arXiv:1802.05113v1 fatcat:faiazufoq5fphmdd4qefdh5aje

VM reassignment in hybrid clouds for large decentralised companies: A multi-objective challenge

Takfarinas Saber, James Thorburn, Liam Murphy, Anthony Ventresque
2018 Future generations computer systems  
Constraints of the problem Every placement of VM on a PM is subject to a set of constraints that we describe below. 1) Capacity constraints: require that VMs do not exceed the resource capacity of the  ...  Example As a motivating example, consider an organisation with various hosting departments, each of them managed by capital allocators with different placement preferences.  ... 
doi:10.1016/j.future.2017.06.015 fatcat:f2e34ncnv5gvlbb5ava3yqrdie
« Previous Showing results 1 — 15 out of 15,234 results