Karma: Resource allocation for dynamic demands

M Vuppalapati, G Fikioris, R Agarwal, A Cidon… - … USENIX Symposium on …, 2023 - usenix.org
The classical max-min fairness algorithm for resource allocation provides many desirable
properties, eg, Pareto efficiency, strategy-proofness and fairness. This paper builds upon the …

Dynamicmr: A dynamic slot allocation optimization framework for mapreduce clusters

S Tang, BS Lee, B He - IEEE Transactions on Cloud …, 2014 - ieeexplore.ieee.org
MapReduce is a popular computing paradigm for large-scale data processing in cloud
computing. However, the slot-based MapReduce system (eg, Hadoop MRv1) can suffer from …

Fair resource allocation for data-intensive computing in the cloud

S Tang, BS Lee, B He - IEEE Transactions on Services …, 2016 - ieeexplore.ieee.org
To address the computing challenge ofbig data', a number of data-intensive computing
frameworks (eg, MapReduce, Dryad, Storm and Spark) have emerged and become popular …

Astraea: A fair deep learning scheduler for multi-tenant gpu clusters

Z Ye, P Sun, W Gao, T Zhang, X Wang… - … on Parallel and …, 2021 - ieeexplore.ieee.org
Modern GPU clusters are designed to support distributed Deep Learning jobs from multiple
tenants concurrently. Each tenant may have varied and dynamic resource demands …

Reciprocal resource fairness: Towards cooperative multiple-resource fair sharing in iaas clouds

H Liu, B He - SC'14: Proceedings of the International …, 2014 - ieeexplore.ieee.org
Resource sharing in virtualized environments have been demonstrated significant benefits
to improve application performance and resource/energy efficiency. However, resource …

Dynamic proportional sharing: A game-theoretic approach

R Freeman, SM Zahedi, V Conitzer… - Proceedings of the ACM on …, 2018 - dl.acm.org
Sharing computational resources amortizes cost and improves utilization and efficiency.
When agents pool their resources together, each becomes entitled to a portion of the shared …

Dynamic job ordering and slot configurations for MapReduce workloads

S Tang, BS Lee, B He - IEEE Transactions on Services …, 2015 - ieeexplore.ieee.org
MapReduce is a popular parallel computing paradigm for large-scale data processing in
clusters and data centers. A MapReduce workload generally contains a set of jobs, each of …

Tempo: robust and self-tuning resource management in multi-tenant parallel databases

Z Tan, S Babu - arxiv preprint arxiv:1512.00757, 2015 - arxiv.org
Multi-tenant database systems have a component called the Resource Manager, or RM that
is responsible for allocating resources to tenants. RMs today do not provide direct support …

Elastic multi-resource fairness: balancing fairness and efficiency in coupled CPU-GPU architectures

S Tang, BS He, S Zhang, Z Niu - SC'16: Proceedings of the …, 2016 - ieeexplore.ieee.org
Fairness and efficiency are two important concerns for users in a shared computer system,
and there tends to be a tradeoff between them. Heterogeneous computing poses new …

A cost model for IaaS clouds based on virtual machine energy consumption

M Hinz, GP Koslovski, CC Miers, LL Pilla… - Journal of Grid …, 2018 - Springer
Cloud Computing has revolutionized the software, platform and infrastructure provisioning.
Infrastructure-as-a-Service (IaaS) providers offer on-demand and configurable Virtual …