121 Hits in 2.8 sec

numpywren: serverless linear algebra [article]

Vaishaal Shankar, Karl Krauth, Qifan Pu, Eric Jonas, Shivaram Venkataraman, Ion Stoica, Benjamin Recht, Jonathan Ragan-Kelley
2018 arXiv   pre-print
We present numpywren, a system for linear algebra built on a serverless architecture.  ...  We also introduce LAmbdaPACK, a domain-specific language designed to implement highly parallel linear algebra algorithms in a serverless setting.  ...  Linear Algebra Algorithms Given the motivating applications, in this work, we broadly focus on the case of large-scale dense linear algebra.  ... 
arXiv:1810.09679v1 fatcat:lxv65s42rzd6vomkg2pff2akhm

Cloud Programming Simplified: A Berkeley View on Serverless Computing [article]

Eric Jonas, Johann Schleier-Smith, Vikram Sreekanti, Chia-Che Tsai, Anurag Khandelwal, Qifan Pu, Vaishaal Shankar, Joao Carreira, Karl Krauth, Neeraja Yadwadkar, Joseph E. Gonzalez, Raluca Ada Popa, Ion Stoica, David A. Patterson
2019 arXiv   pre-print
applications that stretch the current limits of serverless, and then lists obstacles and research opportunities required for serverless computing to fulfill its full potential.  ...  Serverless cloud computing handles virtually all the system administration operations needed to make it easier for programmers to use the cloud.  ...  While these results are promising, using serverless computing for linear algebra has several limitations.  ... 
arXiv:1902.03383v1 fatcat:5lly4vckcnhatp4g3ccdt3us6e

Towards Demystifying Intra-Function Parallelism in Serverless Computing [article]

Michael Kiener, Mohak Chadha, Michael Gerndt
2021 arXiv   pre-print
Serverless computing offers a pay-per-use model with high elasticity and automatic scaling for a wide range of applications.  ...  By parallelizing serverless workloads, we observed cost savings up to 81% for AWS Lambda, 49% for GCF, and 69.8% for GCR.  ...  15] , scientific computing [9, 14, 16, 17] , and linear algebra [31] .  ... 
arXiv:2110.12090v3 fatcat:qeb4k2onnfcvpkur3cx3ii3x34

The Collection Virtual Machine: An Abstraction for Multi-Frontend Multi-Backend Data Analysis [article]

Ingo Müller
2020 arXiv   pre-print
We have used CVM successfully to build specialized backends for platforms as diverse as multi-core CPUs, RDMA clusters, and serverless computing infrastructure in the cloud and expect similar results for  ...  To combine different types of analysis in one system, systems like LaraDB [10] and AIDA [7] integrate relational algebra with linear algebra in a single abstraction.  ...  As we explain in more detail below, analytics in relational algebra, graph analysis, linear algebra, and machine learning work on relations of records, set of vertices and edges, vectors and matrices of  ... 
arXiv:2004.01908v2 fatcat:7k2svf6sobghrexr7oaaaezgeu

FuncPipe: A Pipelined Serverless Framework for Fast and Cost-efficient Training of Deep Learning Models [article]

Yunzhuo Liu, Bo Jiang, Tian Guo, Zimeng Huang, Wenhao Ma, Xinbing Wang, Chenghu Zhou
2022 arXiv   pre-print
The appalling resource mismatch between serverless functions and DL models has stalled the progress on serverless-based training.  ...  With the emergence of serverless computing and its benefits of true pay-as-you-go pricing and scalability, systems researchers have recently started to provide support for serverless-based training.  ...  ., CPU servers with serverless functions. It exploits the inherent features of GNN to separate the computation tasks, and uses serverless functions only for the lightweight linear algebra operations.  ... 
arXiv:2204.13561v1 fatcat:gjrrthzmbndsrdkdmqheggmjua

Faasm: Lightweight Isolation for Efficient Stateful Serverless Computing [article]

Simon Shillaker, Peter Pietzuch
2020 arXiv   pre-print
We introduce Faaslets, a new isolation abstraction for high-performance serverless computing.  ...  Existing serverless platforms isolate functions in ephemeral, stateless containers, preventing them from directly sharing memory.  ...  Prior work has shown how serverless can support map/reduce-style jobs [42, 69] , machine learning training [17, 18] and inference [40] , and linear algebra computation [73, 88] .  ... 
arXiv:2002.09344v2 fatcat:5a62qbbadndwhgaotdorly4dim

OverSketched Newton: Fast Convex Optimization for Serverless Systems [article]

Vipul Gupta, Swanand Kadhe, Thomas Courtade, Michael W. Mahoney, Kannan Ramchandran
2020 arXiv   pre-print
OverSketched Newton leverages matrix sketching ideas from Randomized Numerical Linear Algebra to compute the Hessian approximately.  ...  optimization algorithm to solve large-scale convex optimization problems in serverless systems.  ...  In [4] , the authors observed that many large-scale linear algebra operations on serverless systems take at least 30% more time compared to MPI-based computation on serverful systems.  ... 
arXiv:1903.08857v3 fatcat:luuf7mcdm5dk3l4bhpmf4khmq4

A Serverless-Based, On-the-Fly Computing Framework for Remote Sensing Image Collection

Jin Wu, Mingbo Wu, Haiyan Li, Lijuan Li, Leilei Li
2022 Remote Sensing  
As an advancement of cloud computing, serverless computing makes it possible to realize the on-the-fly computational model.  ...  The proof-of-concept experiments have suggested that the on-the-fly computing model for remote sensing data analysis can be efficiently implemented as a serverless software.  ...  The band-level operators, band math, or map algebra can be regarded as some window operators on linear algebra.  ... 
doi:10.3390/rs14071728 fatcat:f43jkf5c3ng47bqvvzjqrxyq6m

Pay-Per-Request Deployment of Neural Network Models Using Serverless Architectures

Zhucheng Tu, Mengping Li, Jimmy Lin
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations  
We describe a number of techniques that allow efficient use of serverless resources, and evaluations confirm that our design is both scalable and inexpensive.  ...  We demonstrate the serverless deployment of neural networks for model inferencing in NLP applications using Amazon's Lambda service for feedforward evaluation and DynamoDB for storing word embeddings.  ...  Our deployment package is bundled with OpenBLAS to take advantage of optimized linear algebra routines. In both cases, we see that latency increases slightly as throughput ramps up.  ... 
doi:10.18653/v1/n18-5002 dblp:conf/naacl/TuLL18 fatcat:ipm2zinp2nf6bh3n4lykzz7ezy

Serverless Computing: One Step Forward, Two Steps Back [article]

Joseph M. Hellerstein, Jose Faleiro, Joseph E. Gonzalez, Johann Schleier-Smith, Vikram Sreekanti, Alexey Tumanov, Chenggang Wu
2018 arXiv   pre-print
Serverless computing offers the potential to program the cloud in an autoscaling, pay-as-you go manner.  ...  Put together, these gaps make current serverless offerings a bad fit for cloud innovation and particularly bad for data systems innovation.  ...  algebra workloads.  ... 
arXiv:1812.03651v1 fatcat:gnbjeueh4vgtdmmypp3frcmfcq

In Search of a Fast and Efficient Serverless DAG Engine [article]

Benjamin Carver, Jingyuan Zhang, Ao Wang, Yue Cheng
2019 arXiv   pre-print
In this paper, we thoroughly investigate the problem space of DAG scheduling in serverless computing. We identify and evaluate a set of techniques to make DAG schedulers serverless-aware.  ...  The auto-scaling property of serverless computing platforms accommodates short tasks and bursty workloads, while the pay-per-use billing model of serverless computing providers keeps the cost of short  ...  General Matrix Multiplication (GEMM): GEMM, as the core of many linear algebra algorithms, performs matrix multiplication.  ... 
arXiv:1910.05896v2 fatcat:itnhnurocvdozghbvxeqrnwjpi

Computação Serverless: Conceitos, Aplicações e Desafios [chapter]

André Vieira, Gustavo Pantuza, Jean Freire, Lucas Duarte, Racyus Pacífico, Marcos Vieira, Luiz Vieira, José Nacif
2020 Minicursos do XXXVIII Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos  
Serverless computing is a novel application deployment model that provides scaling, finegrained billing, and platform decoupling for developers.  ...  In this short course, we present the Serverless Computing state of the art: concepts, limitations, use cases, research opportunities, caveats, and a view on some current platforms.  ...  [Shankar et al. 2018] criaram Numpywren, um sistema distribuído Serverless que permite executar algoritmos de álgebra linear em larga escala utilizando funções Serverless.  ... 
doi:10.5753/sbc.5033.7.5 fatcat:i44rrkqqmvezpd5puxphb4p7mi

Serverless End Game: Disaggregation enabling Transparency [article]

Pedro García-López, Aleksander Slominski, Simon Shillaker, Michael Behrendt, Barnard Metzler
2020 arXiv   pre-print
We elaborate in this article why resource disaggregation in serverless computing is the definitive catalyst to enable full transparency in the Cloud.  ...  We demonstrate with two experiments that we can achieve transparency today over disaggregated serverless resources and obtain comparable performance to local executions.  ...  In this case, we evaluate a linear algebra algorithm, Matrix Multiplication (GEMM) which is a good use-case for testing parallel processing on large in-memory data structures.  ... 
arXiv:2006.01251v1 fatcat:sklmosqtvjcppf44rr77okd6oa

SeBS: A Serverless Benchmark Suite for Function-as-a-Service Computing [article]

Marcin Copik, Grzegorz Kwasniewski, Maciej Besta, Michal Podstawski, Torsten Hoefler
2021 arXiv   pre-print
To address this challenge, we propose the Serverless Benchmark Suite: the first benchmark for FaaS computing that systematically covers a wide spectrum of cloud resources and applications.  ...  Function-as-a-Service (FaaS) is one of the most promising directions for the future of cloud services, and serverless functions have immediately become a new middleware for building scalable and cost-efficient  ...  algebra).  ... 
arXiv:2012.14132v2 fatcat:i6hehq4zqfht3f6ovuljvkzvsu

Serverless Straggler Mitigation using Local Error-Correcting Codes [article]

Vipul Gupta, Dominic Carrano, Yaoqing Yang, Vaishaal Shankar, Thomas Courtade, Kannan Ramchandran
2020 arXiv   pre-print
Inexpensive cloud services, such as serverless computing, are often vulnerable to straggling nodes that increase end-to-end latency for distributed computation.  ...  The proposed schemes are inspired by error-correcting codes and employ parallel encoding and decoding over the data stored in the cloud using serverless workers.  ...  Generally, computations underlying several linear algebra and optimization problems tend to be iterative in nature.  ... 
arXiv:2001.07490v1 fatcat:ptbzh4ld3jezphosqkylgpadni
« Previous Showing results 1 — 15 out of 121 results