Filters








250 Hits in 3.6 sec

The SPEC cloud group's research vision on FaaS and serverless architectures

Erwin van Eyk, Alexandru Iosup, Simon Seif, Markus Thömmes
2017 Proceedings of the 2nd International Workshop on Serverless Computing - WoSC '17  
We clarify the term serverless, by reducing it to cloud functions as programming units, and a model of executing simple and complex (e.g., workflows of) functions with operations managed primarily by the  ...  An emerging class of cloud-based software architectures, serverless, focuses on providing software architects the ability to execute arbitrary functions with small overhead in server management, as Function-as-a-service  ...  This raises new challenges for granular, function-oriented scheduling. To make data-driven decisions in-time, a challenge of scheduling scalability arises.  ... 
doi:10.1145/3154847.3154848 dblp:conf/middleware/EykIST17 fatcat:d5vcefd6cbei3pv25jw4gcl3ai

Practical Scheduling for Real-World Serverless Computing [article]

Kostis Kaffes and Neeraja J. Yadwadkar and Christos Kozyrakis
2021 arXiv   pre-print
We use these insights to design Hermes, a scheduler for serverless functions with three key characteristics.  ...  Serverless computing has seen rapid growth due to the ease-of-use and cost-efficiency it provides. However, function scheduling, a critical component of serverless systems, has been overlooked.  ...  Nevertheless, existing schedulers use techniques and policies that are ill-suited for the characteristics of serverless functions: Existing Schedulers for Serverless functions: Apache OpenWhisk [5] is  ... 
arXiv:2111.07226v1 fatcat:rfuz2fuskzdulfu5wb32f654yy

OpenDC Serverless: Design, Evaluation, and Implementation of a FaaS Platform Simulator

Soufiane Jounaid, Alexandru Iosup, Erwyn Van Eyk, Georgios Andreadis
2020 Zenodo  
The simulator exposes custom interfaces for the implementation of resource allocation, management, and scheduling policies. It further supports the modification of its core architectural components.  ...  Function-as-a-Service (FaaS) pertains to a recent advancement in cloud computing known as serverless computing.  ...  Naturally, the tracking granularity has been lowered to functions in OpenDC Serverless.  ... 
doi:10.5281/zenodo.4046674 fatcat:5p3maohqhzc6bapixwzmvesyse

Putting the "Micro" Back in Microservice

Sol Boucher, Anuj Kalia, David G. Andersen, Michael Kaminsky
2018 USENIX Annual Technical Conference  
The result is microsecond-granularity launch latency, and microsecondscale preemptive scheduling using high-precision timers.  ...  The most recent face to this trend is the "serverless" model, in which individual functions, or microservices, are executed on demand.  ...  National Science Foundation under award CCF-1535821 and the Intel Science and Technology Center for Visual Cloud Systems.  ... 
dblp:conf/usenix/BoucherKAK18 fatcat:xxl3vw7dx5c4vpfg5etvcntrfa

Interim Report on Adaptive Event Dispatching in Serverless Computing Infrastructures [article]

Manuel Stein
2019 arXiv   pre-print
This interim report investigates the economical importance of the emerging trend and asserts that existing serverless platforms still do not optimize for data locality, whereas a variety of scheduling  ...  Serverless computing is an emerging service model in distributed computing systems.  ...  , which fits with serverless granularity.  ... 
arXiv:1901.02680v1 fatcat:5mofzvliyfbwjlbczj7bvqgwmy

A Literature Review on Serverless Computing [article]

Jinfeng Wen, Zhenpeng Chen, Xuanzhe Liu
2022 arXiv   pre-print
Serverless computing is an emerging cloud computing paradigm. Moreover, it has become an attractive development option for cloud-based applications for software developers.  ...  Finally, based on our analysis, we discuss some key challenges and envision promising opportunities for future research on the serverless platform side, serverless application side, and serverless computing  ...  [121] presented a new scheduling architecture with a centralized core-granular scheduler. Core granular can directly assign functions to individual cores, guaranteeing performance stability.  ... 
arXiv:2206.12275v3 fatcat:uk5rrpxu5vaixlkudoshyjrq64

rFaaS: Enabling High Performance Serverless with RDMA and Decentralization [article]

Marcin Copik, Konstantin Taranov, Alexandru Calotoiu, Torsten Hoefler
2022 arXiv   pre-print
We identify key limitations of modern serverless systems - centralized scheduling and inefficient network transport - and propose an overhaul of FaaS architectures with decentralized allocations and low-latency  ...  Function-as-a-Service, a modern cloud programming paradigm for pay-as-you-go execution of stateless functions, brought the elasticity needed to take advantage of ephemeral resources.  ...  These servers offer idle and unused hardware resources (CPU cores, memory) to support the dynamic execution of serverless functions.  ... 
arXiv:2106.13859v2 fatcat:z54qxj4yovb3rism3sflntwqqy

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

Benjamin Carver, Jingyuan Zhang, Ao Wang, Yue Cheng
2019 arXiv   pre-print
These characteristics make them well-suited for a new cloud computing model called serverless computing or Function-as-a-Service (FaaS), which has become prevalent in recent years.  ...  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.  ...  centralized scheduler.  ... 
arXiv:1910.05896v2 fatcat:itnhnurocvdozghbvxeqrnwjpi

HiveMind: A Scalable and Serverless Coordination Control Platform for UAV Swarms [article]

Justin Hu, Ariana Bruno, Brian Ritchken, Brendon Jackson, Mateo Espinosa, Aditya Shah, Christina Delimitrou
2020 arXiv   pre-print
HiveMind employs an event-driven serverless framework to run tasks on the cluster, guarantees fault tolerance both in the edge devices and serverless functions, and handles straggler tasks and underperforming  ...  There are two main trains of thought for controlling devices in such swarms; centralized and distributed control.  ...  in the cloud, and also launch serverless functions for inference [57] .  ... 
arXiv:2002.01419v1 fatcat:wn6yawh4ezdbjcjlzm7n5zuluu

Fifer: Tackling Underutilization in the Serverless Era [article]

Jashwant Raj Gunasekaran, Prashanth Thinakaran, Nachiappan Chidambaram, Mahmut T. Kandemir, Chita R. Das
2020 arXiv   pre-print
Datacenters are witnessing a rapid surge in the adoption of serverless functions for microservices-based applications.  ...  To address them, we propose Fifer – an adaptive resource management framework to efficiently manage function-chains on serverless platforms.  ...  microservices 1 , also known 1 A microservice is the smallest granularity of an application performing an independent function, a.k.a functions in serverless domain. as microservice-chains [64, 89]  ... 
arXiv:2008.12819v1 fatcat:zcvn6yd2xzfwtaiiqs4attqkua

Archipelago: A Scalable Low-Latency Serverless Platform [article]

Arjun Singhvi, Kevin Houck, Arjun Balasubramanian, Mohammed Danish Shaikh, Shivaram Venkataraman, Aditya Akella
2019 arXiv   pre-print
Our analysis of popular serverless workloads indicates that schedulers need to handle functions that are very short-lived, have unpredictable arrival patterns, and require expensive setup of sandboxes.  ...  The challenge of running a large number of such functions in a multi-tenant cluster makes existing scheduling frameworks unsuitable.  ...  While centralized schedulers can make optimal scheduling decisions, when incoming workload grows arbitrarily, a centralized approach can easily become a scalability bottleneck.  ... 
arXiv:1911.09849v1 fatcat:2lnu4wzowrhxlp27stvotuk4f4

Serverless Computing: A Security Perspective [article]

Eduard Marin, Diego Perino, Roberto Di Pietro
2022 arXiv   pre-print
However, in this quest for efficiency, security is not ranked among the top priorities, also because of the (misleading) belief that current solutions developed for virtualised environments could be applied  ...  Unfortunately, this is not the case, due to the highlighted idiosyncratic features of serverless computing.  ...  Acknowledgements We would like to thank the anonymous reviewers for their comments, that helped us to improve the quality of the manuscript.  ... 
arXiv:2107.03832v2 fatcat:kbidlkkpk5cclbxciy2mrgpepq

SLA-Driven ML Inference Framework for Clouds with Hetergeneous Accelerators

Junguk Cho, Diman Zad Tootaghaj, Lianjie Cao, Puneet Sharma
2022 Conference on Machine Learning and Systems  
Second, ignoring infrastructure heterogeneity for workload scheduling and inference request distribution can lead to further performance inefficiencies.  ...  We schedule functions on the suitable GPU accelerators and proportionally distribute inference requests to the deployed functions based on the autoscaling decision.  ...  ACKNOWLEDGEMENT We thank the anonymous reviewers for their feedback on earlier drafts of this paper. We wish to thank Eric Wu in Hewlett Packard Labs for his support in setting up the testbed.  ... 
dblp:conf/mlsys/ChoTCS22 fatcat:uxfzaro2lza3ti7bfhe3onhqcq

Adaptive Event Dispatching in Serverless Computing Infrastructures [article]

Manuel Stein
2019 arXiv   pre-print
The term "serverless" has become a synonym for the entirely resource-transparent deployment model of cloud-based event-driven distributed applications.  ...  This work investigates how adaptive event dispatching can improve serverless platform resource efficiency and contributes a novel approach that allows for better scaling and fitting of the platform's resource  ...  The first set of tests verifies the worker's work-conserving, processorsharing execution scheduling using the M/M/1 queue model for a single-core test and M/M/c model for multi-core tests.  ... 
arXiv:1901.03086v1 fatcat:wzgif4fsdnhfvhetxxhritsl2m

Memory deduplication for serverless computing with Medes

Divyanshu Saxena, Tao Ji, Arjun Singhvi, Junaid Khalid, Aditya Akella
2022 Proceedings of the Seventeenth European Conference on Computer Systems  
Finally, we develop a simple sandbox management policy that exposes a narrow, intuitive interface for operators to trade-off performance for memory by jointly controlling warm and dedup sandboxes.  ...  Serverless platforms today impose rigid trade-offs between resource use and user-perceived performance.  ...  Acknowledgements We would like to thank the anonymous reviewers of Eu-roSys'22 and the members of UTNS Lab for their regular insightful comments and suggestions.  ... 
doi:10.1145/3492321.3524272 fatcat:otmztsvgwzdijl4wmtj25dbmbq
« Previous Showing results 1 — 15 out of 250 results