Beyond Load Balancing: Package-Aware Scheduling for Serverless Platforms

Aumala, Gabriel; Boza, Edwin F.; Ortiz-Aviles, Luis; Totoy, Gustavo; Abad, Cristina L.

doi:10.1109/ccgrid.2019.00042

Cited by 48 publications

(24 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A developer would specify the QoS requirements for a DAG based application either as a whole [120] or for each individual task. A serverless application could also be monolithic, composed of a single function, representing a single task [63], [8], [125], [86], [88], [69].…”

Section: Application Modelmentioning

confidence: 99%

“…In addition to the above, many studies focus on achieving data locality via load balancing methods to direct new function requests to workers with pro-actively spawned containers or containers that could be re-used [120], [82]. Research works have also focused on attaining data locality among functions in terms of the runtime package dependencies of functions, by routing functions requiring similar packages to the same node for execution [8]. Lee et al propose a greedy load balancing algorithm which tries to maximize locality and improve the cache-hit ratio while also minimizing load imbalance among nodes [79].…”

Section: Workloadmentioning

confidence: 99%

“…One main cause for high application latency in serverless environments is the cold start delay in resource setup, which tends to become signiicant compared to application execution times [88]. Scheduling algorithms along with diferent container pool management techniques, customized sandboxing methods, and low-overhead runtimes to reduce resource setup times are explored in literature to reduce cold start delay [66], [8], [120], [82], [53], [3], [93]. Latency caused by CPU contention is also studied and dynamic resource control methods are proposed for serverless applications [88], [127], [74], [63].…”

Section: Qos Goalmentioning

confidence: 99%

See 2 more Smart Citations

A Holistic View on Resource Management in Serverless Computing Environments: Taxonomy and Future Directions

2022

View full text Add to dashboard Cite

Serverless computing has emerged as an attractive deployment option for cloud applications in recent times. The unique features of this computing model include rapid auto-scaling, strong isolation, fine-grained billing options and access to a massive service ecosystem which autonomously handles resource management decisions. This model is increasingly being explored for deployments in geographically distributed edge and fog computing networks as well, due to these characteristics. Effective management of computing resources has always gained a lot of attention among researchers. The need to automate the entire process of resource provisioning, allocation, scheduling, monitoring and scaling, has resulted in the need for specialized focus on resource management under the serverless model. In this article, we identify the major aspects covering the broader concept of resource management in serverless environments and propose a taxonomy of elements which influence these aspects, encompassing characteristics of system design, workload attributes and stakeholder expectations. We take a holistic view on serverless environments deployed across edge, fog and cloud computing networks. We also analyse existing works discussing aspects of serverless resource management using this taxonomy. This article further identifies gaps in literature and highlights future research directions for improving capabilities of this computing model.

show abstract

Section: Application Modelmentioning

confidence: 99%

Section: Workloadmentioning

confidence: 99%

Section: Qos Goalmentioning

confidence: 99%

See 1 more Smart Citation

A Holistic View on Resource Management in Serverless Computing Environments: Taxonomy and Future Directions

2022

View full text Add to dashboard Cite

show abstract

“…Cold-start latency depends on many additional factors: the language used for programming the function, the number and types of libraries, their dependencies, etc. While many of these factors are under a developer's control and considered for optimization [9,49,54,58,73,86], one can achieve only a limited reduction of the cold-start latency.…”

Section: Background and Related Workmentioning

confidence: 99%

Sledge

Gadepalli

McBride

Peach

et al. 2020

Proceedings of the 21st International Middleware Conference

View full text Add to dashboard Cite

“…• Cristina L. Abad: Beyond Load Balancing: Package-Aware Scheduling for Serverless Platforms [1] to appear in CCGrid 2019 • André Bauer: Chameleon: A Hybrid, Proactive Auto-Scaling Mechanism on a Level-Playing Field [2] in IEEE TPDS 30 4 The workshop will conclude with a discussion session and lightning talks on the recent achievements of the SPEC RG Cloud group: Contributors: Erwin van Eyk, Johannes Grohmann, Simon Eismann, André Bauer, Laurens Versluis, Lucian Toader, Norbert Schmitt, Nikolas Herbst, Cristina L. Abad, and Alexandru Iosup The organizing committee would like to thank all members of the HotCloudPerf program committee for their intense and constructive reviews and discussion. Furthermore, thanks go to the FAS* workshop track chairs Ingrid Nunes and Robert Birke, the FAS* proceedings chair Ahmed Ali-Eldin Hassan and the complete local organisation team.…”

mentioning

confidence: 99%

HotCloudPerf 2019: The Second Workshop on Hot Topics in Cloud Computing Performance

2019

2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems (FAS*W)

View full text Add to dashboard Cite

The organisers of the Second Workshop on Hot Topics in Cloud Computing Performance (HotCloudPerf 2019) are delighted to welcome you to the workshop proceedings as part of the FAS*(IEEE ICAC/SASO) joint conference companion. The HotCloudPerf 2019 workshop is a full day workshop on Sunday, June 16, taking place as part of the FAS*(IEEE ICAC/SASO) conference week in Umea. Each year, the workshop chooses a focus theme to explore; for 2019, the theme is "Performance from the cloud datacenter to the edge." Cloud computing is emerging as one of the most profound changes in the way we build and use IT. The use of global services in public clouds is increasing, and the lucrative and rapidly growing global cloud market already supports over 1 million IT-related jobs. However, it is currently challenging to make the IT services offered by public and private clouds performant (in an extended sense) and efficient. Emerging architectures, techniques, and real-world systems include hybrid deployment, serverless operation, everything as a service, complex workflows, auto-scaling and-tiering, etc. It is unclear to which extent traditional performance engineering, software engineering, and system design and analysis tools can help with understanding and engineering these emerging technologies. The community also needs practical tools and powerful methods to address hot topics in cloud computing performance.

show abstract

Beyond Load Balancing: Package-Aware Scheduling for Serverless Platforms

Cited by 48 publications

References 21 publications

A Holistic View on Resource Management in Serverless Computing Environments: Taxonomy and Future Directions

A Holistic View on Resource Management in Serverless Computing Environments: Taxonomy and Future Directions

Sledge

HotCloudPerf 2019: The Second Workshop on Hot Topics in Cloud Computing Performance

Contact Info

Product

Resources

About