Using a Microbenchmark to Compare Function as a Service Solutions

Timon Back; Vasilios Andrikopoulos

doi:10.1007/978-3-319-99819-0_11

Using a Microbenchmark to Compare Function as a Service Solutions

Timon Back, Vasilios Andrikopoulos

Software Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

38 Citations (Scopus)

971 Downloads (Pure)

Abstract

The Function as a Service (FaaS) subtype of serverless computing provides the means for abstracting away from servers on which developed software is meant to be executed. It essentially offers an event-driven and scalable environment in which billing is based on the invocation of functions and not on the provisioning of resources. This makes it very attractive for many classes of applications with bursty workload. However, the terms under which FaaS services are structured and offered to consumers uses mechanisms like GB–seconds (that is, X GigaBytes of memory used for Y seconds of execution) that differ from the usual models for compute resources in cloud computing. Aiming to clarify these terms, in this work we develop a microbenchmark that we use to evaluate the performance and cost model of popular FaaS solutions using well known algorithmic tasks. The results of this process show a field still very much under development, and justify the need for further extensive benchmarking of these services.

Original language	English
Title of host publication	Service-Oriented and Cloud Computing
Subtitle of host publication	7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings
Editors	Kyriakos Kritikos, Pierluigi Plebani, Flavio De Paoli
Place of Publication	Cham
Publisher	Springer
Pages	146-160
ISBN (Electronic)	978-3-319-99819-0
ISBN (Print)	978-3-319-99818-3
DOIs	https://doi.org/10.1007/978-3-319-99819-0_11
Publication status	Published - 2018
Event	7th IFIP WG 2.14 European Conference, ESOCC 2018 - Como, Italy Duration: 12-Sept-2018 → 14-Sept-2018

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Number	11116

Name	Programming and Software Engineering
Publisher	Springer
Number	11116

Conference

Conference	7th IFIP WG 2.14 European Conference, ESOCC 2018
Country/Territory	Italy
City	Como
Period	12/09/2018 → 14/09/2018

Keywords

Function as a Service (FaaS)
Microbenchmark
Performance evaluation
Cost evaluation

Access to Document

10.1007/978-3-319-99819-0_11

Using a Microbenchmark to Compare Function as a Service SolutionsFinal author's version, 434 KB

Handle.net

Persistent link

Embargoed Document

Using a Microbenchmark to Compare Function as a Service Solutions)
Final publisher's version, 1.36 MB
Request copy

Cite this

Back, T., & Andrikopoulos, V. (2018). Using a Microbenchmark to Compare Function as a Service Solutions. In K. Kritikos, P. Plebani, & F. De Paoli (Eds.), Service-Oriented and Cloud Computing: 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings (pp. 146-160). (Lecture Notes in Computer Science; No. 11116), (Programming and Software Engineering; No. 11116). Springer. https://doi.org/10.1007/978-3-319-99819-0_11

Back, Timon ; Andrikopoulos, Vasilios. / Using a Microbenchmark to Compare Function as a Service Solutions. Service-Oriented and Cloud Computing: 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings. editor / Kyriakos Kritikos ; Pierluigi Plebani ; Flavio De Paoli. Cham : Springer, 2018. pp. 146-160 (Lecture Notes in Computer Science; 11116). (Programming and Software Engineering; 11116).

@inproceedings{2a9cb1e6b41241c0bc95f5d8f2802453,

title = "Using a Microbenchmark to Compare Function as a Service Solutions",

abstract = "The Function as a Service (FaaS) subtype of serverless computing provides the means for abstracting away from servers on which developed software is meant to be executed. It essentially offers an event-driven and scalable environment in which billing is based on the invocation of functions and not on the provisioning of resources. This makes it very attractive for many classes of applications with bursty workload. However, the terms under which FaaS services are structured and offered to consumers uses mechanisms like GB–seconds (that is, X GigaBytes of memory used for Y seconds of execution) that differ from the usual models for compute resources in cloud computing. Aiming to clarify these terms, in this work we develop a microbenchmark that we use to evaluate the performance and cost model of popular FaaS solutions using well known algorithmic tasks. The results of this process show a field still very much under development, and justify the need for further extensive benchmarking of these services.",

keywords = "Function as a Service (FaaS), Microbenchmark, Performance evaluation, Cost evaluation",

author = "Timon Back and Vasilios Andrikopoulos",

year = "2018",

doi = "10.1007/978-3-319-99819-0_11",

language = "English",

isbn = "978-3-319-99818-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

number = "11116",

pages = "146--160",

editor = "Kyriakos Kritikos and Pierluigi Plebani and {De Paoli}, Flavio",

booktitle = "Service-Oriented and Cloud Computing",

note = "7th IFIP WG 2.14 European Conference, ESOCC 2018 ; Conference date: 12-09-2018 Through 14-09-2018",

}

Back, T & Andrikopoulos, V 2018, Using a Microbenchmark to Compare Function as a Service Solutions. in K Kritikos, P Plebani & F De Paoli (eds), Service-Oriented and Cloud Computing: 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings. Lecture Notes in Computer Science, no. 11116, Programming and Software Engineering, no. 11116, Springer, Cham, pp. 146-160, 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, 12/09/2018. https://doi.org/10.1007/978-3-319-99819-0_11

Using a Microbenchmark to Compare Function as a Service Solutions. / Back, Timon; Andrikopoulos, Vasilios.
Service-Oriented and Cloud Computing: 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings. ed. / Kyriakos Kritikos; Pierluigi Plebani; Flavio De Paoli. Cham: Springer, 2018. p. 146-160 (Lecture Notes in Computer Science; No. 11116), (Programming and Software Engineering; No. 11116).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Using a Microbenchmark to Compare Function as a Service Solutions

AU - Back, Timon

AU - Andrikopoulos, Vasilios

PY - 2018

Y1 - 2018

N2 - The Function as a Service (FaaS) subtype of serverless computing provides the means for abstracting away from servers on which developed software is meant to be executed. It essentially offers an event-driven and scalable environment in which billing is based on the invocation of functions and not on the provisioning of resources. This makes it very attractive for many classes of applications with bursty workload. However, the terms under which FaaS services are structured and offered to consumers uses mechanisms like GB–seconds (that is, X GigaBytes of memory used for Y seconds of execution) that differ from the usual models for compute resources in cloud computing. Aiming to clarify these terms, in this work we develop a microbenchmark that we use to evaluate the performance and cost model of popular FaaS solutions using well known algorithmic tasks. The results of this process show a field still very much under development, and justify the need for further extensive benchmarking of these services.

AB - The Function as a Service (FaaS) subtype of serverless computing provides the means for abstracting away from servers on which developed software is meant to be executed. It essentially offers an event-driven and scalable environment in which billing is based on the invocation of functions and not on the provisioning of resources. This makes it very attractive for many classes of applications with bursty workload. However, the terms under which FaaS services are structured and offered to consumers uses mechanisms like GB–seconds (that is, X GigaBytes of memory used for Y seconds of execution) that differ from the usual models for compute resources in cloud computing. Aiming to clarify these terms, in this work we develop a microbenchmark that we use to evaluate the performance and cost model of popular FaaS solutions using well known algorithmic tasks. The results of this process show a field still very much under development, and justify the need for further extensive benchmarking of these services.

KW - Function as a Service (FaaS)

KW - Microbenchmark

KW - Performance evaluation

KW - Cost evaluation

U2 - 10.1007/978-3-319-99819-0_11

DO - 10.1007/978-3-319-99819-0_11

M3 - Conference contribution

SN - 978-3-319-99818-3

T3 - Lecture Notes in Computer Science

SP - 146

EP - 160

BT - Service-Oriented and Cloud Computing

A2 - Kritikos, Kyriakos

A2 - Plebani, Pierluigi

A2 - De Paoli, Flavio

PB - Springer

CY - Cham

T2 - 7th IFIP WG 2.14 European Conference, ESOCC 2018

Y2 - 12 September 2018 through 14 September 2018

ER -

Back T, Andrikopoulos V. Using a Microbenchmark to Compare Function as a Service Solutions. In Kritikos K, Plebani P, De Paoli F, editors, Service-Oriented and Cloud Computing: 7th IFIP WG 2.14 European Conference, ESOCC 2018, Como, Italy, September 12-14, 2018, Proceedings. Cham: Springer. 2018. p. 146-160. (Lecture Notes in Computer Science; 11116). (Programming and Software Engineering; 11116). doi: 10.1007/978-3-319-99819-0_11

Using a Microbenchmark to Compare Function as a Service Solutions

Abstract

Publication series

Conference

Keywords

Access to Document

Handle.net

Embargoed Document

Fingerprint

Cite this