Holistic workload scaling : A new approach to compute acceleration in the cloud
Workload scaling is an approach to accelerating computation and thus improving response times by replicating the exact same request multiple times and processing it in parallel on multiple nodes and accepting the result from the first node to finish. This is not unlike a TV game show, where the same...
| Autores Principales: | , , , |
|---|---|
| Formato: | Artículo (Article) |
| Lenguaje: | Inglés (English) |
| Publicado: |
2018
|
| Materias: | |
| Acceso en línea: | http://repository.urosario.edu.co/handle/10336/19089 |
| id |
ir-10336-19089 |
|---|---|
| recordtype |
dspace |
| spelling |
ir-10336-190892019-09-19T12:37:54Z Holistic workload scaling : A new approach to compute acceleration in the cloud Pérez, Juan F. Chen, Lydia Y. Villari, Massimo Ranjan, Rajiv Cloud Computing Stochastic Systems Cloud Environments Inter Processor Communication Mapreudce Model And Analysis Optimal Approaches Parallelilzation Performance Modeling And Analysis Stochastic Variation Economic And Social Effects Probabilidades & matemáticas aplicadas Sistemas estocásticos Comercio electrónico Workload scaling is an approach to accelerating computation and thus improving response times by replicating the exact same request multiple times and processing it in parallel on multiple nodes and accepting the result from the first node to finish. This is not unlike a TV game show, where the same question is given to multiple contestants and the (correct) answer is accepted from the first to respond. This is different than traditional strategies for parallelization as used in, say, MapReduce workloads, where each node runs a subset of the overall workload. There are a variety of strategies that trade off metrics such as cost, utilization, performance, and interprocessor communication requirements. Performance modeling can help determine optimal approaches for different environments and goals. This is important, because poor performance can lead to application and domain-specific losses, such as e-commerce conversions and sales. Performance modeling and analysis plays an important role in designing and driving the selection of resource scaling mechanisms. Such modeling and analysis is complex due to time-varying workload arrival rates and request sizes, and even more complex in cloud environments due to the additional stochastic variation caused by performance interference due to resource sharing across co-located tenants. Moreover, little is known on how to multi-scale, i.e., dynamically and simultaneously scale resources vertically, horizontally, and through workload scaling. In this article, we first demonstrate the effectiveness of multi-scaling in reducing latency, and then discuss the performance modeling challenges, particularly for workload scaling. © 2014 IEEE. 2018 2019-02-15T19:41:21Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion 2325-6095 http://repository.urosario.edu.co/handle/10336/19089 10.1109/MCC.2018.011791711 eng info:eu-repo/semantics/openAccess application/pdf Metrics, K., Blog, , https://blog.kissmetrics.com/loading-time |
| institution |
EdocUR - Universidad del Rosario |
| collection |
DSpace |
| language |
Inglés (English) |
| topic |
Cloud Computing Stochastic Systems Cloud Environments Inter Processor Communication Mapreudce Model And Analysis Optimal Approaches Parallelilzation Performance Modeling And Analysis Stochastic Variation Economic And Social Effects Probabilidades & matemáticas aplicadas Sistemas estocásticos Comercio electrónico |
| spellingShingle |
Cloud Computing Stochastic Systems Cloud Environments Inter Processor Communication Mapreudce Model And Analysis Optimal Approaches Parallelilzation Performance Modeling And Analysis Stochastic Variation Economic And Social Effects Probabilidades & matemáticas aplicadas Sistemas estocásticos Comercio electrónico Pérez, Juan F. Chen, Lydia Y. Villari, Massimo Ranjan, Rajiv Holistic workload scaling : A new approach to compute acceleration in the cloud |
| description |
Workload scaling is an approach to accelerating computation and thus improving response times by replicating the exact same request multiple times and processing it in parallel on multiple nodes and accepting the result from the first node to finish. This is not unlike a TV game show, where the same question is given to multiple contestants and the (correct) answer is accepted from the first to respond. This is different than traditional strategies for parallelization as used in, say, MapReduce workloads, where each node runs a subset of the overall workload. There are a variety of strategies that trade off metrics such as cost, utilization, performance, and interprocessor communication requirements. Performance modeling can help determine optimal approaches for different environments and goals. This is important, because poor performance can lead to application and domain-specific losses, such as e-commerce conversions and sales. Performance modeling and analysis plays an important role in designing and driving the selection of resource scaling mechanisms. Such modeling and analysis is complex due to time-varying workload arrival rates and request sizes, and even more complex in cloud environments due to the additional stochastic variation caused by performance interference due to resource sharing across co-located tenants. Moreover, little is known on how to multi-scale, i.e., dynamically and simultaneously scale resources vertically, horizontally, and through workload scaling. In this article, we first demonstrate the effectiveness of multi-scaling in reducing latency, and then discuss the performance modeling challenges, particularly for workload scaling. © 2014 IEEE. |
| format |
Artículo (Article) |
| author |
Pérez, Juan F. Chen, Lydia Y. Villari, Massimo Ranjan, Rajiv |
| author_facet |
Pérez, Juan F. Chen, Lydia Y. Villari, Massimo Ranjan, Rajiv |
| author_sort |
Pérez, Juan F. |
| title |
Holistic workload scaling : A new approach to compute acceleration in the cloud |
| title_short |
Holistic workload scaling : A new approach to compute acceleration in the cloud |
| title_full |
Holistic workload scaling : A new approach to compute acceleration in the cloud |
| title_fullStr |
Holistic workload scaling : A new approach to compute acceleration in the cloud |
| title_full_unstemmed |
Holistic workload scaling : A new approach to compute acceleration in the cloud |
| title_sort |
holistic workload scaling : a new approach to compute acceleration in the cloud |
| publishDate |
2018 |
| url |
http://repository.urosario.edu.co/handle/10336/19089 |
| _version_ |
1645141929865773056 |
| score |
12,131701 |