Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Web portals for high-performance computing: a survey
Calegari P., Levrier M., Balczyski P.  ACM Transactions on the Web 13 (1): 1-36, 2019. Type: Article
Date Reviewed: Sep 24 2021

High-performance computing (HPC) offers researchers the performance they need to solve complex problems. However, as these researchers are specialized in areas not related to HPC, web interfaces facilitate the submission of their jobs to HPC systems, as well as the visualization of their results. Web HPC systems are the topic of this survey.

Three parts can be distinguished in this paper. First is a presentation of HPC portals, covering sections 2 through 4. The state of the art of HPC portals can be found in section 5. And, finally, section 6 presents XCS3, the HPC portal in which the authors are involved.

The presentation of HPC portals starts in section 2, with a use-case overview of HPC portals, discussing their differences with other portals. HPC portal requirements complete this presentation. These requirements are classified in two types: functional features and nonfunctional features. Functional features, covered in section 3, are organized in categories: mandatory features, key features, and nice-to-have features. Nonfunctional features are collected in section 4. They include categories such as security, usability, performance, and reliability.

The state of the art in section 5 is focused on functional features. However, some nonfunctional features, easy to check, are also included: three from the category of security authorizations and three from the usability area. Table 2 summarizes these results. The portals reviewed are selected with these criteria: they are multidomain (that is, community-specific web portals are intentionally excluded); they use a graphical user interface (GUI) (command-line interface–based portals are not considered); and they are not the HPC offerings of major cloud players such as Microsoft Azure Big Compute, Google Cloud Platform (GCP), or Amazon Web Service (AWS). Twenty-three portals (24 including the XCS3 portal described in the third part) are reviewed, checking the features present in each of them. Some trends of HPC portals are pointed out in this section.

The presentation of the XCS3 portal in the third part of this paper covers its history and support of HPC portal functional features (section 6), its architecture (section 7), and the revision of its support of nonfunctional features (section 8). The evolution of the XCS3 portal shows what features were initially included in previous versions of this portal and how it has been enriched over time with additional features. This evolution is the result of the evolution of the HPC portals market. HPC portals evolve to provide new features that their users want. Nonfunctional features in section 8 include all features presented in section 4, even those that were not checked in the state of the art of section 5.

The distinction of three parts in this paper is helpful for identifying potential readers. As for the first part, it is a good introduction to HPC portals for any person aiming to learn a bit about these systems. It is clear, well organized, and easy to follow. Besides the main features of HPC portals, the paper clearly explains the differences of the portals surveyed, with offerings from major cloud players that general readers are more familiar with (Microsoft Azure Big Compute and so on). This can be helpful to get a good idea of what HPC portals are and are not.

No previous knowledge of the field is needed. As for the second part on the state of the art and the description of the XSC3 portal, it is clear that it is more appropriate for people familiar with HPC portals, either developers or users.

Reviewer:  Mercedes Martínez González Review #: CR147362
Bookmark and Share
 
Cloud Computing (C.2.4 ... )
 
 
Graphical User Interfaces (GUI) (H.5.2 ... )
 
 
Model Classification (I.6.1 ... )
 
 
Types Of Simulation (Continuous And Discrete) (I.6.1 ... )
 
 
Web-Based Services (H.3.5 ... )
 
 
General (I.6.0 )
 
  more  
Would you recommend this review?
yes
no
Other reviews under "Cloud Computing": Date
Orchestrating big data analysis workflows in the cloud: research challenges, survey, and future directions
Barika M., Garg S., Zomaya A., Wang L., Moorsel A., Ranjan R.  ACM Computing Surveys 52(5): 1-41, 2019. Type: Article
Mar 2 2021
Secure sensor cloud
Kumar V., Sen A., Madria S.,  Morgan&Claypool Publishers, San Rafael, CA, 2018. 126 pp. Type: Book (978-1-681734-68-2)
Dec 21 2020
 Zooming in on wide-area latencies to a global cloud provider
Jin Y., Renganathan S., Ananthanarayanan G., Jiang J., Padmanabhan V., Schroder M., Calder M., Krishnamurthy A.  SIGCOMM 2019 (Proceedings of the ACM Special Interest Group on Data Communication, Beijing, China,  Aug 19-23, 2019) 104-116, 2019. Type: Proceedings
Feb 27 2020
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright © 2000-2021 ThinkLoud, Inc.
Terms of Use
| Privacy Policy