Application Form for the call for proposals of HPCI projects in 2014
2013/10/25 rev.1.1
Call for Proposals of projects using the HPCI System
for the fiscal year 2014
Application form and additional sheets Preparation Instructions
Instructions
Screening of the proposals of projects using the HPCI System for the fiscal year 2014 for awarding will be based on three documents; (1) proposal application form created via online proposal submission system on the website, (2) Additional Information Sheet 1 starting from page 2 of this document, and (3) Additional Information Sheet 2 in a Excel file format.
In preparing proposal application documents, please read following instructions carefully.
1. Instructions
1) Proposals must be prepared and submitted by Project Representatives.
2) Proposals consists of an application form created via online proposal submission system by entering basic project information, and Additional Information Sheets filled by the applicants and uploaded via online proposal submission system. The application form created via online system printed out and signed by the project representative and a head of the organization the project representative belongs to must be sent to the HPCI Operations Office by registered mail.
3) Application form to be created via online proposal submission system
Following basic project information must be registered via online proposal submission system.
(1) Information about a project representative
(2) Project information such as project name, brief summary of the project, and area of project research.
(3) Project members’ names including the name of a deputy project representative.
(4) Contact person’s name and contact information
(5) Usage information such as category of use, period of use, requesting resources.
*HPCI-IDs and 4 digits reference codes of all the project members are necessary for the item (1) and (3).
**Please make sure that the contact person is a reliable point of contact. Contact persons do not have to possess the HPCI-IDs.
4) Additional Information Sheets
Additional Information Sheet1:
1 Detailed summary of the project research
2 Usage plan of requesting resources and reasons why necessary
3 Organizational plan of the project execution and staffing plan
Additional Information Sheet 2: Please fill in the Excel file attached as appendix. If multiple programs will be used, please add sheets as necessary. Please refer to the
4 Programming information
5 Program readiness
[uploading of the additional information sheets]
Please convert the Word file of Additional Information Sheet 1 to a PDF file. Then combine the Sheet1 (PDF file) and the Sheet2 (Excel file) into a ZIP file for uploading.
2. In preparing Additional Information Sheets 1 and 2.
In preparing Additional Information Sheets 1 and 2, please make sure that 2.(2) usage conditions in the Proposal Preparation Instructions are clearly explained and responded.
The usage conditions are written in red on the Additional Information Sheet 1 and 2. Please respond clearly to each item. If space is not sufficient, please add more space. You could delete these conditions written in red after you fill in the sheets.)
Application Form for the call for proposals of HPCI projects in 2014 - Additional Information Sheet 1
2013/9/2 rev.1.0
Project information
Project Representative / Affiliation： / Name：
Descriptions of the project (Significance, objectives, features, expected achievements, plans of publications)
※ In the category of General Projects of the K computer use, scientific excellence, social significance, expected breakthroughs must be shown. In the category of Junior Researcher Promotion Project, please show that the research project is based on excellent ideas and further development in the future can be expected.
1) Significance of the research
2) Research Objectives
3) Particular features of the research
4) Expected achievements
※ Please describe expected achievements in the first half year and at the end of the full year separately. If you wish to use a secondary choice of the machine in case the first choice will not be awarded, please describe expected achievements from the use of the secondary choice.
[Expected achievements from the first choice of the machine]
[Expected achievement from the secondary choice of the machine]
5) Plan of publications
6) Others
* Please show that the execution of the project and utilization of the results will be limited to peaceful purposes and proper in terms of Basic Act on Science and Technology and social standards.
Spin off effects of the project
※ In case of the industrial use project, please describe exit strategies and marketing strategies. Also, perspectives and planning for the practical use must be written.
The necessity of the requested computational resources.
※ In case of the K computer use, please explain the reason why the K computer is really needed.
※ In the category of computers other than the K computer, show that the project is quite challenging and large scale and capacity computing will be needed.
※ In case of industrial use projects, please show that the scale and complexity of the computation is beyond the in-house capacity of industrial companies.
※ If the project is collaborating with large experimental facility projects by complementarily enhancing the effectiveness of the experimental approache with numerical simulations and excellent results can be expected, please mention about the collaborative project (name of the facility, project name and project IDs).
Utilization plan of the requested computational resources
※ Please describe your plan and schedule to use the computational resources in your research project, and provide the basis of computational resource estimation.
Example of descriptions
List of input data set
Input Data set name / data size (# of elements) / # of time-steps / Already computed?
Data-A (small-scale) / 1.44 million / 100 / yes
Data-B (small-scale) / 38.8 million / 600 / yes
Data-C (middle-scale) / 144 million / 1,000 / no
Data-D (middle-scale) / 144 million / 50,000 / no
Data-E (large-scale) / 1440 million / 1,000 / no
Data-F (large-scale) / 1440 million / 100,000 / no
(Usage Plan of the first half-year)
1. Program Porting and Tuning
(1) Porting the software and test runs with existing input data set:
Data-A: 12 node x 360 sec x 20 cases = 24 node•hour
Data-B: 324 node x 1000 sec x 20 cases = 1,800 node•hour
(2) Performance analysis and speed up of the software : Jun,2014-Jul,2014
Data-B: 324 node x 1000 sec x 30 cases = 2,700 node•hour
2. Middle-scale Input Data test runs and production runs
(1) Preparation of middle-scale input data (Data-C): Jul,2014
(2) Test runs (Data-C) and verification of the results: Aug,2014
Data-C: 1,200 node x 0.5 hour x 20 cases = 12,000 node•hour
(3) Production runs and the analysis of the results: Aug,2014
Data-D: 1,200 node x 19 hour x 40 cases = 912,000 node•hour
3. Preparation of the large-scale input data
(1) Preparation of the large scale input data: Sep,2014
======
Requested resources of the first half-year (Apr,2014- Sep,2014): 928,524 node•hour
======
(Usage plan of the second half-year)
4. Large-scale input data test runs and production runs
(1) Test runs and verification of the results: Oct,2014
Data-E: 12,000 node x 1 hour x 8 cases = 96,000 node•hour
(2) Production runs and the analysis of the results: Nov,2014-Jan,2015
Data-F: 12,000 node x 40 hour x 8 cases = 3,800,000 node•hour
5. Finalization
(1) Additional runs and finalization of the results: Jan,2015- Mar,2015
Additional runs: 175,476 node•hour
======
Requested resources of the 2nd half-year (Oct,2014-Mar,2015): 4,071,476 node•hour
Total (Apr,2014-Mar,2015): 5,000,000 node•hour
======
Organizational plan of the project execution and staffing plan
An example of description>
List of project members
Specialization / Roles
Researcher A / Computational Fluid Dynamics / project leader
Researcher B / numerical analysis / development of computational models, problem definition, preparation of input data and production run
IT Technician C / program porting
External Supporting staff D / HPC / program tuning
Collaborator E / turbulence simulation / verification and evaluation of the computational results
Researcher A has past experiences in the field of computational fluid dynamics …
Researcher B …
IT Technician C has 5-years of experience in program porting …
…
(Staffing plan & training)
Apr,2014: B and C will be given “K-computer users’ Lecture”
Apr, 2014: External support staff D will be contracted. Addition of the project member will be requested.
Mar, 2014: D will be given “K-computer users’ Lecture”.
The requirements of the requested resource use / If the K computer and other HPCI computers are planned to be used at the same time, please clarify the relation between the resource of the first choice and the resources of the secondary choices. You have to clarify whether you wish to use multiple resources at the same time (Simultaneous Use) or you wish to use the secondary choices as alternatives to the first choice in case you will not be awarded with the first choice (Alternative Use). Necessary amount of resources will have to be specified via online proposal submission system on the website. In this sheet, details of the resource uses must be described.
In the Simultaneous Use, the goal can be only achieved by the simultaneous use of multiple resources. If the resource of the first choice will not be awarded, the secondary resources will not be awarded either. For example, other HPCI resources are requested for the purpose of pre/post processing of the computational data from the K computer.
In the Alternative Use, the secondary resources can be awarded in case the first choice will not be awarded. For example, if the resource of the K computer will not be awarded, awarding of other HPCI computational resources will be considered through the screening process. If the resource of the first choice will be awarded, the secondary resources will not be awarded.
Basis of the storage disk space estimation (The maximum necessary disk space and basis for the estimation with each provider)
Necessary local storage with each provider / GB
Basis
※ The quantity of variables and frequency of data accesses must be specified quantitatively.
Shared storage / GB
Basis
In case of the “CPU Simultaneous Use”, there must be a clear usage plan of the HPCI shared computational resource; data analysis must be executed as planned; the data on the HPCI shared storage should be moved and stored in the storage facility of the user after completion of the project.
In case of “Sharing Use”, the storage should serve for sharing of the computational data among the members of the community to jointly analyze the data. Please show the number of users for sharing, easiness of the use, and significance of the sharing.
Related Information
If this is a proposal of continuing project, please write the project ID, project name of presently on-going or already completed project together with the achievement or actual performance.
1) Project ID and Project Name of the continuing project
2) Achievements and the ratio of actually consumed computational resources with respect to the amount of resources awarded in per cent.
HPCI User Information for Junior Researcher Promotion Project Applicant
Name / HPCI-ID
Age / (as of April 1st, 2014)
* Users with age of 39 and under as of April 1st 2014 ( born on and after April 2nd 1974) will be eligible for applying to Junior Researcher Promotion Project.
Project name for public disclosure (Please specify this, in case of proprietary use category of the industrial use project, and if the original project name needs to be kept confidential)※ In case of proprietary use category of the industrial use project, project representatives can specify other project names to be used for the purpose of public disclosure in place of the original project name
Project nameSubstitute name for public disclosure
8
2014 Application Form Supplement
2013/9/2 rev.1.0
Supplementary instructions for preparing "program information" and "current status of program preparation" on the Additional Information Sheet 2Major changes from the previous call are summarized below.
1. Entry form (Additional Information Sheet 2) has become an Excel file sheet.
You have to fill in blue cells by typing in, and you have to choose from pull down list for beige cells. For green cells, please follow instructions below.
2. Confirmation of “current status of program preparation” and skills of project members are required now. You have to provide this confirmation together with other program information respectively.
3. For the box of actual performance, please describe the measured performance by using the same type of input data a) as the input data for the planned production run. Performance information measured in the past can be also used here.
4. You can choose the strong scaling or the weak scaling for performance measurement of your program b).
5. The extent of parallelism, parallelization rate, and parallelization ratio, and parallelization efficiency are automatically calculated in the Excel sheet form.
a):“the same type of input data” are defined as the data used to perform the same routine or function in the program.
b): Strong scaling is defined as how the elapsed time varies with the number of processors for a fixed total problem size. Weak scaling is defined as how the elapsed time varies with the number of processors for a fixed problem size per processor.
*1 Method of numerical analysis
Describe the method of numerical analysis employed in the program.
Examples: Finite Difference Method, Finite Element Method, Finite Volume Method, Density Functional Theory, Molecular Orbital method, Molecular Dynamics method, Monte Carlo method.
*2 Used resource
Please describe the name of the computer used for performance measurements. If the computer of the HPCI computational resources is used, please choose from drop-down list.
*3 Parallelization method
If you use MPI, OpenMP or Automatic parallelization for parallelization of your program, please select “use”, and if not, please select “no use”.
If the other parallelization method is used, please describe the name of the method.
*4 Performance measurement method
Select either the Strong scaling or Weak Scaling as the method of measurement.
*5 Two-point measurements
To estimate scalability of parallelization, measure performance at two different degrees of parallelism (m, n). Set ‘n’ more than twice as large as ‘m’ if possible.