Utilization of JSS2 for AMSR-E L1 data processing

JAXA Supercomputer System Annual Report April 2016-March 2017

Report Number: R16E0093

Responsible Representative: Teruyuki Nakajima(Earth Observation Research Center)
Contact Information: Kazuyoshi Nakamura(nakamura.kazuyoshi@jaxa.jp)
Members: Shin Ishida, Norio Saito, Susumu Saitoh, Makoto Imanaka, Kazuyoshi Nakamura
Subject Category: Space(Satellite utilization)

Abstract

We will provide AMSR-E product in the same format as AMSR2 which is succeeding sensor, and expedite the provision of products to users. And expedite the provision of products to users.

Goal

AMSR-E estimates various geophysical parameters by measuring radio waves in the weak microwave band radiated from the Earth’s surface and the atmosphere with multiple frequencies and multiple polarizations.

Geophysical parameters include water vapor, cloud liquid water, precipitation, sea surface temperature, sea surface wind speed, sea ice concentration, snow water equivalent, and soil moisture.

The long-term geophysical record will play an important role in climate change monitoring and will provide valuable information for understanding the Earth’s climate system, including water and energy circulation.

By reprocessing the AMSR-E product in the same format as AMSR2 which is succeeding sensor, we provide a long-term data set useful for users.

Objective

Utilizing the AMSR-E product generation environment already built on JSS2 last fiscal year, this fiscal year we will reprocess the product in the same format as AMSR2.

Also, shortening reprocessing time will realize to provide products more quickly.

References and Links

Please refer ‘AMSR/AMSR-E Website‘.

Use of the Supercomputer

We will use JSS2 as a computer resource when reprocessing the AMSR-E product to format for long term data set and will reprocess it in a short period of time.

Necessity of the Supercomputer

– By using JSS2 with a large scalability processing environment and executing processing of enormous scenes in parallel, it becomes possible to provide products to users more quickly.

– Cost reduction effect can be expected by utilizing shared processing environment without having processing environment dedicated to project.

Achievements of the Year

In this fiscal year, we reprocessed product that observation period two years worth (2002/06/01 – 2004/05/31) and created L1B, L1R products in hdf5 format. In the next fiscal year, we will reprocess throughout the whole period and promptly release it.

Also, we verified the processing period under parallel execution and and confirmed that the total processing time is shortened.

Number of processing scenes : 20452

Elapsed processing time : About 131.3 hours

Utilizing the JSS2 environment, We were able to improve the cost effectiveness in providing products.

Publications

N/A

Computational Information

Parallelization Methods: Serial
Process Parallelization Methods: n/a
Thread Parallelization Methods: n/a
Number of Processes: 16
Number of Threads per Process: 1
Number of Nodes Used: 2
Elapsed Time per Case (Hours): 0.0064
Number of Cases: 20452

Resources Used

Total Amount of Virtual Cost(Yen): 354,317

Breakdown List by Resources

Computational Resources
System Name	Amount of Core Time(core x hours)	Virtual Cost(Yen)
SORA-MA	0.00	0
SORA-PP	18,857.82	161,008
SORA-LM	0.00	0
SORA-TPP	0.00	0

SORA-FS File System Resources
File System Name	Storage assigned(GiB)	Virtual Cost(Yen)
/home	127.62	1,203
/data	5,053.43	47,669
/ltmp	14,973.97	141,249

J-SPACE Archiving System Resources
Archiving System Name	Storage used(TiB)	Virtual Cost(Yen)
J-SPACE	1.03	3,186

Note: Virtual Cost=amount of cost, using the unit price list of JAXA Facility Utilization program(2016)