Step 1. Initiate Proxy

After logining, you have to import your usercert.pem, userkey.pem, and enter passphrase to initiate proxy. System will verify your identity, then help you to submit jobs.

**The Proxy Initial page.**
(Click the picture to enlarge)

Field Name	Description
Your Certificate	Select the `usercert.pem` file.
Your Private	Select the `userkey.pem` file.
Passphrase	The Certificate password.
Lifetime	Available time after proxy initial, default time is 72 hours, maximum time: 168 hours, minimum time: 1 hour.

Step 1.1 First Initiate Proxy

First proxy initial, you need to setup some field of Profile page.

**Check your identity.**
(Click the picture to enlarge)

Step 2. Upload File

At Data Management page, upload your input files or binary files before job submission.

**Prepare your source file.**
(Click the picture to enlarge)

Attention:
☆ File name cannot be same
☆ A Dataset is defined as a set of File
☆ A Container is defined as a set of Dataset
☆ Dataset and Container can submit job in batch mode
☆ Container include Container and Dataset, but Dataset only include File
☆ Unlimited multi-layer(but only Container can repeat)
☆ Cannot change the name of Container, Dataset and File
☆ Only File can be removed
☆ If you have several input files, you should compress them (tar) by yourself before submission, and write decompress command (tar –zxf filename) in your script.

Step 3. Job Submission

**Start to work**
(Click the picture to enlarge)

The describe of parameters are as follow:

Task Name: Defined name for task. Note that task name cannot be the same.
Executable: Executable files (binary or script)
User’s Library: Function library used for executing task, can be left blank or select your input files.
Input Dataset: Choose input file, dataset or left blank.
☆ If you left blank, can upload files up to 1000.
☆ Select file or dataset, system may calculate number of files, the limit is
```
“Num of Jobs” × “Maximum files per job” ~= “Num of Input Files”
```
Output Dataset Name: Dataset name for output after job finish. Default name is Task Name.
Computing Site: Select which site you want to submit. Default site is to select automatically the most free site to submit your job.
Job Type: Select your job type.
Number of Jobs: Resource used for the task. Single job submission is set to 1; Batch job submission will separate according to the number of input files.
Maximum file per job: Set the maximum input files for one job.
Job Parameter: Provides advanced option (please using blank between parameters)

Step 4. Job Monitoring

At Job Monitoring page, you can monitor job status and according time to search.

**Six statuses of Job.**
(Click the picture to enlarge)

Step 5. Download File

Results can be downloaded at Data Management page or perform other operations when job is completed.

**Download it.** (Click the picture to enlarge)

Other Explanations:

#Scope

To divide into two part
- User’s scope (for example: twgrid-user-mandy4318)
- Shared scope
System only allow copy files in shared scope
You can copy File to other Dataset or copy File from shared scope to Dataset of your user’s scope.
You can copy Dataset to other Container or copy Dataset from shared scope to Container of your user’s scope.