[Draft Only] DRMAAv2 Implemantion for PBSPro

Target release	1.0
Epic
Document status	DRAFT
Document owner	Vinod (Deactivated)
Designer	Ram Pranesh Suresh Thelkar (Deactivated) Arun Grover (Deactivated) Vinod (Deactivated)
Developers	Ram Pranesh Suresh Thelkar (Deactivated) Arun Grover (Deactivated) Vinod (Deactivated)
QA	<TBD>

Goals

Implementation of DRMAAv2 Specification

Background and strategic fit

The Distributed Resource Management Application API Version 2 (DRMAA) specification defines an interface for tightly coupled, but still portable access to the majority of DRM systems. PBS is Distributed Resource Management system(DRMS). Standardized access to PBS product through a DRMAAv2 implementation(API). High-level API designers, meta-scheduler architects and end users can rely on DRMAAv2 implementations for a unified access to execution resources.

DRMAAv2 Specification (Distributed Resource Management Application API 2)

Overview

As we all know open standards are taking major role in the exponential growth in many areas. DRMAA2 is such an open standard. It is the successor of the wide-spread DRMAA. DRMAA is generally used for submitting jobs (or creating job work-flows) into a compute cluster by using a cluster resource management system like PBS Professional. Unlike DRMAA with DRMAA2 you can not only submit jobs, you can also get cluster related information, like getting host names, types and status information or insight about queues configured in the resource management system. You can also monitor jobs not submitted within the application. Overall it covers many more use cases than the old DRMAA standard.

More details of DRMAA2 specification can be found from the following document.

https://www.ogf.org/documents/GFD.231.pdf

Assumptions

Requirements

The implementation of DRMAA2 specification for PBS Pro is broadly divided into the following requirements/user stories.

#	Title	User Story	Importance	Notes
1	Control and monitoring of Jobs
2	Control and monitoring of Job arrays
3	Control and monitoring of Reservations
4	Control and monitoring of Nodes
5	Control and monitoring of Queues
6	Asynchronous event notification to DRMAAv2 application
7	Re-entrancy support for the DRMAAv2 API
8	Support for Autotools build infrastructure
9	Support for RPM package
10	Support for unit testing(White-box) using cppunit framework
11	Support for unit testing(black-box)

Sample Example

Example1: DRMAA2 Simple Application

 	drmaa2_jsession js; /* Job session object*/		
    drmaa2_jtemplate jt; /* Job session template*/
    drmaa2_j j;  /* Job object*/
    ...
    ...
	...
    /* Create and open job session */
    js = drmaa2_create_jsession("testsession", DRMAA2_UNSET_STRING);
    if (js == NULL) {
   		 ...
       return;
    }
    ...
    ...
	...

    /* Create job template */
    jt = drmaa2_jtemplate_create();
    jt->remoteCommand = strdup("/bin/date");
    j = drmaa2_jsession_run_job(js, jt);

    ...
	...
    drmaa2_j_wait_terminated(j, DRMAA2_INFINITE_TIME);
    drmaa2_jtemplate_free(&jt);
    drmaa2_destroy_jsession("testsession");
    ...
    ...
	...
    drmaa2_j_free(&j);
    drmaa2_jsession_free(&js);
    ...
    ...
	...

Example2: DRMAA2 Advance Application

 	...
    ...
	...
    drmaa2_jtemplate        jt = drmaa2_jtemplate_create(); /* job template object */
    drmaa2_rtemplate        rt = drmaa2_rtemplate_create(); /* reservation template object */
    drmaa2_string_list      cl = drmaa2_list_create(DRMAA2_STRINGLIST, NULL); / *create list of strings */
    drmaa2_dict            env = drmaa2_dict_create(NULL); /* create a dictionary */

    drmaa2_jsession js = drmaa2_create_jsession("myjsession", NULL); /* open sessions to DRM system */
    if (js == NULL) {
        ...
        ...
        return;
    }
    drmaa2_rsession rs = drmaa2_create_rsession("myrsession", NULL); /* create a reservation session */
    if (rs == NULL)
    {
        ...
        ...
        return;
    }
    drmaa2_msession ms = drmaa2_open_msession(NULL); 					 /*create and open monitoring session */

    ml = drmaa2_msession_get_all_machines(ms, DRMAA2_UNSET_LIST);        /* determine name of first machine */
    if (drmaa2_list_size(ml) < 1) {
    	...
        return;
    }
    m = (drmaa2_machineinfo)drmaa2_list_get(ml, 0);
    drmaa2_list_add(cl, m->name);

    rt->maxSlots = 4;                                                    /* perform advance reservation */
    if (DRMAA2_TRUE == drmaa2_supports(DRMAA2_RT_MACHINEOS)) 
        rt->machineOS=DRMAA2_LINUX;
    rt->candidateMachines = cl;
    r = drmaa2_rsession_request_reservation(rs, rt);

    jt->remoteCommand = strdup("/bin/date");                             /* submit job */
    jt->reservationId = drmaa2_r_get_id(r);
    drmaa2_dict_set(env, "PBS_SCP", "/usr/bin/scp");
    jt->jobEnvironment = env;
    j = drmaa2_jsession_run_job(js, jt);

    drmaa2_j_wait_terminated(j, DRMAA2_INFINITE_TIME);                  /* Wait for termination and print exit status */
    ji = drmaa2_j_get_info(j);
    ...
	...
	...

    /* close sessions, cleanup */
    drmaa2_jtemplate_free(&jt);
    drmaa2_rtemplate_free(&rt);
    drmaa2_jinfo_free(&ji);
    drmaa2_j_free(&j);
    drmaa2_r_free(&r);

    drmaa2_close_msession(ms);
    drmaa2_close_rsession(rs);
    drmaa2_close_jsession(js);
    drmaa2_msession_free(&ms);
    drmaa2_rsession_free(&rs);
    drmaa2_jsession_free(&js);

    ...
    ...

Architecture and design

Application stack

<<Image>

Internal architecture