510 likes | 641 Views
Schedulers and Resource Brokers. Scheduler. Job manager submits jobs to scheduler. Scheduler assigns work to resources to achieve specified time requirements. Scheduling. From "Introduction to Grid Computing with Globus," IBM Redbooks. Why scheduling?.
E N D
Schedulers and Resource Brokers Grid Computing, B. Wilkinson, 2004
Scheduler • Job manager submits jobs to scheduler. • Scheduler assigns work to resources to achieve specified time requirements. Grid Computing, B. Wilkinson, 2004
Scheduling From "Introduction to Grid Computing with Globus," IBM Redbooks Grid Computing, B. Wilkinson, 2004
Why scheduling? • Efficient use of Grid resources requires powerful and flexible Grid scheduling • For Grid technology to be successful, there must automatic features to determine available Grid resources and to coordinate the allocation of these resources in accordance with the requirements, dependencies, and objectives of the user. From GGF7-workshop on Grid scheduling Grid Computing, B. Wilkinson, 2004
Scheduling architecture • Is a current area of study for Grid • Eventually, there must be a definition of a scheduling architecture • The cooperation of different scheduling instances for arbitrary resources available in the grid • The interaction of local resource management (e.g., PBS, LoadLeveler) and data management Grid Computing, B. Wilkinson, 2004
Service Level Agreements • Resource SLA, (RSLA) i.e., reservation • A promise that a resource will be available when it is needed • The client will utilize the promise in subsequent SLA’s • Task SLA, (TSLA) i.e., execution • A promise to perform a task • There may be complex task requirements and may reference an RSLA implicitly Grid Computing, B. Wilkinson, 2004
SLA’s, continued • Binding SLA (BSLA), i.e., a claim • Binds a resource capability to a TSLA • May reference an RSLA, or be implicit • May be created lazily to provision the task Grid Computing, B. Wilkinson, 2004
Advance Reservation • Requesting actions at times in future. (“A service level agreement in which the conditions of the agreement start at some agreed-upon time in the future” [2]) [2] “The Grid 2, Blueprint for a New Computing Infrastructure,” I. Foster and C. Kesselman editors, Morgan Kaufmann, 2004. Grid Computing, B. Wilkinson, 2004
Resource Broker • “A scheduler that optimizers the performance of a particular resource. Performance may be measured by such criteria as fairness (to ensure that all requests for the resources are satisfied) or utilization (to measure the amount of the resource used).” [2] Grid Computing, B. Wilkinson, 2004
Community Scheduling • Individual users • Require service • Have application goals • Community schedulers • Broker service • Aggregate scheduling • Individual resources • Provide service • Have policy autonomy • Serve higher-level layers Grid Computing, B. Wilkinson, 2004
Scheduling in Globus (not) • Fully-fledged scheduler/resource broker not in Globus. • For example, Globus does not currently have advance reservation. • Scheduler/resource broker need to be provided separately on top of Globus, using basic services provided in Globus. Grid Computing, B. Wilkinson, 2004
Resource Broker Examples • Condor-G, Nimrod/G, Grid Canada Grid Computing, B. Wilkinson, 2004
Condor • System first developed at University of Wisconsin-Madison in mid 1980’s to convert a collection of distributed workstations and clusters into a high-throughput computing facility. • Key concept - using wasted computer power of idle workstations. Grid Computing, B. Wilkinson, 2004
Condor • Converts collections of distributed workstations and dedicated clusters into a distributed high-throughput computing facility. Grid Computing, B. Wilkinson, 2004
Features • Include: • Resource finder • Batch queue manager • Scheduler • Checkpoint/restart • Process migration Grid Computing, B. Wilkinson, 2004
Intended to run job even if: • Machines crash • Disk space exhausted • Software not installed • Machines are needed by others • Machines are managed by others • Machines are far away Grid Computing, B. Wilkinson, 2004
Uses • Consider following scenario: • I have a simulation that takes two hours to run on my high-end computer • I need to run it 1000 times with slightly different parameters each time. • If I do this on one computer, it will take at least 2000 hours (or about 3 months) From: “Condor: What it is and why you should worry about it,” by B. Beckles, University of Cambridge, Seminar, June 23, ,2004 Grid Computing, B. Wilkinson, 2004
Suppose my department has 100 PCs like mine that are mostly sitting idle overnight (say 8 hours a day) • If I could use them when their legitimate users are not using them, so that I do not inconvenience them, I could get about 800 CPU hours/day. • This is an ideal situation for Condor. • I could do my simulations in 2.5 days. From: “Condor: What it is and why you should worry about it,” by B. Beckles, University of Cambridge, Seminar, June 23, ,2004 Grid Computing, B. Wilkinson, 2004
How does Condor work? • A collection of machines running Condor called a pool. • Individual pools can be joined together in a process called flocking. From: “Condor: What it is and why you should worry about it,” by B. Beckles, University of Cambridge, Seminar, June 23, ,2004 Grid Computing, B. Wilkinson, 2004
Machine Roles • Machines have one or more of four roles: • Central manager • Submit machine (Submit host) • Execution machine (Execute host) • Checkpoint server Grid Computing, B. Wilkinson, 2004
Central Manager • Resource broker for a pool. Keeps track of which machines are available, what jobs are running, negotiates which machine will run which job, etc. • Only one central manager per pool. Grid Computing, B. Wilkinson, 2004
Submit Machine • Machine which submits jobs to pool. • Must be at least one submit machine in a pool, and usually more than one. Grid Computing, B. Wilkinson, 2004
Execute Machine • Machine on which jobs can be run. • Must be at least one execute machine in a pool, and usually more than one. Grid Computing, B. Wilkinson, 2004
Checkpoint Server • Machine which stores al checkpoint files produced by job which checkpoint. • Can only be one checkpoint machine in a pool. • Optional to have a checkpoint machine. Grid Computing, B. Wilkinson, 2004
Possible Configuration • A central manager. • Some machine that can only be submit hosts. • Some machine that can be only execute hosts. • Some machines that can be both submit and execute hosts. Grid Computing, B. Wilkinson, 2004
Submitting a job • Job submitted to submit host • Submit host tells the central ,manager about job using Condors “ClassAd” Mechanism which may include: • What it requires • What it desires • What it prefers, and • What it will accept Grid Computing, B. Wilkinson, 2004
1. Central manager monitoring execute hosts so knows what is available and what type of machines each execute host is, and software. 2. Execute hosts periodically send a ClassAd describing themselves to the central manager. Grid Computing, B. Wilkinson, 2004
3. At times, the central manager enters a negotiation cycle where it matches waiting jobs with available execute hosts. 4. Eventually job is matched with a suitable execute host (hopefully) . Grid Computing, B. Wilkinson, 2004
5. Central manager informs chosen execute host that is has been claimed and gives it a ticket. 6. Central manage informs submit host which execute host to use and gives it a matching ticket. Grid Computing, B. Wilkinson, 2004
7. Submit host contacts execute host presenting its matching ticket and transfers job’s executable and date files to execute host if necessary. (shared file system also possible.) 8. When job finished, results returned to submit host (unless shared file system in use between submit and execute hosts). Grid Computing, B. Wilkinson, 2004
Connections • Connection between submit and execute host usually done with a TCP connection. • If connection dies, job resubmitted to Condor pool. • Some jobs might access files and resources on submit host via remote procedure calls. Grid Computing, B. Wilkinson, 2004
Checkpointing • Certain jobs can checkpoint, both periodically for safety and when interrupted. • If checkpointed job interrupted, it will resume at the last checkpointed state when it starts again. • Generally no change to source code - need to link Condor’s Standard Universe support library (see later). Grid Computing, B. Wilkinson, 2004
Types of Jobs • Classified according to environment it provides. Currently seven environments: • Standard • Vanilla • PVM • MPI • Globus • Java • Scheduler Grid Computing, B. Wilkinson, 2004
Standard • For jobs compiled with Condor libraries • Allows for checking pointing and remote system calls. • Must be single threaded. • Not available under Windows. Grid Computing, B. Wilkinson, 2004
Vanilla • For jobs that cannot be compiled with Condor libraries, and for shell scripts and Windows batch files. • No checkpointing or remote system calls. Grid Computing, B. Wilkinson, 2004
Job Universes continued PVM For PVM programs. MPI For MPI programs (MPICH). Globus For submitting jobs to resources managed by Globus (version 2.2 and higher). Grid Computing, B. Wilkinson, 2004
Java For Java programs (written for Java Virtual Interface). Scheduler A universe not normally used by end-user. Ignores any requirements and runs job on submit host. Never preempted. Grid Computing, B. Wilkinson, 2004
Directed Acyclic Graph Manager (DAGMan) • Allows one to specify dependencies between Condor Jobs. Example “Do not run Job B until Job A completed successfully” Especially important to jobs working together (as in Grid computing). Grid Computing, B. Wilkinson, 2004
Directed Acyclic Graph(DAG) • A data structure used to represent dependencies. • Each job is a node in the DAG. • Each node can have any number of parents and childred as long as there are no loops (Acyclic graph). Grid Computing, B. Wilkinson, 2004
Defining a DAG • DAG defined by a .dag file, listing each of the nodes and their dependencies Example # diamond.dag Job A a.sub Job B b.sub Job C c.sub Job D d.sub Parent A Child B C Parent B C Child D Job A Job B Job C Job D Grid Computing, B. Wilkinson, 2004
Running a DAG • DASGMan acts as a scheduler managing the submission of jobs to Condor based upon DAG dependencies. • DAGMan holds and submits jobs to Condor queue at appropriate times. Grid Computing, B. Wilkinson, 2004
Job Failures • DAGMan continues until it cannot make progress and then creates a rescue file holding current state of DAG. • When failed job ready to re-run, rescue file used to restore prior state of DAG. Grid Computing, B. Wilkinson, 2004
ClassAd Matchmaking Used to ensure job done according to constraints of users and owners. Example of user constraints “ I need a Pentium IV with at least 512 Mbytes of RAM and speed of at least 3.5 Ghz Example of machine owner constraints “Never run jobs owned by Fred” Grid Computing, B. Wilkinson, 2004
Condor Submit Description File Describes job to Condor. Used with Condor _submit command. Description File Example # This is a comment, condor submit file Universe = vanilla Executable = /home/abw/condor/myProg Input = myProg.stdin Output = myProg.stdout Error = myProg.stderr Arguments = -arg1 -arg2 InitialDir = /home/abw/condor/assignment4 Queue Grid Computing, B. Wilkinson, 2004
Submitting Multiple Jobs • Submit file can specify multiple jobs Queue 500 will submit 500 jobs at once • Condor calls groups of jobs a cluster • Each job within cluster called a process • Condor job ID is the cluster number, a period and process number, for example 26.2 • Single jobs also a cluster but with a single process (process 0) Grid Computing, B. Wilkinson, 2004
Specifying Requirements • A C/Java-like Boolean expression that evaluates to TRUE for a match. # This is a comment, condor submit file Universe = vanilla Executable = /home/abw/condor/myProg InitialDir = /home/abw/condor/assignment4 Requirements = Memory >= 512 && Disk > 10000 queue 500 Grid Computing, B. Wilkinson, 2004
Summary of Key Condor Features • High throughput computing using an opportunitistic environment. • Matchmaking • Checkpointing • DAG scheduling Grid Computing, B. Wilkinson, 2004
Condor-G • Grid enabled version of Condor. • Uses Globus Toolkit for: • Security (GSI) • managing remote jobs on grid (GRAM) • file handling and remote I/O (GSI-FTP) Grid Computing, B. Wilkinson, 2004
Remote execution by Condor-G on Globus-managed resources From:”Condor-G A Computation Management Agent for Multi-Institutional Grids” by J. Frey, T. Tannenbaum, M. Livny, I. Foster and S. Tuecke. Figure probably refers to Globus version 2. Grid Computing, B. Wilkinson, 2004