Douglas Thain, John Bent Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, Miron Livny Computer Sciences Department, UW-Madison Gathering at the Well: Creating.

Slides:



Advertisements
Similar presentations
Jaime Frey Computer Sciences Department University of Wisconsin-Madison OGF 19 Condor Software Forum Routing.
Advertisements

Dan Bradley Computer Sciences Department University of Wisconsin-Madison Schedd On The Side.
CPU Scheduling Questions answered in this lecture: What is scheduling vs. allocation? What is preemptive vs. non-preemptive scheduling? What are FCFS,
High Performance Computing Course Notes Grid Computing.
Introduction CSCI 444/544 Operating Systems Fall 2008.
Condor-G: A Computation Management Agent for Multi-Institutional Grids James Frey, Todd Tannenbaum, Miron Livny, Ian Foster, Steven Tuecke Reporter: Fu-Jiun.
Workload Management Workpackage Massimo Sgaravatto INFN Padova.
Introduction and Overview “the grid” – a proposed distributed computing infrastructure for advanced science and engineering. Purpose: grid concept is motivated.
1 Draft of a Matchmaking Service Chuang liu. 2 Matchmaking Service Matchmaking Service is a service to help service providers to advertising their service.
Workload Management Massimo Sgaravatto INFN Padova.
The Condor Data Access Framework GridFTP / NeST Day 31 July 2001 Douglas Thain.
Efficiently Sharing Common Data HTCondor Week 2015 Zach Miller Center for High Throughput Computing Department of Computer Sciences.
The Difficulties of Distributed Data Douglas Thain Condor Project University of Wisconsin
Virtual Network Servers. What is a Server? 1. A software application that provides a specific one or more services to other computers  Example: Apache.
Jim Basney Computer Sciences Department University of Wisconsin-Madison Managing Network Resources in.
The SAM-Grid Fabric Services Gabriele Garzoglio (for the SAM-Grid team) Computing Division Fermilab.
Miron Livny Computer Sciences Department University of Wisconsin-Madison Harnessing the Capacity of Computational.
Alain Roy Computer Sciences Department University of Wisconsin-Madison An Introduction To Condor International.
Distributed Systems Early Examples. Projects NOW – a Network Of Workstations University of California, Berkely Terminated about 1997 after demonstrating.
Module 3: Business Information Systems Chapter 11: Knowledge Management.
History of the National INFN Pool P. Mazzanti, F. Semeria INFN – Bologna (Italy) European Condor Week 2006 Milan, 29-Jun-2006.
Gilbert Thomas Grid Computing & Sun Grid Engine “Basic Concepts”
Douglas Thain, John Bent, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, and Miron Livny WiND and Condor Projects 6 May 2003 Pipeline and Batch Sharing in.
Introduction and Overview Questions answered in this lecture: What is an operating system? How have operating systems evolved? Why study operating systems?
 What is OS? What is OS?  What OS does? What OS does?  Structure of Operating System: Structure of Operating System:  Evolution of OS Evolution of.
Grid Data Management A network of computers forming prototype grids currently operate across Britain and the rest of the world, working on the data challenges.
Workload Management WP Status and next steps Massimo Sgaravatto INFN Padova.
Networked Storage Technologies Douglas Thain University of Wisconsin GriPhyN NSF Project Review January 2003 Chicago.
Profiling Grid Data Transfer Protocols and Servers George Kola, Tevfik Kosar and Miron Livny University of Wisconsin-Madison USA.
ARGONNE  CHICAGO Ian Foster Discussion Points l Maintaining the right balance between research and development l Maintaining focus vs. accepting broader.
The Data Grid: Towards an Architecture for the Distributed Management and Analysis of Large Scientific Dataset Caitlin Minteer & Kelly Clynes.
1 Performance Evaluation of Computer Systems and Networks Introduction, Outlines, Class Policy Instructor: A. Ghasemi Many thanks to Dr. Behzad Akbari.
Y. Kotani · F. Ino · K. Hagihara Springer Science + Business Media B.V Reporter: 李長霖.
Grid Workload Management & Condor Massimo Sgaravatto INFN Padova.
Miron Livny Computer Sciences Department University of Wisconsin-Madison Welcome and Condor Project Overview.
Introduction to dCache Zhenping (Jane) Liu ATLAS Computing Facility, Physics Department Brookhaven National Lab 09/12 – 09/13, 2005 USATLAS Tier-1 & Tier-2.
Virtual Data Grid Architecture Ewa Deelman, Ian Foster, Carl Kesselman, Miron Livny.
ROOT and Federated Data Stores What Features We Would Like Fons Rademakers CERN CC-IN2P3, Nov, 2011, Lyon, France.
Tevfik Kosar Computer Sciences Department University of Wisconsin-Madison Managing and Scheduling Data.
Flexibility, Manageability and Performance in a Grid Storage Appliance John Bent, Venkateshwaran Venkataramani, Nick Leroy, Alain Roy, Joseph Stanley,
Storage Research Meets The Grid Remzi Arpaci-Dusseau.
AN SLA-BASED RESOURCE VIRTUALIZATION APPROACH FOR ON-DEMAND SERVICE PROVISION Gabor Kecskemeti MTA SZTAKI International Workshop on Virtualization Technologies.
CEDPS Data Services Ann Chervenak USC Information Sciences Institute.
6/23/2005 R. GARDNER OSG Baseline Services 1 OSG Baseline Services In my talk I’d like to discuss two questions:  What capabilities are we aiming for.
Peter Couvares Associate Researcher, Condor Team Computer Sciences Department University of Wisconsin-Madison
Nick LeRoy Computer Sciences Department University of Wisconsin-Madison Hawkeye.
A Fully Automated Fault- tolerant System for Distributed Video Processing and Off­site Replication George Kola, Tevfik Kosar and Miron Livny University.
Globus and PlanetLab Resource Management Solutions Compared M. Ripeanu, M. Bowman, J. Chase, I. Foster, M. Milenkovic Presented by Dionysis Logothetis.
Super Computing 2000 DOE SCIENCE ON THE GRID Storage Resource Management For the Earth Science Grid Scientific Data Management Research Group NERSC, LBNL.
6 march Building the INFN Grid Proposal outline a.ghiselli,l.luminari,m.sgaravatto,c.vistoli INFN Grid meeting, milano.
Miron Livny Computer Sciences Department University of Wisconsin-Madison Condor and (the) Grid (one of.
Managing Network Resources in Condor Jim Basney Computer Sciences Department University of Wisconsin-Madison
Douglas Thain, John Bent Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, Miron Livny Computer Sciences Department, UW-Madison Gathering at the Well: Creating.
April 25, 2006Parag Mhashilkar, Fermilab1 Resource Selection in OSG & SAM-On-The-Fly Parag Mhashilkar Fermi National Accelerator Laboratory Condor Week.
NeST: Network Storage John Bent, Venkateshwaran V Miron Livny, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau.
George Kola Computer Sciences Department University of Wisconsin-Madison Data Pipelines: Real Life Fully.
Capacity Planning in a Virtual Environment Chris Chesley, Sr. Systems Engineer
Condor on Dedicated Clusters Peter Couvares and Derek Wright Computer Sciences Department University of Wisconsin-Madison
Condor Week May 2012No user requirements1 Condor Week 2012 An argument for moving the requirements out of user hands - The CMS experience presented.
Workload Management Workpackage
Condor A New PACI Partner Opportunity Miron Livny
Condor: Job Management
Ákos Frohner EGEE'08 September 2008
Class project by Piyush Ranjan Satapathy & Van Lepham
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Basic Grid Projects – Condor (Part I)
NeST: Network Storage Technologies
Cloud Computing Architecture
GLOW A Campus Grid within OSG
L. Glimcher, R. Jin, G. Agrawal Presented by: Leo Glimcher
Presentation transcript:

Douglas Thain, John Bent Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, Miron Livny Computer Sciences Department, UW-Madison Gathering at the Well: Creating Communities for Grid I/O

New framework needed › Remote I/O is possible anywhere › Build notion of locality into system? › What are possibilities?  Move job to data  Move data to job  Allow job to access data remotely › Need framework to expose these policies

Key elements › Storage appliance, interposition agents, schedulers and match-makers › Mechanism not policies › Policies are exposed to an upper layer  We will however demonstrate the strength of this mechanism

To infinity and beyond › Speedups of 2.5x possible when we are able to use locality intelligently › This will continue to be important  Data sets are getting larger and larger  There will always be bottlenecks

Outline › Motivation › Components › Expressing locality › Experiments › Conclusion

I/O communities › Mechanism which allow either  jobs to move to data, or  data to move to jobs, or  data to be accessed remotely › Framework to evaluate these policies

Grocers, butchers, cops › Members of an I/O community  Storage appliances  Interposition agents  Scheduling systems  Discovery systems  Match-makers  Collection of CPU’s

Storage appliances › Should run without special privilege  Flexible and easily deployable  Acceptable to nervous sys admins › Should allow multiple access modes  Low latency local accesses  High bandwidth remote puts and gets

NeST Common protocol layer GFTPChirpHTTPFTP Dispatcher Storage Manager Physical storage layer Multiple concurrencies Transfer Manager Control flow Data flow

Interposition agents › Thin software layer interposed between application and OS › Allow applications to transparently interact with storage appliances › Unmodified programs can run in grid environment

PFS: Pluggable File System

Scheduling systems and discovery › Top level scheduler needs ability to discover diverse resources › CPU discovery  Where can a job run? › Device discovery  Where is my local storage appliance? › Replica discovery  Where can I find my data?

Match-making › Match-making is the glue which brings discovery systems together › Allows participants to indirectly identify each other  i.e. can locate resources without explicitly naming them

Condor and ClassAds

Outline › Motivation › Components › Expressing locality › Experiments › Conclusion

I/O Communities UW INFN

Two I/O communities › INFN Condor pool  236 machines, about 30 available at any one time  Wide range of machines and networks spread across Italy  Storage appliance in Bologna 750 MIPS, 378 MB RAM

Two I/O communities › UW Condor pool  ~900 machines, 100 dedicated for us  Each is 600 MIPS, 512 MB RAM  Networked on 100 Mb/s switch  One was used as a storage appliance

Who Am I This Time? › We assumed the role of an Italian scientist › Database stored in Bologna › Need to run 300 instances of simulator

Hmmm…

Three way matching Machine NeST Job Ad Machine Ad Storage Ad match Refers to NearestStorage. Knows where NearestStorage is.

Two way ClassAds Type = “job” TargetType = “machine” Cmd = “sim.exe” Owner = “thain” Requirements = (OpSys==“linux”) Job ClassAd Type = “machine” TargetType = “job” OpSys = “linux” Requirements = (Owner==“thain”) Machine ClassAd

Three way ClassAds Type = “job” TargetType = “machine” Cmd = “sim.exe” Owner = “thain” Requirements = (OpSys==“linux”) && NearestStorage.HasCMSData Job ClassAd Type = “machine” TargetType = “job” OpSys = “linux” Requirements = (Owner==“thain”) NearestStorage = ( Name = “turkey”) && (Type==“Storage”) Machine ClassAd Type = “storage” Name = “turkey.cs.wisc.edu” HasCMSData = true CMSDataPath = /cmsdata” Storage ClassAd

Outline › Motivation › Components › Expressing locality › Experiments › Conclusion

BOOM!

CMS simulator sample run › Purposefully choose a run with high I/O to CPU ratio › Accesses about 20 MB of data from a 300 MB database › Writes about 1 MB of output › ~160 seconds execution time  on a 600 MIPS machine with local disk

Policy specification › Run only with locality  Requirements = (NearestStorage.HasCMSData) › Run in only one particular community  Requirements = (NearestStorage.Name == “nestore.bologna”) › Prefer home community first  Requirements = (NearestStorage.HasCMSData)  Rank = (NearestStorage.Name == “nestore.bologna” ) ? 10 : 0 › Arbitrarily complex  Requirements = ( NearestStorage.Name == “nestore.bologna”) || ( ClockHour 18 )

Policies evaluated › INFN local › UW remote › UW stage first › UW local (pre-staged) › INFN local, UW remote › INFN local, UW stage › INFN local, UW local

Completion Time

CPU Efficiency

Conclusions › I/O communities expose locality policies › Users can increase throughput › Owners can maximize resource utilization

Future work › Automation  Configuration of communities  Dynamically adjust size as load dictates › Automation  Selection of movement policy › Automation

For more info › Condor  › ClassAds  › PFS  › NeST 

Local only

Remote only

Both local and remote

I/O communities are an old idea, right? › File servers and administrative domains › No, not really. We need  more flexible boundaries  simple mechanism by which users can express I/O community relationships  hooks into system that allow users to use locality

Grid applications have demanding I/O needs › Petabytes of data in tape repositories › Scheduling systems have demonstrated that there are idle CPUs › Some systems  move jobs to data  move data to jobs  allow job remote access to data › No one approach is always “best”

Easy come, easy go › In a computation grid, resources are very dynamic › Programs need rich methods for finding and claiming resources  CPU discovery  Device discovery  Replica discovery

Bringing it all together CPU Discovery System Replica Discovery System Device Discovery System JobAgent Execution site Storage appliance Distributed Repository Short-haul I/O Long-haul I/O

Conclusions › Locality is good › Balance point between staging data and accessing it remotely is not static  depends on specific attributes of the job data size, expected degree of re-reference, etc  depends on performance metric CPU efficiency or job completion time

Implementation › NeST  storage appliance › Pluggable File System (PFS)  interposition agent built with Bypass › Condor and ClassAds  scheduling system  discovery system  match-maker

Jim Gast and Bart say... › Too many bullet slides › Contributions  scientist doesn’t want to name bec resources are dynamic and name is irrelevant  hooks into system to allow users to express and take advantage of locality

Jim Gast and Bart say...  everyone knows locality is good - but there is no way to express this and run jobs on the grid  I/O communities are mechanism by which user can use locality and specify policies to optimize job performance

4 earth-shattering revelations 1) The grid is big. 2) Scientific data-sets are large. 3) Idle resources are available. 4) Locality is good.

Mechanisms not policies › I/O communities are a mechanism not a policy › A higher layer is expected to choose application appropriate policies › We will however demonstrate the strength of the mechanism by defining appropriate policies for one particular application

Experimental results › Implementation › Environment › Application › Measurements › Evaluation