Cloud Computing Application in High Energy Physics Yaodong Cheng IHEP, CAS 2012-4-23.

Slides:



Advertisements
Similar presentations
System Center 2012 R2 Overview
Advertisements

Profit from the cloud TM Parallels Dynamic Infrastructure AndOpenStack.
Cloud Computing to Satisfy Peak Capacity Needs Case Study.
CERN IT Department CH-1211 Genève 23 Switzerland t CERN-IT Plans on Virtualization Ian Bird On behalf of IT WLCG Workshop, 9 th July 2010.
Virtualization and the Cloud
M.A.Doman Model for enabling the delivery of computing as a SERVICE.
SPRING 2011 CLOUD COMPUTING Cloud Computing San José State University Computer Architecture (CS 147) Professor Sin-Min Lee Presentation by Vladimir Serdyukov.
1 Bridging Clouds with CernVM: ATLAS/PanDA example Wenjing Wu
Copyright © 2010 Platform Computing Corporation. All Rights Reserved.1 The CERN Cloud Computing Project William Lu, Ph.D. Platform Computing.
Opensource for Cloud Deployments – Risk – Reward – Reality
Cyberaide Virtual Appliance: On-demand Deploying Middleware for Cyberinfrastructure Tobias Kurze, Lizhe Wang, Gregor von Laszewski, Jie Tao, Marcel Kunze,
Cloud Computing for the Enterprise November 18th, This work is licensed under a Creative Commons.
INTRODUCTION TO CLOUD COMPUTING CS 595 LECTURE 7 2/23/2015.
 Cloud computing  Workflow  Workflow lifecycle  Workflow design  Workflow tools : xcp, eucalyptus, open nebula.
1 port BOSS on Wenjing Wu (IHEP-CC)
A Cloud is a type of parallel and distributed system consisting of a collection of inter- connected and virtualized computers that are dynamically provisioned.
Cloud Computing. What is Cloud Computing? Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable.
+ CS 325: CS Hardware and Software Organization and Architecture Cloud Architectures.
Appendix B Planning a Virtualization Strategy for Exchange Server 2010.
Introduction to Cloud Technology StratusLab Tutorial (Orsay, France) 28 November 2012.
BESIII distributed computing and VMDIRAC
M.A.Doman Short video intro Model for enabling the delivery of computing as a SERVICE.
Presented by: Sanketh Beerabbi University of Central Florida COP Cloud Computing.
1 Evolution of OSG to support virtualization and multi-core applications (Perspective of a Condor Guy) Dan Bradley University of Wisconsin Workshop on.
Large Scale Sky Computing Applications with Nimbus Pierre Riteau Université de Rennes 1, IRISA INRIA Rennes – Bretagne Atlantique Rennes, France
Wenjing Wu Andrej Filipčič David Cameron Eric Lancon Claire Adam Bourdarios & others.
Plan  Introduction  What is Cloud Computing?  Why is it called ‘’Cloud Computing’’?  Characteristics of Cloud Computing  Advantages of Cloud Computing.
WNoDeS – Worker Nodes on Demand Service on EMI2 WNoDeS – Worker Nodes on Demand Service on EMI2 Local batch jobs can be run on both real and virtual execution.
Predrag Buncic (CERN/PH-SFT) WP9 - Workshop Summary
Jose Castro Leon CERN – IT/OIS CERN Agile Infrastructure Infrastructure as a Service.
WLCG Overview Board, September 3 rd 2010 P. Mato, P.Buncic Use of multi-core and virtualization technologies.
2012 Objectives for CernVM. PH/SFT Technical Group Meeting CernVM/Subprojects The R&D phase of the project has finished and we continue to work as part.
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Development of e-Science Application Portal on GAP WeiLong Ueng Academia Sinica Grid Computing
Workload management, virtualisation, clouds & multicore Andrew Lahiff.
Web Technologies Lecture 13 Introduction to cloud computing.
CLOUD COMPUTING WHAT IS CLOUD COMPUTING?  Cloud Computing, also known as ‘on-demand computing’, is a kind of Internet-based computing,
OpenNebula: Experience at SZTAKI Peter Kacsuk, Sandor Acs, Mark Gergely, Jozsef Kovacs MTA SZTAKI EGI CF Helsinki.
The CernVM Project A new approach to software distribution Carlos Aguado Jakob Predrag
36 th LHCb Software Week Pere Mato/CERN.  Provide a complete, portable and easy to configure user environment for developing and running LHC data analysis.
Predrag Buncic (CERN/PH-SFT) CernVM Status. CERN, 24/10/ Virtualization R&D (WP9)  The aim of WP9 is to provide a complete, portable and easy.
Breaking the frontiers of the Grid R. Graciani EGI TF 2012.
INTRODUCTION TO GRID & CLOUD COMPUTING U. Jhashuva 1 Asst. Professor Dept. of CSE.
© 2012 Eucalyptus Systems, Inc. Cloud Computing Introduction Eucalyptus Education Services 2.
Claudio Grandi INFN Bologna Virtual Pools for Interactive Analysis and Software Development through an Integrated Cloud Environment Claudio Grandi (INFN.
1 Globe adapted from wikipedia/commons/f/fa/ Globe.svg IDGF-SP International Desktop Grid Federation - Support Project SZTAKI.
DIRAC for Grid and Cloud Dr. Víctor Méndez Muñoz (for DIRAC Project) LHCb Tier 1 Liaison at PIC EGI User Community Board, October 31st, 2013.
Update on Computing/Cloud Marco Destefanis Università degli Studi di Torino 1 BESIII Ferrara, Italy October 21, 2014 Stefano Bagnasco, Flavio Astorino,
Trusted Virtual Machine Images the HEPiX Point of View Tony Cass October 21 st 2011.
Building on virtualization capabilities for ExTENCI Carol Song and Preston Smith Rosen Center for Advanced Computing Purdue University ExTENCI Kickoff.
DIRAC Distributed Computing Services A. Tsaregorodtsev, CPPM-IN2P3-CNRS FCPPL Meeting, 29 March 2013, Nanjing.
St. Petersburg, 2016 Openstack Disk Storage vs Amazon Disk Storage Computing Clusters, Grids and Cloud Erasmus Mundus Master Program in PERCCOM Author:
Predrag Buncic, CERN/PH-SFT The Future of CernVM.
Introduction to Cloud Technology
C Loomis (CNRS/LAL) and V. Floros (GRNET)
Use of HLT farm and Clouds in ALICE
The advances in IHEP Cloud facility
AWS Integration in Distributed Computing
Cloud Challenges C. Loomis (CNRS/LAL) EGI-TF (Amsterdam)
Blueprint of Persistent Infrastructure as a Service
Dag Toppe Larsen UiB/CERN CERN,
Progress on NA61/NA49 software virtualisation Dag Toppe Larsen Wrocław
Dag Toppe Larsen UiB/CERN CERN,
StratusLab Final Periodic Review
StratusLab Final Periodic Review
CernVM Status Report Predrag Buncic (CERN/PH-SFT).
WLCG Collaboration Workshop;
Red Hat User Group June 2014 Marco Berube, Cloud Solutions Architect
Introduction to Cloud Computing
Exploit the massive Volunteer Computing resource for HEP computation
Presentation transcript:

Cloud Computing Application in High Energy Physics Yaodong Cheng IHEP, CAS

2 Cloud and Grid workshop, Joint CHAIN/EPIKH School Outline  From Grid to Cloud  Some cloud projects in HEP  Cloud activities at IHEP, CAS

3 Cloud and Grid workshop, Joint CHAIN/EPIKH School Outline  From Grid to Cloud  Some cloud projects in HEP  Activities at IHEP, CAS

4 Cloud and Grid workshop, Joint CHAIN/EPIKH School Terminology  What is a grid?  A platform for scientific collaboration  A scientific tool to help with data manipulation and processing  A computational platform  And how it compares with a cloud?  A source for computing and storage capacity Flexible, easy to access resource  And a cluster?  A building block for both grids and clouds

5 Cloud and Grid workshop, Joint CHAIN/EPIKH School Grid as a collaborative platform  State before the Grid:  Scientists/teams have resources or access to them  Teams are working independently, they do not share their resources (no technology support)  Data sharing very simply, e.g. secure copy (scp) between teams  State with a Grid:  Team’s resources are connected  Sharing is easy  Scientists could focus on their science and not on the technology behind it

6 Cloud and Grid workshop, Joint CHAIN/EPIKH School Grid View

7 Cloud and Grid workshop, Joint CHAIN/EPIKH School Clouds  A rather recent commercial platform  (Large) Pool of virtualized servers  Users submits not jobs, but full virtual machines with jobs inside them  Targets real-time requirements  Fast deployment of new virtual server  Can quickly react on user’s changing requirements  Standard clouds rather simple  Easy to use web interface  No collaboration support (standard Consumer – Provider model, ideal for commercial use)

8 Cloud and Grid workshop, Joint CHAIN/EPIKH School What is cloud  The NIST definition lists five essential characteristics of cloud computing:  on-demand self-service  broad network access  resource pooling  rapid elasticity or expansion  measured service  Three "service models"  software, platform and infrastructure  Four "deployment models"  private, community, public and hybrid

9 Cloud and Grid workshop, Joint CHAIN/EPIKH School Cloud and Grid: A Comparison Grid Middleware Computing /Data Center Cloud Middleware Computing /Data Center

10 Cloud and Grid workshop, Joint CHAIN/EPIKH School From Grid to Cloud  Grid has been the necessary infrastructure for many scientific research, e.g.. HEP  But, there are still some disadvantages  How to schedule jobs efficiently to improve the resource utilization (vs. static policy)  diversified service model on demand (vs. job submission)  Compatible with legacy programs (vs. unified system environment)  …  Virtualization/Cloud is feasible solution

11 Cloud and Grid workshop, Joint CHAIN/EPIKH School Outline  From Grid to Cloud  Some cloud projects in HEP  Activities at IHEP, CAS

12 Cloud and Grid workshop, Joint CHAIN/EPIKH School CernVM  CernVM is a baseline Virtual Software Appliance for the participants of CERN LHC experiments  Motivation  large, complicated Install/update/configure, …  Multi-core with hardware support for virtualization  Using virtualization and extra cores to get extra comfort zero configuration, reduce compiler-platform combinations  CernVM: Build a “thin” Virtual Software Appliance for use by the LHC experiments  provide a complete, portable and easy to configure user environment  independent of physical software and hardware platforms cernvm.cern.ch/

13 Cloud and Grid workshop, Joint CHAIN/EPIKH School “Thin” Software Appliance JeOS (based on rPath Linux) rAArAA KERNEL fuse module FILESYSTEMFILESYSTEM rAA plugin Extra Libs & Apps Cache HTTPDHTTPD Software Repository 10 GB1 GB0.1 GB LAN/WAN (HTTP)

14 Cloud and Grid workshop, Joint CHAIN/EPIKH School CVMFS: CernVM File System App On same host: /opt/lcg -> /chirp/localhost/opt/lcg open(“/opt/lcg”) On File Server /opt/lcg -> /grow/host/opt/lcg Cache Kernel NFSLFSFUSE CernVM Fuse !Cache

15 Cloud and Grid workshop, Joint CHAIN/EPIKH School Bridging Grids & Clouds  Volunteer Computing  uses computers belonging to ordinary people  BOINC  Open-source software for Volunteer Computing and Grid computing  CernVM is extended to support BOINC client  CernVM CoPilot development  Based on BOINC, experience and CernVM image  Image size is of utmost importance to motivate volunteers  Can be easily adapted to Pilot Job frameworks (AliEn,Dirac, PanDA) BOINC PanDA Pilot

16 Cloud and Grid workshop, Joint CHAIN/EPIKH School CernVM CoPilot Architecture

17 Cloud and Grid workshop, Joint CHAIN/EPIKH School lxcloud  CERN Internal Cloud  Highly scalable, Linux (KVM) based cloud-like infrastructure  Optimized for efficiency/speed

18 Cloud and Grid workshop, Joint CHAIN/EPIKH School Resource Pool details  Quattor managed pool of resources  Hardware: (cheap) CPU server type, local disks  LANDB integration  Pre-allocation of VM “slots” in landb  Hypervisor “knows” the name of guests  Disk management  Use of LVM snapshots  All free disk space in one big LV  Pre-stage raw images on LV on the hypyerviors  Fast installation of VMs Using LV snapshots

19 Cloud and Grid workshop, Joint CHAIN/EPIKH School Image management  Central image catalogue (VMIC)  Close collaboration with HEPiX  No direct user access/user images  Images require endorsement by IT  Image distribution system  Image distribution repository of trusted images  Fast distribution using Bit-torrent (rtorrent)  Pull model: Hypervisors ask if there are updates  Transparent update of images using LV tools  Hypervisors advertise existing images

20 Cloud and Grid workshop, Joint CHAIN/EPIKH School Virtual Machine Management  OpenNebula  an open source Cloud Data Center Management Solution  provides a powerful, scalable and secure multi-tenant cloud platform for fast delivery and elasticity of virtual resources  OpenStack  The Open Source Cloud Operating System  The Main components: Compute, Object Storage, Image service  A interesting product worth to be checked

21 Cloud and Grid workshop, Joint CHAIN/EPIKH School Lxcloud ecosystem lxcloud Physical Resource OpenNebula ONE EC2 Interface ONE 3.0 Master Image creation and endorsement Enduser VO Application manager VM Provisioning Image repository (VMIC) Image repository (VMIC) CernVM Golden Nodes Quattor Image creation

22 Cloud and Grid workshop, Joint CHAIN/EPIKH School Clever: A New VIM  CLEVER: A CLoud-Enabled Virtual EnviRonment  To simplify the access management of private/hybrid clouds  To provide simple and easily accessible interfaces to interact with different “interconnected” clouds, deploy Virtual Machines and perform load balancing through migration

23 Cloud and Grid workshop, Joint CHAIN/EPIKH School Clever on Grid Host1 Host Manager (HM) Host Manager (HM) Ejabberd XMPP Server Ejabberd XMPP Server Sedna Distributed Databases Sedna Distributed Databases job Submission CLEVER.jar and X.509 Certificate User Interface Resource Broker CE HM Worker Nodes jobs Running tiny.vdi Host2 Host Manager (HM) Host Manager (HM) HostN-1 Cluster Manager (CM) Cluster Manager (CM) HostN Host Manager (HM) Host Manager (HM) … Matchmaking and Jobs Scheduling Storage Element Computing Element XQuery/XPath Administration tool XMPP

24 Cloud and Grid workshop, Joint CHAIN/EPIKH School Outline  From Grid to Cloud  Some cloud projects in HEP  Activities at IHEP, CAS

25 Cloud and Grid workshop, Joint CHAIN/EPIKH School Virtual Cluster  Motivation  Build Virtual machine pool on physical machines, elastic to expand or shrink on demand  Flexible to support more kinds of applications  Compatible with legacy programs  R&D cloud for users  Key technologies  hypervisors (KVM, XEN, …) evaluation suitable for HEP  VIM management (OpenNebula, OpenStack, …)  Monitoring and accounting  Interface to PBS, WLCG, and other services  dynamic scheduling Live migration VM resource adjustment (CPU, Memory, Network, …)

26 Cloud and Grid workshop, Joint CHAIN/EPIKH School Architecture of Virtual Server PBS Client Physical Machine VM Physical Machine VM VIM (VM create, start, pause, destroy, migration) Scheduler PBS Server Submit Job Query and Modify Queue Scheduling policy Power Management WLCG Grid Job

27 Cloud and Grid workshop, Joint CHAIN/EPIKH School PBS/Torque integration  Each batch queue has basic resources (physical nodes or Virtual machines)  If the jobs are too many in one queue, the scheduler will create some extra virtual machines according with scheduling policy and requirements, then added the new resources into the queue  The queue with higher priority needs more resources, the VM resources in queues with lower priority will be paused, even destroyed  Fair scheduling is very important here!  WLCG interface is simply via PBS/torque

28 Cloud and Grid workshop, Joint CHAIN/EPIKH School GUI

29 Cloud and Grid workshop, Joint CHAIN/EPIKH School BESIII Cloud  Integrated with Grid, volunteer computing, and virtualization  User submits jobs to BESIII portal, then these jobs will be dispatched to different computing resource  Volunteer computing (small sites and personal computers)  Local cluster (managed by LRMS)  WLCG  CNGrid  plugin framework  gLite, PBS, GOS plugins already completed!  Recently, BESIII Offline Software System (BOSS) has successfully run on CernVM-based  BONIC plugin is ready!

30 Cloud and Grid workshop, Joint CHAIN/EPIKH School  is the first volunteer computing platform in China  Use BOINC as its middleware  Launched by IHEP in January 2010  To help scientists from CAS or other research organizations in China to to run their scientific researches on volunteer computing resources  More than 9,000 user, 16,000 computer joined

31 Cloud and Grid workshop, Joint CHAIN/EPIKH School Architecture of BESIII Cloud BESIII portal BOINC Server BOINC Server PBS Server PBS Server gLite WMS GOS Local ClusterWLCG CNGrid Small sites and Personal Computer Plugins (gLite, GOS, PBS, BOINC, …)

32 Cloud and Grid workshop, Joint CHAIN/EPIKH School Future: Cloud-Grid Integration Cloud + Grid Computing Service Catalog Computing center Infrastructure Virtual Client service Web Application Service Compute Service Database service Storage service Content Classification Storage backup, archive… service Job Scheduling Service Collaboration Services Datacenter Infrastructure

33 Cloud and Grid workshop, Joint CHAIN/EPIKH School Thanks! Questions?