die kooperation von forschungszentrum karlsruhe gmbh und universität karlsruhe (th) the cirrus...

Die Kooperation von Forschungszentrum Karlsruhe GmbH

und Universität Karlsruhe (TH)

The Cirrus & Cumulus Project: Build a Scientific Cloud for a Data CenterLizhe Wang and Marcel KunzeSteinbuch Centre for Computing (SCC)Karlsruhe Institute of Technology (KIT) Germany

KIT - Die Kooperation von Forschungszentrum Karlsruhe GmbH


Karlsruhe Institute of Technology (KIT)

Cooperation between Research Centre Karlsruhe und Karlsruhe University

Largest scientific center in Germany

8.000 scientists, 18.000 students

Annual budget: > 500 Million Euro

R&D focus: Energy research and Nanotechnology

> +


und Universität Karlsruhe (TH)3 | Autor | Steinbuch Centre for Computing | 06.09.2007

Google Trends: Grid Computing vs. Cloud Computing



OpenCirrus Cloud Computing Research Testbedhttp://cloudtestbed.com

An open, internet-scale global testbed for cloud computing research

a tool for collaborative research focus: data center management & cloud services

Resources: Multi-continent, multi-datacenter, cloud computing system “Centers of Excellence” around the globe

each with 100–400+ nodes and up to ~2PB storage and running a suite of cloud services

Structure: a loose federation Sponsors: HP Labs, Intel Research, Yahoo! Partners: UIUC, Singapore IDA, KIT, NSF Members: System and application development

Available towards end of the year Great opportunity for Cloud R&D

4 April 10, 2023 © Hewlett-Packard Company



Physical and Virtual Resource Sets (PRS, VRS)

experiment save/restore



A possible Definition of Cloud Computing

Definition A computing Cloud is a set of

network enabled services, providing scalable, QoS guaranteed, inexpensive computing platforms on demand, which could be accessed in a simple and pervasive way.

Functionalities SaaS: Software as a Service HaaS: Hardware as a Service DaaS: Data as a Service PaaS: Platform as a Service IaaS: Infrastructure as a Service



Anatomy of Cloud Computing

Key features User-centric access On-demand service provisioning QoS guaranteed offer Autonomy Scalability & flexibility

Enabling technologies Virtualization Service flow/workflow orchestration Web service & SOA Web 2.0 World-wide distributed storage & file system Parallel & distributed programming model



Clouds vs. Grids: A Comparison

Cloud Computing Grid Computing

Objective Provide desired computing platform via network enabled services

Resource sharing

Job execution

Infrastructure One or few data centers, heterogeneous/homogeneous resource under central control,

Industry and Business

Geographically distributed, heterogeneous resource, no central control, VO

Research and academic organization

Middleware Proprietary, several reference implementations exist (e.g. Amazon)

Well developed, maintained and documented

Application Suited for generic applications Special application domains like High Energy Physics

User interface Easy to use/deploy, no complex user interface required

Difficult use and deployment

Need new user interface, e.g., commands, APIs, SDKs, services …

Business Model Commercial: Pay-as-you-go Publicly funded: Use for free

Enabling technology Virtualization, SaaS, Web 2.0, Web service, …

HPC, Grid infrastructure, middleware, …

QoS Possible Little support

On-demand provisioning Yes No



Cumulus: A Cloud Computing Prototype



Re-engineering the Globus Virtual Workspace Service (GVWS)

GVWS Limitations Force all backends to install GVWS control agents Cloud users need to select network solutions Cloud users need to prepare VM images

Re-engineering the GVWS for the Cumulus Remove GVWS control agents, GVWS frontend talks directly to

LVMS, e.g., OpenNEbula Consider the similar scenario: Globus + PBS

The “forward” network requirements to LVMS Prepare VM image via OS Farm for users



OpenNEbula as a Backend

The Cumulus frontend communicates with OpenNEbula via: SSH XML-RPC

Improvement NIS LDAP: shared user management - oneadmin NFS OCFS (Oracle Cluster File System):

shared volumes for VM images

Can develop other plug-in, embedded in the Cumulus frontend, for other LVMS, e.g., VMware Virtual Infrastructure



OS Farm as a VM Provisioning Tool

Development of CERN OpenLab

OS Farm as server to generate VM images

Can accept HTTP requirements, or via wget:

wget http://www.fzk.de/osfarm/create?name=&transfer=http&

class=slc_old&arch=i386&filetype=.tar&group=core&group=base

A Java client is embedded in the Cumulus frontend to invoke OS Farm service dynamically.



The “forward” Network Solution

Users do not need to specify the network configuration of a VM

Users only get network access to VM, e.g., IP address or hostname + domain name

Cumulus forwards the network requirements to backend (LVMS), e.g., OpenNEbula

Network solutions on Cumulus backend: OpenNEbula starts a VM, listens to central DHCP server, arranges

an IP address to VM, returns it to Cumulus frontend Set hostname when the virtual machine is created Lease an IP address from the IP pool

Cumulus frontend returns the IP address or the host name to the client.



Thank you! Question?

Contact: Lizhe Wang, Marcel KunzeSteinbuch Centre for Computing (SCC) Karlsruhe Institute of Technology (KIT) Hermann-von-Helmholtz-Platz 1 D-76344 Eggenstein-Leopoldshafen, Germany [email protected]@kit.edu

die kooperation von forschungszentrum karlsruhe gmbh und universität karlsruhe (th) the cirrus...

Documents

computing cloud

cloud computing slide

die kooperation

service slide

scientific cloud

cloud rd

network solutions cloud

service haas