Your SlideShare is downloading. ×
2012.05.11 - IBM SmartCloud Provisioning by Rossella de Gaetano - Forum du Club Cloud des Partenaires
Upcoming SlideShare
Loading in...5

Thanks for flagging this SlideShare!

Oops! An error has occurred.

Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

2012.05.11 - IBM SmartCloud Provisioning by Rossella de Gaetano - Forum du Club Cloud des Partenaires


Published on

Support exploité par Rossella de Gaetano lors de l'atelier SmartCloud Provisioning - 9ème Forum du Club Cloud des Partenaires

Support exploité par Rossella de Gaetano lors de l'atelier SmartCloud Provisioning - 9ème Forum du Club Cloud des Partenaires

Published in: Technology
1 Like
  • Be the first to comment

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

No notes for slide


  • 1. May 2012Build a low-touch, highly scalable cloud withIBM SmartCloud ProvisioningRossella De Gaetano : © 2012 IBM Corporation
  • 2. Agenda IBM SmartCloud Provisioning overview Case studies IBM SmartCloud Monitoring integration with IBM SmartCloud Provisioning How to position IBM SmartCloud Provisioning Demo Upcoming features2 © 2012 IBM Corporation
  • 3. Companies need an affordable entry point for cloudOrganizations need a low-cost, low-risk entry point to cloud computing technologywith compelling features that achieves their business goals. Customer goals: • Provide agile service delivery that caters to changing business needs. • We know virtualization helps reduce capital expense but how do I control my operational cost? • Need to increase the performance, utilization, reliability and scalability of IT resources while reducing IT complexity • Ability to expand capabilities VISIBILITY CONTROL AUTOMATION See services in real time & Better manage assets, Achieve greater efficiency better predict business service & compliance. and service quality outcomes3 © 2012 IBM Corporation
  • 4. IBM SmartCloud Provisioning is a true Infrastructure-as-a-Service cloud, reducingcost and providing a highly scalable, rapid-deployment environment with near-zero downtime and automated recovery across heterogeneous platforms.Key Benefits: Key Differentiators: Distributed architecture for solution resilience.  Hypervisor agnostic supporting KVM, ESX, Xen Rapid scalable deployment designed to deliver  Reduced hypervisor licensing by accessing the near-instant deployment of 100s of virtual hypervisor directly without going through the machines in seconds instead of mins or hours. licensed (and costly) management components. Continuous operations during upgrades and  Hardware agnostic enabling choice of maintenance resulting in no outages or supporting your current hardware. downtime.  Advanced Image lifecycle management & image Reliable, non-stop cloud capable of composition tooling. automatically tolerating and recovering from  Intelligent load balancing during provisioning. software and hardware failures. Save IT labor resources at scale by enabling self-  Open source based providing and easy service request and highly automated operations extensible platform utilizing existing. Reduce complexity through ease of use and  Small footprint of code with core components improve time to value. for the Cloud management less than 200Mb.4 © 2012 IBM Corporation
  • 5. Quickly get started with IBM SmartCloud Provisioning• Quickly stand up a cloud • Start small and scale based on need • No additional pre-reqs such as databases, app Provisioning servers, messaging middleware • Freedom of choice for hypervisors • Avoid expensive vendor lock in • VMWare ESX, KVM, Xen Compute node Compute node• Highly cost effective solution • Requires no additional hypervisor management tools • Requires no HA hardware or software First Box• Rich set of interfaces into the cloud • Web Interface, scripting and web services Storage node • All function can be driven by a user or by scripts Storage node for complete automation • Out-of-the-box and running in less than 4 hours • Get started with only 4 servers5 © 2012 IBM Corporation
  • 6. Provides Unparalleled Scalability, Speed and Fault Tolerance• It’s Fast • Can start 100 VMs in under 3 mins • Can start a single VM and load OS in under 10 seconds • Can go from bare metal to ready for work in under 5 minutes• It Scales up to and beyond 50,000 VMs in an hour (50 nodes) • Add capacity by simply plugging in a blade or server • Writes only the data you change • Peer-peer architecture to avoid traditional bottlenecks• Its Fault-Tolerant Provisioning • “Live Update” capability to patch or upgrade the Cloud • No single point of failure • Automatic failure recovery Bo Bo t VM Bo Mgt VM t Hypervisor Hypervisor t VM Bo Hypervisor Hardware t VM Bo Hardware VM Bo t Hardware t Hypervisor Hypervisor Bo Hypervisor Hardware VM Bo Hardware t Bo t HardwareRequested VMs will be up Hypervisor Hypervisor Hypervisor tand running under a minute Hardware Hardware Hardwareusing standard HW6 © 2012 IBM Corporation
  • 7. IBM SmartCloud Provisioning key concepts..• IBM SmartCloud Provisioning coordinates three indipendent ingredients: • Virtual servers • Network addresses • Storage volumes When a request is made, a given virtual server is attached to one or more network addresses and one or more storage volumes• This has several advantages: • Quickly recover from failures – cheaper to replace than to fix: - Swap out old VM and replace with fresh new VM on different hardware  Dramatically simplifies patch, image and change management: - New versions can be easily swapped in, and old versions can be easily put back if problems arise - Servers can be restarted and then reattached to their previous storage volumes and addresses to quickly restore their state7 © 2012 IBM Corporation
  • 8. IBM SmartCloud Provisioning vs traditional hierarchical architecture VV M VV MV MMM VV M VV M V MMM VV M VV M V MMM VV MVV MVMMM V End Users V M VV M V End Users MMM VV M VV M V MMM VV VV V V V V V V V V MVV MV M VV MV MV MVV M VV MVV M M MV MV M MV M MV Server machines Storage devices M M M M M M MM M M M M M (iSCSI) No single points of failure, no bottlenecks in data Failures that will impact your users, slowdowns that yourserving/processing, no intervention to repair broken parts! users will notice, and extra work for the admin team! Multiple, load balanced instances of all services  Single instance of critical services Parallel processing against storage  Serialized processing Self-adapting peer-to-peer communication & coordination  No automatic restart and rerouting of requests Recovery oriented computing  Patches and upgrades go everywhere Network deployed software & image updating  Requires very high cost hardware Distributed request processing, data storage and messaging  The bigger the cloud the worse the damage Designed to run on lower cost hardware and storage devices  Request for images bound to a single location All services monitor and restart each other, and dynamically  Serialized, slow access and susceptible to peak overloadselect the leader  Patches/upgrades require taking down the Cloud to apply and they Base software is loaded via network boot (PXE) need to be carefully applied Services are images, so update themselves by restarting with  When a service breaks, that part of the system is outnew image version  Admins have to troubleshoot and fix8 © 2012 IBM Corporation
  • 9. Failure mode and automatic recovery 1. Leader fails • triggers reelection and another takes over 2. Agent fails • watchdog restarts Standard IaaS Interface 3. Storage node fails • Multi-path continues uninterrupted • Agent connects to another replica V V V V M M M M 0 0 0 0 • Leader tries to restart V V V V M M M M VM3 VM2 VM1 Dom0 ∆‘s ∆‘s ∆‘ s 0 0 0 0 ∆‘s ∆‘ iSCSI s ∆‘s iSCSI Hypervisor Hypervisor Compute 4. Compute node failure Dynamic P2P Nodes • Partial faults cause agent to stop accepting requests Management network overlay V V V V Agent M M M M 0 0 0 0 to evacuate iSCSI ∆‘ iSCSI ∆‘s ∆‘ s Hy isor perv s • Once empty will restart (PXE fresh image) ∆‘s ∆‘s ∆‘s V M 0 V M 0 V M 0 V M 0 • Full failure detected by leader which stops sending V V V V M M M M V V V V 0 0 0 0 M M M M requests iSCSI ∆‘s ∆‘ ∆‘s Hy isor perv s iSCSI ∆‘ s ∆‘ s ∆‘ s 0 0 0 0 Hypervisor Hypervisor iSCSI ∆‘s ∆‘ ∆‘s s • Leader cycle’s server power (PXE fresh image) Hy isor perv Image streaming V V V V M M M M 0 0 0 0 5. Service VM fails ∆‘ iSCSI ∆‘ s s ∆‘s Hy isor perv • All services are clustered to tolerate individual faults • Leader VM detects and replaces failed VM Storage Nodes 6. Network fault9 • Redundant interfaces automatically take over © 2012 IBM Corporation
  • 10. Disaster recovery & volumes high availability Persistent volumes needs to be manually backed-up and made highly available: – Integrate with Tivoli Storage Manager or any other backup/recovery tool – Use file systems on a SAN and mount them on demand on the appropriate storage node Disaster recovery can be easily achieved using two VM regions – Data to be synchronized is the content of the storage nodes10 © 2012 IBM Corporation
  • 11. Usage and reporting Usage Data is stored in HBASE table Realtime data can be extracted via CLI (iaas-describe-resources-inuse-by- access) Metering information can be generated using cronjobs11 © 2012 IBM Corporation
  • 12. Nice to knowLeverage Open Source software: • HBase™: a scalable, distributed database that supports structured data storage for large tables • ZooKeeper™: A high-performance coordination service for distributed applications • OpenLDAP™ Software: an open source implementation of the Lightweight Directory Access Protocol • Dojo Toolkit: an open source modular JavaScript library • HTMLSmall footprint … less than 200MB of which 80MB for Java™ libraryOffer a variety of interfaces: UI, command line, REST APIsIBM Infrastructure as a Service APIs submitted to DMTF (Desktop Management Task Force), also used by IBM SmartCloud Enterprise12 © 2012 IBM Corporation
  • 13. What’s the problem with Virtualization and Standardization? 1 In the beginning, there was the perfect image… 2 Then users starting making changes and “snapshots”… …and what they put in the images is unknown… 3 Then they get copied to multiple locations… …and some change again… 4 Then you need to apply a critical security patch…how? … where?13 © 2012 IBM Corporation
  • 14. IBM SmartCloud Provisioning Puts You Back in Control Control over Image Versions, Content and Locations – Image library allows check in, check out, and tracking of versions in the environment •Changes can finally be tracked – Powerful image analytics finds the content you need and can show the specific changes from one image to another •Encourages reuse and gives needed visibility to analyzing whole systems at a glance – Images are tracked across multiple Clouds and/or multiple sites •Critical for disaster recovery arrangements and decentralized use Control over Image Construction – Build images using Company-certified OS, middleware, and application packages •Avoids images being taken out of production due to non-compliance14 © 2012 IBM Corporation
  • 15. Two optionally installable components: •Image Construction and Composition Tool •Import images from etherogeneous providers •Extend images Avoid images being taken out of production due to non-compliance •Image library •Import & export images across an etherogeneous set of repositories •Assign version to images •Search for images •Compare images15 Get full control on image lifecycle © 2012 IBM Corporation
  • 16. Use cases: from the setup to the use Manage End user tasksCompare Index Check-in Image Administrative tasks Images Images Images Versioning Image management Federate Start/Stop Search Import Check-out Allocate Create Image Deployed Images Images Images vLANs Volumes Repositories VMs Register Build Register Deploy Attach Install IPs Images Images VMs VolumesRegister Create Modify Capture Allocate Terminate Monitor vLANs Users/Groups Images Images IPs VMs User workload Assign Monitor Associate Delete Usage Quotas cloud IPs images16 © 2012 IBM Corporation
  • 17. Use Case: Rapid deployment & low touch resource additionUse Case: User driven rapid deployment offew or 100’s of vm’s across heterogeneousplatforms and hypervisorsIT Admin needs to provision few to hundreds ofVMs in minutes across multiple hypervisors andplatforms in a fault-tolerant environment . ITAdmin requests deployment of an image acrossseveral VM instances. Deploy fails due to lack ofresources. New blade is powered on and allowedPXE boot to our infrastructure. In less than 6minutes with no manual configuration, resource isavailable in cloud to be consumed. IT Adminresubmits provisioning job, Image is selected fromthe image library and provisioned across rackswith each node provisioning VMs. A hardwarefailure occurs and a blade is removed. The work isautomatically balanced across remaining blades Customer Valuewith no disruptions in running the workloads. Power up and forget: Scale up cloud infrastructure with no manual configuration and rapid crossChallenge(s) this Scenario Addresses: domain provisioningScaling up cloud resources  Choice through extensive hypervisor and platform support17 © 2012 IBM Corporation 17
  • 18. Use Case : Build and manage a fault tolerant cloudUse Case: Deploy highly available cloudinfrastructure that automatically tolerates HW andSW failures and scales with no manualconfigurationsCloud Admin sets up a infrastructure-as-a-servicecloud which allows cloud users to deploy applicationsin the cloud. He is responsible to mange andmaintain availability for this infrastructure so thecloud has high reliable and available. To achieve thishe will use the OOTB fault tolerance infrastructureprovided by IBM SmartCloud Provisioning. He makessure depending on number of Compute nodes andstorage nodes in his environment he is able tomaintain a 7x24 up and running cloud. IT Adminrequests deployment of an image across several VMinstances. Image is selected from the image libraryand provisioned across racks with each node Customer Valueprovisioning VMs. A hardware failure occurs and a  Zero downtime, tolerates hardware failures drivingblade is removed. The work is automatically balanced higher customer satisfactionacross remaining blades with no disruptions in  Power up and forget: Scale up cloud infrastructure withrunning the workloads. no manual configuration and rapid cross domain provisioningChallenge(s) this Scenario Addresses:  Choice through extensive hypervisor and platformNon availability of cloud infrastructure in event of supporthardware failure which impacts SLA’s 18 © 2012 IBM Corporation 18
  • 19. Use Case: image Lifecycle ManagementUse case: Control image sprawl and reducebusiness risk with rich analytics, imageversioning and federated image library tostandardize imagesOrganization has virtualized IT, but needs toimplement preventive measures to control imagesprawl. IT Architect designs and models the fullstack image according to company policies,synchronizes the image , and captures instancesas cloud ready image to publish to master imagerepository. The IT Admin controls image versionsand federates across all images by convertingimage formats and editing and analyzing image tooptimize impact on storage, and then selects animage from the library to deploy to repositories inthe cloud. Customer ValueChallenge(s) this Scenario Addresses: • Reduced data storage through smarter analytics and single instance storage can reduce image storage costsImage Sprawl and drift by up to 80%  Reduced risk of using non-compliant images that could have security exposure  Over 70% reduction in image provisioning time and labor costs through automated image lifecycle management19 © 2012 IBM Corporation 19
  • 20. DutchCloud Case Study © 2012 IBM Corporation
  • 21. About Dutch CloudDutch Cloud • Founded in 2009 with HQ in The Netherlands. • Team with long-term experience on Cloud Computing. • 100% committed to IBM. • Delivering “Private Clouds” (from a shared environment).Our Focus on • IaaS (Infrastructure as a Service). • the SMB Market in The Netherlands. • Partner Delivery Model (including Resellers). • Complex architectures. • Automation & Standardisation. • Adding network integration (Dutch Cloud is also ISP). • Adding simple tools; easy to use and easy to maintain.21 © 2012 IBM Corporation
  • 22. Dutch Cloud’s requirements• Rapid service delivery with high degrees of automation.• Customer isolation for multi-tenancy.• Customer and management traffic separation.• Integration with IBM V7000 storwize for non-local storage.• Easily extensible platform, supporting simple customisation.• Highly scalable and able to recovery autonomously from failures without interruptions to the service (no outages).• Ability to “brand” the portal/GUI for specific customers.• Ability to support a reseller model, and segregate resources.• It works…consistently, reliably, quickly, and with minimal administration.22 © 2012 IBM Corporation
  • 23. Solution Architecture Users Reseller Cloud Admin Linux VM Window VM IaaS VM WebConsole Web Services Zoo Keeper HBase x3650 M3 x3650 M3 iSCSI Compute Nodes Storage Nodes Storwize V7000 PXE Server Persistent Storage2323 © 2012 IBM Corporation
  • 24. Customer Deployment ScenariosRapid service delivery of IaaS & PaaS Disaster Recovery of IaaS & PaaSProblem: Customers want to respond quickly to Problem: Customer wants DR capability for IaaSbusiness events, and need to provision new for the provision of 200 machines within an SLAserver resources in a few minutes. of 60 mins. Typically this is done by having dedicated hardware on warm/cold standby.Benefit: SCP Allows us to provide a new level ofresponsiveness and agility that customers are Benefit: SCP means that we do not needfinding extremely beneficial to them, and driving dedicated hardware, but just ensuring we havemore revenue for us. (It’s a differentiator) sufficient total capacity available. This increases our utilisation rates / improves costs.Partner Reseller Model Development of Sharepoint ServicesProblem: Business partners don’t want to own Problem: One of customers uses high end laptopsidle capacity, but do want to scale up quickly to for the development of Sharepoint sites for itsrespond to their customer needs. customers – due to their hardware & storage constrained IT environment.Benefit: SCP supports a reseller model wherepresentation UI can be branded, quotas set for Benefit: SCP allow us to offer Sharepoint PaaSsoft limits and dedicated resources can be images that can not only be provided quickly, butassigned to support delivery for different with regular versioning on images for This offers a huge cost saving to the customer and24 improved agility. © 2012 IBM Corporation
  • 25. Leveraging IBM SmartCloud Provisioningthroughout the Development LifecycleCase Study © 2012 IBM Corporation
  • 26. Business Challenge Solution Hardware constraints Leverage SmartCloud Provisioning to rapidly  Significant number of machines required to support stand up development and test development life-cycle environments  Low HW utilization rates Currently using SmartCloud in 28 development  Forced to use outdated HW and test projects in Tivoli Time constraints  Required time to setup a complex topologies Reduced setup time from 2 hours to 5 minutes  Significant time spent on setup rather than testing Returned 356 old physical machines to surplus Skills requirements Enables more time to be spent validating  Training engineers to setup complex software stack software instead of setting up environments Before Avg: 2 hours Test Environment (physical machine) Request HW for TestingTester Now Avg: 5 mins Test Environment (SmartCloud instance) IBM SmartCloudTester Provisioning Launch instance 26 © 2012 IBM Corporation
  • 27. SmartCloud Monitoring integration with SCP © 2012 IBM Corporation
  • 28. Service Health for IBM SmartCloud Provisioning Pre-built integration with SmartCloud Monitoring Downloadable for free from http://www- The installation is based on IBM Installation Manager Leverages SmartCloud Provisioning PXE server to install and configure monitoring agents in the infrastructure Pre-built workspace to provide monitoring information on key SCP components Built-in monitoring situations: – Low disk space on compute nodes and storage nodes – Kernel Service VM failure – Compute node/storage node failure Supported – Unresponsive hyper-bot hypervisors: – Hbase/zookeeper processes failed ESXi, KVM28 © 2012 IBM Corporation
  • 29. 29 © 2012 IBM Corporation
  • 30. How to position SCP with customers? © 2012 IBM Corporation
  • 31. Selling SCP or ISDM? Key DifferentiatorsSmartCloud Provisioning Service Delivery Manager• Entry level offering • Mid/High-end level offering• Small/Medium Business / MSP • Medium / Enterprise / CSPs• Focus: IaaS, and simple PaaS • Focus: Iaas, PaaS, and complex PaaS +++• Provisioning: VMs only, thin provisioning, • Provisioning: VMs & Physical*, thick rapid deployment (15 secs – 20 mins) provisioning, 5-20 min deployment time• No workflow & approval • Strong Workflow engine (manual & system)• No reservations • Reservation based scheduling.• Flat service catalog / image library • Flexible & configurable service catalog• Basic service provider capabilities • Strong service provider focus (fully multi-tenant)• No monitoring (can upsell SCM) • Integrated Monitoring with ITM• No metering (can add TUAM) • Integrated Metering with TUAM• Multi-hypervisor (KVM, Xen, ESXi) • Multi-hypervisor, strong System P management• No hypervisor manager required (plus ESX, KVM, Xen)• No OOB extensions (yet) • Control of hypervisor through manager• Strong image lifecycle tooling (VIL / ICON) • Many extensions: NetApp storage cloudlet,• Distributed architecture network integration, firewalls, hybrid • Library of automation packages (thru TPM) • Centralised architecture31 © 2012 IBM Corporation
  • 32. Upcoming features © 2012 IBM Corporation
  • 33. Upcoming featuresAttention: References in content to IBM products, software, programs, services or associated technologies do not imply that they will beavailable in all countries in which IBM operates. Content, including any plans contained in content, may change at any time at IBMs solediscretion, based on market opportunities or other factors, and is not intended to be a commitment to future content, including product or featureavailability, in any way. Statements regarding IBMs future direction or intent are subject to change or withdrawal without notice and representgoals and objectives only. Please refer to the developerWorks terms of use for more information.  Merge with IBM Workload Deployer  Group Level Administration  Static IP support  Live migration  VMs take over  MS Active Directory integration  Image Registration  Single sign-on and launch out for the adminconsole  Diagnostic tool for problem determination 33 © 2012 IBM Corporation
  • 34. Upcoming features (contd.)Attention: References in content to IBM products, software, programs, services or associated technologies do not imply that they will beavailable in all countries in which IBM operates. Content, including any plans contained in content, may change at any time at IBMs solediscretion, based on market opportunities or other factors, and is not intended to be a commitment to future content, including product or featureavailability, in any way. Statements regarding IBMs future direction or intent are subject to change or withdrawal without notice and representgoals and objectives only. Please refer to the developerWorks terms of use for more information.  Virtual image library enhancements: • Instance-level authorization • Portability checks and remediation • Standalone installation  Image Construction and Composition Tool enhancements: • Add software to Windows images • Command line support 34 © 2012 IBM Corporation
  • 35. Merge with IWDAttention: References in content to IBM products, software, programs, services or associated technologies do not imply that they will be available in all countries in which IBM operates. Content,including any plans contained in content, may change at any time at IBMs sole discretion, based on market opportunities or other factors, and is not intended to be a commitment to future content,including product or feature availability, in any way. Statements regarding IBMs future direction or intent are subject to change or withdrawal without notice and represent goals and objectives only.Please refer to the developerWorks terms of use for more information. Self Service and Admin UI REST APIs Virtual Virtual Appliance Systems Placement/Deployment Hypervisors & IAAS Management SHIMS VMControl vCenter IAAS API Shim SHIM SHIM Virtualization Management Based Provisioning Power VM VM vCenter HSLT KVM, Xen, ESXi 35 © 2012 IBM Corporation
  • 36. Collaborating with us If you would like to:  Hear the latest news on IBM SmartCloud Provisioning  Familiarize yourself with the product functionalities  Help IBM to improve product functionality and usability  Get in touch with our Subject Matter Experts  You can participate in any of our Customer Interaction Program activities, which includes:  Open Beta (  See the new features under development, with recorded Demos (  Work with a preconfigured Virtual Beta environment. Explore the product, either with our direct support or in an unattended mode, without the effort of installation and configuration.  For more details go to Service Management Connect* at *Service Management Connect: You can connect, learn, and share with Integrated Service Management (ISM) professionals in Service Management Connect. Get access to developers and technical experts who provide their perspectives and expertise to help you implement ISM solutions.36 © 2012 IBM Corporation
  • 37. Additional resources  IBM SmartCloud Provisioning information center: %20SmartCloud%20Provisioning/page/Home  IBM SmartCloud Provisioning WIKI: %20SmartCloud%20Provisioning/page/Home • Additional resources • Customer interaction • Demos • Upcoming features  IBM SmartCloud Provisioning forum: © 2012 IBM Corporation