1. IBM Labs in Haifa © 2008 IBM Corporation RESERVOIR – Bringing the Cloud down to Earth Eliot Salant – salant@il.ibm.com RESERVOIR Project Coordinator Manager, Virtualization Technologies, IBM Haifa Lab June 4, 2008 The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 215605

2. IBM Labs in Haifa © 2008 IBM Corporation 2 RESERVOIR @ NORTUR 2008 Agenda  What is Cloud Computing?  Evolution of Cloud Computing  The RESERVOIR vision

3. IBM Labs in Haifa © 2008 IBM Corporation 3 RESERVOIR @ NORTUR 2008 What is Cloud Computing?  The definition depends on whom you talk to…  Utility Computing: A pool of virtualized computer resources that IT can access on demand (Example: IBM Blue Cloud, Google App Engine, Amazon EC2…)  Software as a Service (SaaS)/On Demand Software : Delivers a single application through the browser to thousands of customers using a multitenant architecture (Example: salesforce.com, Google docs…)  To quote Ian Foster: So is “cloud computing” just a new name for grid? In information technology, where technology scales by an order of magnitude, and in the process reinvents itself, every five years, there is no straightforward answer to such questions.

4. IBM Labs in Haifa © 2008 IBM Corporation 4 RESERVOIR @ NORTUR 2008 “You say ‘tomato’, I say tomahto…”  Consider cloud computing as providing a service for users to run complete applications from centralized servers sharing resources such as memory, bandwidth, cpu and storage.  Grid computing provides a mechanism for running processes across multiple compute resources.

5. IBM Labs in Haifa © 2008 IBM Corporation 5 RESERVOIR @ NORTUR 2008 What are the requirements on clouds?  Demand puts requirements on scalability, reliability…  One billion people worldwide have access to the Web  MySpace signs up an average of 300,000 new users every day with 65 billion page views per month.  In 2Q 2006, 50 million blogs were created at the rate of 2 per second.  And what will happen as millions (billions?) of inexpensive sensors (“smart dust”) start connecting to the Web?  Web 2.0 best practice principles will also drive infrastructure requirements:  Release early, release often  Operations are a core competency  High availability, systems monitoring and management…

6. IBM Labs in Haifa © 2008 IBM Corporation 6 RESERVOIR @ NORTUR 2008 The next generation Cloud has many challenges to tackle  Security  Scalability  Availability  Reliability  Cost Efficient  Data Intensive  Personalization  Mobility  Latency  Manageability

7. IBM Labs in Haifa © 2008 IBM Corporation 7 RESERVOIR @ NORTUR 2008 But, the financial incentives are there…  OnDemand apps like Salesforce.com can be provisioned for as little as $300-500 per subscriber after fully costing hardware, software and service vs. as much as $8,000-10,000/user for OnPremise clientserver apps.  Merrill Lynch estimates that today’s $2 billion market in on-demand applications will expand to a $165 billion market opportunity.

8. IBM Labs in Haifa © 2008 IBM Corporation 8 RESERVOIR @ NORTUR 2008 Who are the main commercial clouds players today?  Amazon Web Service (AWS) offers Elastic Compute Cloud (EC2), and Simple Storage Service (S3), Simple Db and more  EC2: Can hire small, large or extra large instances which gives set configurations for memory, storage and EC2 Compute Units (1.0 – 1.2 GHz 2007 Opteron or 2007 Xeon processor. )  Google offers a range of cloud apps, and platforms (Google App Engine, Google Apps)  Google and IBM Academic Cluster Computing initiative  IBM Blue Cloud – offers infrastructure and platform support  Salesforce.com – offers Force.com - a development platform in the Cloud  Microsoft has some offerings, such as Office Live for small businesses  Activision – World of Warcraft have over 10 million paying users

9. IBM Labs in Haifa © 2008 IBM Corporation 9 RESERVOIR @ NORTUR 2008 The Commoditization of ICT  As far back as 1960, John McCarthy predicted that “computation may someday be organized as a public utility”.  In fact, in early grid days, the computing grid was envisioned as being analogous in form and function to the electric grid.  In 2003, The Harvard Business Review published an article by Nicholas Carr entitled, “IT doesn’t matter”.  Carr argued that once IT’s power and presences reach a widespread enough state, it simply becomes a commodity – a cost of doing business – rather than an advantage to a single player

10. IBM Labs in Haifa © 2008 IBM Corporation 10 RESERVOIR @ NORTUR 2008 The Evolution of the Power Grid http://www.pbase.com/rbenny/image/29116201 http://www.anl.gov/Media_Center/logos22-1/electricity.htm The US National Power Grid http://www.rootsweb.com/~nytigs/BurdenPayrollRecords.htm The Burden Iron Works Water Wheel http://ieee-virtual-museum.org/collection/event.php?id=3456876 The Pearl Street Station Make your own infrastructure Not the company’s main business but a considerable competitive advantage The utility industry Metering Limited reach Reproducible (yet costly) Efficient distribution Federation of providers The diversity factor Economies of scale

11. IBM Labs in Haifa © 2008 IBM Corporation 11 RESERVOIR @ NORTUR 2008 The Evolution of the Compute Grid Make your own infrastructure Not the company’s main business but a considerable competitive advantage The utility industry Metering Limited reach Reproducible (yet costly) Efficient distribution Federation of providers The diversity factor Economies of scale http://www.by-star.net/techspeak/datacenter/ http://www.smcplus.com/applications.asp?id=32 http://www.informationweek.com/galleries/showImage.jhtml?galleryID=62&imageID=13 Google @ The Dulles, OR R E S E R V O I R “… today’s commercial clouds have not been open and general purpose, but instead been mostly proprietary and specialized for the specific internal uses (e.g., large-scale data analysis) of the companies that developed them. The idea that we might want to enable interoperability between providers (as in the electric power grid) has not yet surfaced …” “…will move towards a mix of microproduction and large utilities, with increasing numbers of small-scale producers co-existing with large-scale regional producers, and load being distributed among them dynamically …” There’s Grid in then thar CloudsThere’s Grid in then thar Clouds - Ian Foster

12. IBM Labs in Haifa © 2008 IBM Corporation 12 RESERVOIR @ NORTUR 2008 The Reservoir Vision - Positioning  RESERVOIR is an aggressive research attempt to meet the emerging needs of the service-based economy sponsored by the EU  Provide revolutionary foundation for a new European infrastructure where resources and services can be transparently and dynamically managed, provisioned and relocated like utilities – virtually “without borders”  No single facility/provider can create a seemingly infinite infrastructure capable of serving massive amounts of users at all times, from all locations  Federation of clouds  Leverage the diversity factor to achieve economies of scale  Leverage locality  There are many other solutions out there - so what’s new in RESERVOIR ?

13. IBM Labs in Haifa © 2008 IBM Corporation 13 RESERVOIR @ NORTUR 2008 The RESERVOIR Vision  Goal: Create the next generation Compute Cloud for the reliable and effective delivery of IT services as utilities  Example: EU Winter Olympics Scenario to highlight competitive differentiation vs. present technologies, e.g. EC2

14. IBM Labs in Haifa © 2008 IBM Corporation 14 RESERVOIR @ NORTUR 2008 EU Olympics Scenario – Service Definition … … … Web site service for EU Olympics 1. The Olympic committee uses client tools to generate the service definition. Includes: Tier definition (web servers, application servers, databases) Required Virtual Execution Environments (VEEs) Software Images Storage Network Required configuration Inter-tier relations Required QoS.

15. IBM Labs in Haifa © 2008 IBM Corporation 15 RESERVOIR @ NORTUR 2008 EU Olympics Scenario – Service Deployment 2. The committee negotiates and ships the service definition to a primary RESERVOIR site (PRS) 3. The PRS automatically deploys the complex service on its own site: Configure required storage & network, creates VEEs selecting proper physical resources to meet QoS Install required images, software according the service definition Apply the required configuration Setup the monitoring and billing … … … web servers App servers Network DB servers VEE Phys server PRS

16. IBM Labs in Haifa © 2008 IBM Corporation 16 RESERVOIR @ NORTUR 2008 RESERVOIR Differentiator: Service definition language enabling automatic deployment of complex services over virtual infrastructure

17. IBM Labs in Haifa © 2008 IBM Corporation 17 RESERVOIR @ NORTUR 2008 EU Olympics Scenario – Service Cooperation PRS RS1 RS2 <service.. -- - - - - - - - -- - - - - - - - <service.. 4. For HA and assuring the SLA, the PRS negotiates with two other RS (RESERVOIR site), and ships the service definition to them 5. Each RS deploys the service (according to the contracted resources) in its site similar to what the PRS did 6. The PRS and RS1, RS2 maintain a service cooperation relationship for the EU games service Overlay network Content distribution Image and software updates Load balancing <service.. -- - - - - - - - -- - - - - - - - <service.. Service Cooperation

18. IBM Labs in Haifa © 2008 IBM Corporation 18 RESERVOIR @ NORTUR 2008 RESERVOIR Differentiator: Inter-domain management site protocols that enable multiple management sites to cooperate in providing a single service, where the cooperation is automatically driven from a service definition document.

19. IBM Labs in Haifa © 2008 IBM Corporation 19 RESERVOIR @ NORTUR 2008 EU Olympics Scenario – HA with Live VM Migration PRS RS1 RS2 7. PRS site suffers electricity problems and needs to power off physical servers. 8. PRS negotiates for additional resources at RS1 employing the RS-RS protocol 9. PRS evacuates the VEEs on the servers to be powered off, migrating them to RS1 - Live migration to maintain application servers’ states and client connections

20. IBM Labs in Haifa © 2008 IBM Corporation 20 RESERVOIR @ NORTUR 2008 RESERVOIR Differentiator: Live migration without borders: Cross geographical, network and management domains

21. IBM Labs in Haifa © 2008 IBM Corporation 21 RESERVOIR @ NORTUR 2008 EU Olympics Scenario – On Demand Service Expansion 10. Load increases and PRS realizes that the available resources at the 3 sites are not enough 11. PRS negotiates with additional RS3, and ships it the service definition 12. RS3 deploys the service (according to the contracted resources), and dynamically joins the service cooperation relationship for the EU Olympics service PRS RS1 RS2 RS3

22. IBM Labs in Haifa © 2008 IBM Corporation 22 RESERVOIR @ NORTUR 2008 RESERVOIR Differentiator: The ability to dynamically hire additional 'service power‘ from a new management site, fully automated, using the service definition language and the inter-domain site protocols

23. IBM Labs in Haifa © 2008 IBM Corporation 23 RESERVOIR @ NORTUR 2008 The Reservoir Vision -the technical goals  Minimize over-provisioning of resources  Dynamic allocation and re-provisioning can get better utilization out of existing resources  Break down platform and geography barriers  Adhere to SLA constraints through intelligent placement and relocation algorithms  Address cross administrative domain SLA  Domains may be in different organizations  Create standards to allow for interoperability between administrative domains  Must be able to transfer information to allow applications to run on different domains  SLA, billing, application meta-data…

24. IBM Labs in Haifa © 2008 IBM Corporation 24 RESERVOIR @ NORTUR 2008 Approach  Focus on technologies that enable to build cooperating computing clouds  Connect computing clouds to create an even bigger cloud  Integration of virtualization technologies with grid computing driven by new techniques for business service management   The Service Oriented Infrastructure (SOI) equation:  Building on this equation we will architect and implement a platform for supporting complex services, which  Enables dynamic deployment of complex multi-tier services across heterogeneous administration domains  Uses virtualization of servers, storage and network to allow migration without borders  Supports service definition, SLA management, accounting and billing

25. IBM Labs in Haifa © 2008 IBM Corporation 25 RESERVOIR @ NORTUR 2008 Evolution of a solution: In the beginning, there was Grid Computing Grid node or Service Site Physical Resources Service Tasks

26. IBM Labs in Haifa © 2008 IBM Corporation 26 RESERVOIR @ NORTUR 2008 Evolution: Virtualization technologies for Intel machines take hold: Grid Computing + Virtualization Improved isolation, Relax dependencies, Well defined billing units Virtual Execution Environment (VEE)

27. IBM Labs in Haifa © 2008 IBM Corporation 27 RESERVOIR @ NORTUR 2008 Policy 1: If possible keep VEEs from the same organization in the same physical box Evolution - SOI: Grid Computing + Virtualization + BSM

28. IBM Labs in Haifa © 2008 IBM Corporation 28 RESERVOIR @ NORTUR 2008 Policy 1: If possible keep VEEs from the same organization in the same physical box SOI: Grid Computing + Virtualization + BSM Policy 2: Turn off underutilized physical boxes

29. IBM Labs in Haifa © 2008 IBM Corporation 29 RESERVOIR @ NORTUR 2008 Policy 2: Turn off underutilized physical boxes Policy 1: If possible keep VEEs from the same organization in the same physical box SOI: Grid Computing + Virtualization + BSM Local optimizations (within a single site) : placement, power, etc.

30. IBM Labs in Haifa © 2008 IBM Corporation 30 RESERVOIR @ NORTUR 2008 Policy 3: If possible keep VEEs in “owning” organization RESERVOIR: Grid Computing + Virtualization + BSM – Boundaries

31. IBM Labs in Haifa © 2008 IBM Corporation 31 RESERVOIR @ NORTUR 2008 Policy 3: If possible keep VEEs in “owning” organization Policy 4: If possible keep VEEs in least number of external organizations RESERVOIR: Grid Computing + Virtualization + BSM – Boundaries

32. IBM Labs in Haifa © 2008 IBM Corporation 32 RESERVOIR @ NORTUR 2008 Policy 3: If possible keep VEEs in “owning” organization Policy 4: If possible keep VEEs in least number of external organizations RESERVOIR: Grid Computing + Virtualization + BSM – Boundaries

33. IBM Labs in Haifa © 2008 IBM Corporation 33 RESERVOIR @ NORTUR 2008 Policy 5: “Follow” the service customer RESERVOIR: Grid Computing + Virtualization + BSM – Boundaries Migration across sites  Global optimizations: placement, cost, bandwidth, etc.

34. IBM Labs in Haifa © 2008 IBM Corporation 34 RESERVOIR @ NORTUR 2008 Virtualize the Network Create virtual networks connecting VEEs regardless of physical server location

35. IBM Labs in Haifa © 2008 IBM Corporation 35 RESERVOIR @ NORTUR 2008 Virtualize the Network and the Storage Enable secure access to relevant data regardless of storage location

36. IBM Labs in Haifa © 2008 IBM Corporation 36 RESERVOIR @ NORTUR 2008 Looking a bit beneath the RESERVOIR covers  Security requirements  Challenges  Architecture  Use cases  Testbed  Partners

37. IBM Labs in Haifa © 2008 IBM Corporation 37 RESERVOIR @ NORTUR 2008 RESERVOIR Security considerations 1.Guarantee the security of applications and associated data, allowing end users to specify requirements for service tasks  Protecting a service from other services running in the same virtual environment  Protect confidentiality of stored service data  Need to protect service data relating to amount of resources consumed, accrued billing...  Handle requirements induced by multi-tenancy  The Service Definition will need to support special requirements/restrictions due to multi-tenancy  Example: I don’t want my data residing on the same physical storage as my competitor  Protecting a VEE from other VEEs running in the same compute node

38. IBM Labs in Haifa © 2008 IBM Corporation 38 RESERVOIR @ NORTUR 2008 RESERVOIR Security continued Guarantee the ability of SOI vendors to interoperate in a secure way, building mutual trust and defending themselves from misbehaving vendors or end users.  Ensure the authenticity and integrity of management entities, compute nodes and VEEs.  Secure communication of sensitive end user and vendor data over local and wide area networks (message integrity and confidentiality)  Protecting the access to the management interfaces 3.Security policies for a site must be securely discoverable in order for cross-domain migration i.e. only allow migration to sites with the same security policy

39. IBM Labs in Haifa © 2008 IBM Corporation 39 RESERVOIR @ NORTUR 2008 Security hot spots in RESERVOIR VEEMS-VEEMS Service security specifications Service tasks isolation VEE isolation VEE – Host protection VEEMS internal management Service provider interface VEEMS admin interface Network, storage setup

40. IBM Labs in Haifa © 2008 IBM Corporation 40 RESERVOIR @ NORTUR 2008 RESERVOIR Challenges – At the Service Level  Translate business concept requirements to infrastructure requirements  Define a Service Definition Language to characterize all information and context required to enable lifecycle management of services across RESERVOIR sites  Must be able to handle rollback on deployment failures  Determine the mapping of high level service requirements and metrics (e.g., response time) to infrastructure level requirements and metrics (e.g., CPU utilization)  Support multiple levels of QoS

41. IBM Labs in Haifa © 2008 IBM Corporation 41 RESERVOIR @ NORTUR 2008 Management Challenges  Support policy based management across administrative domains (clouds)  Service level utility analog of electricity power,  Dynamically automatically hire additional 'power‘ from a another cloud  Enable intra-site and inter-site workload optimization, HA and SLA management. (Or, maybe not…)  Create an inter-site protocol to allow for federation of RESERVOIR sites  Protect Service Level Agreements  Detect violations (SLA monitoring)  Provide for dynamic relocation of resources  Provide accountability  Bill for services used, even across RESERVOIR sites  Different billing and accounting systems may be used.

42. IBM Labs in Haifa © 2008 IBM Corporation 42 RESERVOIR @ NORTUR 2008 Challenges at the Infrastructure level  Provide for relocation of resources without boundaries  Live migration across subnet boundaries  Migration to a different physical host without shared storage  Provide standardized interfaces for lifecycle management to Virtualized Execution Environment (virtualized machines, Virtualized Java Service Containers)  Analyze end-to-end performance in a virtualized environment to understand bottlenecks  Be able to handle surges in 3-5 orders of magnitude in service requests

43. IBM Labs in Haifa © 2008 IBM Corporation 43 RESERVOIR @ NORTUR 2008 The Reservoir Architecture Monitor service and enforce SLA compliance by managing capacity of Service Components (VEEs) or/and size of Service Tiers Deals with translation/mapping of service concepts/metrics (response time) to infrastructure concepts/metrics (VEE size) Monitor VEEs and find best VEE placement that meet constraint satisfaction problem Deals federation of domains

44. IBM Labs in Haifa © 2008 IBM Corporation 44 RESERVOIR @ NORTUR 2008 Project Structure A4: Service Management (TID) A2: VEE Infrastructure (IBM) A3: VEE Management (UCM) A1: Architecture (IBM) A5: Testbed and Scenarios (UniMe) A6: Dissemination (CETIC)

45. IBM Labs in Haifa © 2008 IBM Corporation 45 RESERVOIR @ NORTUR 2008 The RESERVOIR use cases  We have four use cases from industrial partners  SAP – Business application  Multi-tenant service delivery for SMBs in a data center  Managing thousands of different service components that comprise a variety of service applications executed by thousands of VEEs.  Deployment of a business application with one click  Deployment based on Service Manifest  Relocation of a multi-tiered business application  Sun –  Utility computing (example: digital content creation – such as computer generated animated films or special effects), or Web 2.0 application  Test performance under the following conditions:  Frequent change of resource use  Unpredictable loads  Pay as you go use.

46. IBM Labs in Haifa © 2008 IBM Corporation 46 RESERVOIR @ NORTUR 2008 Use cases cont.  Telefonica – Telco application  Test a mash-up application on top of the RESERVOIR infrastructure  Lifecycle management in a highly dynamic environment  Accounting, billing and business management from different gadget providers.  Thales – e-Government  Three-tiered application  Tests handling dynamic loads – cyclic demands on a user application to reflect hourly/seasonal peaks  Maintenance scenarios to physical resources  Application manageability – QoS tradeoffs, large number of simultaneous connections etc.

47. IBM Labs in Haifa © 2008 IBM Corporation 47 RESERVOIR @ NORTUR 2008 The RESERVOIR development process  There will be 3 main cycles of code development and delivery to the testbed  The testbed will integrate the code, and then test against the use cases  Additional testbed experiments will:  Compare performance against the native use case environment  Feedback defects to the developers  Analyze for scaleability, bottlenecks etc.

48. IBM Labs in Haifa © 2008 IBM Corporation 48 RESERVOIR @ NORTUR 2008 The RESERVOIR Consortium PartnerRoleComment IBM HRLTechnologyProject Lead, Virtualization/SOA Infrastructure Telefonica I+DTechnologyService Technology, Billing Infrastructure UCMTechnologyGrid, Dynamic Allocation Technology ThalesTechnologySecurity, Virtualization Infrastructure, Hosting SAPUse-CasesUse-Cases, Contribution to Requirement an Standards Sun MicrosystemsUse-Cases + TechContribution to Standards, Java Services, Monitoring DATAMATTechnologyService Management Technologies University LuganoTechnologyPartner, Monitoring and SLA Management University UMEATechnologyMonitoring, Measuring and Billing Technology University MessinaTechnologyGrid Experience, Testbed Development, UC LondonTechnologyVirtualization Technology CETICTechnologySecurity OGFStandardizationGrid and Virtualization Standards

49. IBM Labs in Haifa © 2008 IBM Corporation 49 RESERVOIR @ NORTUR 2008 Project Status  3 Year FP7 project started in February  Kickoff meeting hosted by IBM in Haifa at the beginning of February  21 participants from abroad, representing all 12 partners (+local IBMers)  Architecture work-package started at month 1, others now starting  First version of architectural specification is out  Started working on building the testbed  15 machines at UniMe, 4 machines at IBM (2 more on the way) and 8 machines on the way to UCM  Web site up and running  Come visit us at http://www.reservoir-fp7.eu/http://www.reservoir-fp7.eu/

50. IBM Labs in Haifa © 2008 IBM Corporation 50 RESERVOIR @ NORTUR 2008 There are no stupid questions or stupid answers

51. IBM Labs in Haifa © 2008 IBM Corporation 51 RESERVOIR @ NORTUR 2008 A2: VEE Infrastructure  Virtual Machine Technologies (IBM)  Improve performance of VEE execution for typical RESERVOIR workloads  Provide VEEMS enablement layer for virtual machines  Relocation Enablement (IBM)  Network Virtualization  Storage Virtualization  Java Service Containers (Sun)  Provide VEEMS enablement layer for virtual java service containers

52. IBM Labs in Haifa © 2008 IBM Corporation 52 RESERVOIR @ NORTUR 2008 A3: VEE Management  VEE Provisioning and Supervision (UCM)  Image management  Monitoring  Allocation Policy Management (Datamat)  Policy based placement and migration  Federation of Management Domains (UCM)  Built atop WSRF interfaces to access remote VEE Supervisors  Push new and leverage existing OGF/DMTF/OASIS standards  Interoperability between administrative domains and scheduling heuristics on federated and utility architectures.

53. IBM Labs in Haifa © 2008 IBM Corporation 53 RESERVOIR @ NORTUR 2008 A4: Service Management  Service Definition (UCL)  Design a new service description language that will allow the description of service interfaces, service lifecycle, interface bindings to implementations, service deployment, SLA requirements for a service, rules for VEEs (re)configuration and (re)organisation and service components distribution and configuration  Revisit the service lifecycle definition and extend it to accommodate the influence of virtualisation  Extend tools available for service design (for example the Eclipse Web Tools Platform)  Standardize the service description language  Service Management (TID)  SLA monitoring across administrative domains settings and service-oriented architectures.  Integrate monitoring with resource allocation and scheduling and take explicit account of the potentially synchronous nature of service invocations.  Automatic deployment of services based on complex service definition  Accounting, Billing and Payment (TID)  Accounting and billing arrangements for outsourced services are based on raw machine resource consumption (CPU-time, storage capacity etc)  RESERVOIR will pursue the definition of a framework that allows accounting and billing in terms of the services that were completed, taking into consideration the quality of service that was provided.

54. IBM Labs in Haifa © 2008 IBM Corporation 54 RESERVOIR @ NORTUR 2008 A5: Experimentation and Validation  Testbed (UniMe)  Create the necessary environment for testing and validation  A multi-site environment running the last release of RESERVOIR middleware to evaluate (built on physical resources distributed/owned/managed by some of the project’s partners)  Scenario 1: eGov application (Thales)  Automatic adjustment of resources and domains cooperation.  Scenario 2: SAP business application (SAP)  Business application oriented use cases and the opportunities to execute them on a flexible infrastructure.  Scenario 3: Utility computing (Sun)  Deploy arbitrary operating system and application stacks on remote resources. Provide secure and seamless access to them. Adjust resource allocation on-demand without the end user noticing disruption of service  Scenario 4: Telco application (TID)  Hosting web sites that deals with massive access (e.g., the Olympics games)  High degree of personalization and support for mashups