General Manager Financial Services by mer14442

VIEWS: 0 PAGES: 22

General Manager Financial Services document sample

More Info
									Grid 2.0 : Entering the new age of Grid in
Financial Services

Jim Mancuso, General Manager Financial Services




September 5, 2008
            Grid 1.0: the Results
 Elapsed
  Time




                                          Symphony provides business advantage
                                           sub-millisecond latency
                                           > 90 MB/sec throughput per client
                                           <5 seconds to re-allocate 2,000 CPUs




           Normal Application Execution             Symphony Execution

Imagine the impact that instantaneous response and nearly unlimited compute
           power could have on your business critical application!
        Business Demands
• Beat the Competition
  – Financial institutions are creating more complex
    products and increasing volumes

• Manage Risk
  – Address new regulatory requirements and be able to
    respond in real-time to global volatility

• Be Cost Effective
  – IT decisions are scrutinized for the business value
    they add, the potential ROI to the enterprise, and the
    Total Cost of Ownership
       Successful Grid Deployments

Front Office                  Middle Office                Back Office
Trading                       Risk/End of day report       Clearance & Settlement
Investment Mgmt               Market Research              Batch Reporting
Exchange Pricing              Financial Accounting         Batch processing




                    Capital
                    Market                                            Vanilla
                                                                       Flow
      Asset                                             Exotic
                                     Hedge
    Management                                         Derivative                Risk
                                     Funds
                                                                                Analysis

                                                          Model         Basket
           Option                                       Calibration     Product
                              Exchanges
           Houses
       Going Real-time

• Strategic investment to optimize trade
  results
  – Significantly reduce calculation time to
    improve trade effectiveness
  – Pre-trade risk analysis, backtesting, and other
    processes typically done post-trade
• Calculation profiles require low latency
  workload distribution
  – For complex products, < 2ms runtime per
    calculation
       Reech AIM – A Back-testing Example

• Historical Environment
           Overnight                       Daily



    Backtest           Selection          Execution

• Intraday Grid
                             Daily



     B1        S1       E1           Bn     Sn        En
                 Grid 1.0: Utilization Still Too Low



Unpredictable
     Demand


                                                                   Desktops/Workstations


Un-scalable!                                                                               Untapped
                                                                                           Potential



                                   X                               DR Sites and Spares



 Un-shareable
   Resources
     (“Silos”)
                   Linux/UNIX Clusters        Windows Clusters         UAT Servers

                                         Underutilized Resources
                       IT Challenges


                             Required Resources: CPU, Memory, and Overhead




                                         Environment 3                      Finance, Payroll
Service levels                                                                                       Multiple
and required                                                                                        domains,
 resources                                                                                        departments
                                                                                                       &
                                                                                                  technologies
                                         Environment 2                      Development




                                         Environment 1                      Website

                 7am   8am   9am   10am 11am 12pm 1pm   2pm   3pm   4pm   5pm   6pm   7pm   8pm
                                           Time of Day
                       Automate to Reduce Costs


                             Required Resources: CPU, Memory, and Overhead

                                                 $ savings

                  Adjusted Resources: VMs, CPU, Memory, and Overhead with Automation
Service levels                                                                                       Multiple
and required                                                                                        domains,
 resources                                                                                        departments
                                                                            Finance, Payroll           &
                                                                                                  technologies

                                                                            Development



                                                                            Website

                 7am   8am   9am   10am 11am 12pm 1pm   2pm   3pm   4pm   5pm   6pm   7pm   8pm
                                           Time of Day
Making Cost Reduction a Reality
              Making Cost Reduction a Reality
  Costs             Data Center          Silo Grids      Utility Grid
 (1 Year)
            Environment (CPU’s)                  4,000            2,500   Server Consolidation –
                                                                          resource sharing
CAPEX       Hardware + Chassis            $20,000,000     $12,500,000     Movement to more
                                                                          commodity hardware
            OS support                       $400,000         $250,000    Reduced OS
                                                                          Subscriptions
            Storage                           $86,400          $86,400
OPEX        Data Center (power,             $1,822,080      $1,138,800    Utility Grid = true Green
            cooling)                                                      Computing
            Data Center (rental)             $600,000         $600,000    not going to reduce the
                                                                          size of the DC
            System Administrator            $2,500,000      $1,750,000    More CPU’s managed
                                                                          per Admin
            Application Dev/Support         $4,000,000      $4,000,000
            team
            Total                         $29,408,480     $20,325,200


                                   Year 1 Savings of $9M!
Making Grid 2.0 a Reality: the Internal Cloud
                Four Phases of Grid Adoption
         Phase 1        Phase 2              Phase 3                  Phase 4

                           LOB                LOB              LOB
                                                              LOB             LOB
        LOB              LOB                LOB




                        A A             A A A A             A A A A A A
        A
                        A A             A A A A             A A A A A A


Phase 1: Silo Grid – Grid enable application running on a commodity cluster
Phase 2: LOB Grid – Resource sharing among multiple applications
Phase 3: Enterprise Grid – Utility computing; enterprise scale and management
Phase 4: Cloud Computing – Beyond HPC: commercial applications on grid
          Citi
• Challenge: Meet business growth
  objectives and reduce computing costs                        “Our enterprise grid
                                                               is enabling our IT
   – Make better use of underutilized computing
                                                               users to be more
     resources                                                 creative in how they
   – Provide additional processing power required              share computing
     to improve performance of key business                    resources to do
                                                               their development
     applications                                              and testing work,
• Solution:                                                    thus helping us
                                                               achieve our
   – Built a utility grid, combining 20+ major intra           expensive efficiency
     and end of day pricing and risk applications              targets."
                                                                       John van Uden
                                                                             SVP, Citi
           • Utilization went from 20% to over 80%
           • Infrastructure management is simplified
           • A global utility-computing platform is in place
             – providing on-demand processing power at
             significantly reduced cost
Citi’s Grid Experience – What’s Next?


                             Move to shared resource
                             environment in a
                             controlled manner




                             End State: all resources
                             are shared with grid
                             scheduling and policies
                             ensuring SLA’s are met
            The Platform Solution


PLATFORM ACCELERATE                            PLATFORM MANAGE

        Platform Symphony                        Platform Analytics


   LSF        LSF                    LSF
                                                   Platform RTM
                      LSF Multi-
 License    Process                 Session
                       Cluster
Scheduler   Manager                Scheduler


                                                 Platform Manager
        Platform LSF Family


                 Platform Enterprise Grid Orchestrator (EGO)
                           Platform VM Orchestrator
                        Platform Management Console




                                                               6/23/2011   17
        Platform Unique Benefits
• Cost Savings
  – Cut your silo server farms in half
  – The only true utility computing model

• Speed
  – The lowest-latency HPC solutions
  – 5 x 10 times faster than alternative environments

• Openness
  – Easy for developers to build and deploy their HPC
    applications
  – Free tools to download with no time or node-count
    restrictions
    Leader in HPC


4,000,000   Managed CPUs

   2,000    Customers worldwide

     500    Employees in 15 offices

      16    Years of profitable growth

        1   Leader in HPC
Financial Services Customers
Partnerships & ISV Integrations
Questions ?




   Contact email : jmancuso@platform.com
   Contact tel : 646-290-5159
   www.platform.com

								
To top