High performance data center computing using manageable distributed computing
Download as pptx, pdf3 likes9,624 views
Terrapin Trading Show Chicago, Thursday, June 4 Andy Bach, FSI Architect, Juniper Networks Distributed computing concepts (QFX5100-AA) Scale and performance enhancements (QFX10000 Series) Automation capabilities (tie in QFX-PFA) Larry Van Deusen, Director of the Network Integration Business Unit, Dimension Data Automation Value Added Partner Services
![[APIdays INTERFACE 2021] Now that we have K8s, can we stop re-inventing API p...](https://cdn.slidesharecdn.com/ss_thumbnails/apidaysk8platform-210708083659-thumbnail.jpg?width=560&fit=bounds)
![[WSO2Con EU 2017] Building Smart, Connected Products with WSO2 IoT Platform](https://cdn.slidesharecdn.com/ss_thumbnails/kjpxgjy0rds38akb7ykr-signature-9119aa91aa49883e48c4941d40fb9827f8ba8afecf167aa90a325ffcd0cac924-poli-171107140509-thumbnail.jpg?width=560&fit=bounds)
More Related Content
What's hot (20)
Viewers also liked (19)
Similar to High performance data center computing using manageable distributed computing (20)
More from Juniper Networks (20)
Recently uploaded (20)
High performance data center computing using manageable distributed computing
- 1. High Performance Data Center Computing using Manageable Distributed Computing Andrew Bach, Chief Architect for Financial Services, Juniper Networks Lawrence Van Deusen II , Director – Networking Business Unit, Dimension Data Chicago Trading Show Thursday , June 4th, 2015, 3:10 – 3:40
- 2. Agenda Redistribute Computing as a Solution The Limitations Today Automation in a distributed environment The Challenges
- 3. Agenda Redistribute Computing as a Solution The Limitations Today Automation in a distributed environment The Challenges
- 4. • Requirement for long term retention of data • Requirement to archive meta data • Real time risk management is now required • Requirement for precision timing • Increasing focus on cyber security • Time to market must be faster • Product life time is shorting • Margins are driving OPEX reduction • Regulation is increasing the cost of business • Technology is a strategic weapon • Fueling the race to the triple crown of technology (0 cost, 0 latency, 0 time to market) • Bandwidth demand continues to grow at >30% per year • Rate of data growth is also increasing • Analytics, automation, orchestration are critical • Exchange trading floors continue to fade away • Machine assisted trading dominates • Heuristic trading based on hyper-contextual information is growing • Risk assessment is now routinely based on massive near real time data bases • Shared public/private cloud solutions • Operational excellence is now table stakes Regulatory Model is Driving Change FSS is Evolving to a Commodity Industry Technology Adoption is Accelerating to Meet Accelerating Business Needs Business Model Continues to Evolve Financial Services Sector (FSS) Challenges
- 5. Next Challenge – Heuristic Based Trading Market data feeds News feeds Social feeds Streaming news video Ticker Plant Heuristic Processing Heuristic Processing Heuristic Processing Heuristic Processing Historical Data Base Heuristic Processing Human Oversight and control Execution Venues Trade Plant
- 6. Agenda Redistribute Computing as a Solution The Limitations Today Automation in a distributed environment The Challenges
- 7. Slowing of Processor Speed Gigahertz
- 8. Latency Reduction Trends Industry tracking to ≈ 450ns
- 9. Agenda Redistribute Computing as a Solution The Limitations Today Automation in a distributed environment The Challenges
- 10. Customer Risk Management embedded in Data center Switch Symbol routing embedded in Data center Switch Data pre/post processing Customer 40% less ports Latency down 50µS Total Latency ≈100.0µS Trade Plant Size: ≈ 60 Servers ≈ 1000 Ports Total Latency ≈150.0µS Trade Plant Size: ≈ 100 Servers ≈ 1500 Ports A Different Approach: Distributed Computing for an Exchange Gateway Trade Plant
- 11. A Different Approach: Market Data Ticker Plant Internal Users Feed handler & publishing servers Pub/sub server farm Client subscriber servers World Wide Market Data Feeds Internal Users World Wide Market Data Feeds Feed handler & publishing in top of rack switch Pub/sub server farm Client subscriber imbedded in top of rack switch
- 12. Network Switch/Application Acceleration/Packet Forwarding Accelerator PFE Intel X86 , DDR3, SSD HYPERVISOR NETWORK OS ACTIVE NETWORK OS STANDBY GUEST APPLICATION(S) FPGA
- 13. At about 200 – 500NS for a reasonable switch Need to focus on a different approach Lower latency Eliminate servers Reduce network ports Application can be embedded in A VM in the switch A FPGA in the data path of the switch ASIC Packet forwarding engine (PFE) The Race to Zero is Ending Embed Application Snippets into the Switching Fabric Embed Snippets at the Control or Data Plane of the Network Compute-Integrated Networking The new way to reduce latency and cost.
- 14. • Processors and network switches are hitting natural limits • To achieve a high performance infrastructure compute resource must be distributed • Optimize computing in the server, NIC (FPGA), and data flow engines • Architect compute resources across the data center - not just one place Centralize Processing Where You Can, Distribute Where You Must Impact on Compute and Network
- 15. Agenda Redistribute Computing as a Solution The Limitations Today Automation in a distributed environment The Challenges
- 16. Automation Our networking Skills Transformation
- 17. Dimension Data today 2014 global revenues of USD 6.8 billion 73% of Global Fortune 100 and 59%of Global Fortune 500 are Dimension Data clients Client-centric, services-focused business Extensive experience in emerging markets Over 25,000 employees with operations in 54 countries across 5 regions Over 6,000 clients across all industry sectors
- 18. accelerate your ambition 18 Evolving Networking skills new products & services are software centric clients are becoming dev centric bus function abstracted from infrastructure through software engineering skillsets switching to dev Moving from configuration to policy management
- 19. Evolving Networking skills challenge Open APIs based on REST or other standards Vendor APIs Cisco OnePK Standard Protocols OpenFlow, Netconf Virtual Overlays VXLAN / NVGRE / STT Data Centre Controller WAN Controller Campus Controller Virtual Overlay Controller Control Plane Network services (FW, IPS, ADC, DDI, etc) Data Plane Automation ToolsOrchestration Tools Application Plane NWBU business of the future NWBU business today New world with new skills, new language, new approach
- 20. The Dimension Data Value proposition The ability to build SDN applications The experience working with OpenDaylight and OpenStack The consulting services to evolve IT operational models (business process) Transforming from engineering to development
- 21. Our Managed Cloud Platform network is SDN-enabled USD 20m SDN revenue in FY14 Contribution to and projects with Broad SDN project experience NSX trained engineers 30+ Investments in software development and lifecycle management USD 1.5m investment in multi-vendor SDN labs Our SDN capabilities have depth
- 22. SDN development model Strategy definition, assist with business case and roadmap Detailed Design & Migration Plan Production SDN Deployment Scale out pilot environments Evaluation Lab Working test environment (simulators, cloud labs, vendor POCs) SDN readiness assessment Applications, Integration, operations, migration Limited Scale Pilot and POC’s Dimension Data packaged POC service Dimension Data SDN Services
- 23. Lawrence Van Deusen II Director – Networking Business Unit 212 560 7973 Email: [email protected] Andy Bach Chief Architect for Financial Services, Juniper Networks 516-804-9500 Email: [email protected] Thank You