Data Center Megatrends Opportunity in 2015 Jim Henrys Chief Strategist Intel Corporation
A History of Collaboration Cloud, Big Data,. DC Modernization Unified Computing System Fabric Based Datacenter Networking Goes IA Routers, Switches, Security Appliances, Collaboration Verified for Centrino Business Class Wireless Suite 802.11n WiMAX OPA & IOT Wireless Connectivity EtherChannel Gigabit Ethernet InterSwitch Link Data Center Ethernet FCoE 1995 2005 2012 Ethernet Evolution Continuing Collaboration Driving Data Center Evolution
What s Going On?
Connected Devices: Depend on Data Centers SMARTPHONES ~400 DEVICES drives 1 SERVER 1 CONNECTED FACTORY TOOLS ~40 DEVICES drives 1 SERVER 2 MEDICAL WEARABLE DEVICES ~100 DEVICES drives 1 SERVER 2 DIGITAL SIGNS ~20 DEVICES drives 1 SERVER 2 We all truly live in an Application Economy now John Chambers, CEO, Cisco 1: Intel and 3rd party analysis 2: Intel estimate based on various end user proofs of concepts
Data Center Impact For Smart Phones Alone 1.9B Global Smartphones 26 Apps per Smartphone 20 Daily Transactions with Data Center per App 1 Trillion Daily Data Center Transactions *Sources: Google: Our Mobile Planet (26 apps per phone), Intel estimates
Industry Transformation to Hybrid Cloud Hybrid Cloud w/ Bursting Capability 30% Global Hybrid Cloud CAGR from 2013-2018 1 U U Private Cloud Workload optimized Control, Sensitive IP Desired DC Utilization Spikes Total Datacenter Workloads Bursty Workloads Delivered With Public Cloud Public Cloud Easy app migration Basic Security Guaranteed SLAs Scale/Demand Predictable App utilization (high) Base Base workloads delivered with efficient Private Cloud Limited / No SLAs Scale/Demand Unpredictable App utilization (low) 1 Source: TechNavio, 2014 Time 6
Summary Challenges for Data Center Architecture 2x 2x 2x 8x 50% Data Volume Every 18 Months Application Growth Every 2 Years Operational Costs Every 8 Years Increase in Management & Administration 1 Reduction in Compute Costs Every 2 Years More Capacity More Complexity More Resources Challenge: Decrease Time, Cost, Labor Increase Agility, Scalability, Performance 1 IDC Directions 14-2014 Source: Worldwide and Regional Public IT Cloud Services 2013-2017 Forecast. IDC (August 2013) http://www.idc.com/getdoc.jsp?containerid=242464
Data Center Inflection Point?
Today a Static, Manual, Hardware Defined Tomorrow a Automated, Dynamic, Software Defined Network EMAIL ONLINE SALES CRM ERP App App App App App App Compute ORCHESTRATION LAYER Storage Architectural Shift RESOURCE POOLS SOFTWARE DEFINED SERVERS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK
Software Defined Infrastructure Applications define the resources needed EMAIL ONLINE CRM ERP SALES ORCHESTRATION LAYER RESOURCE POOLS. Infrastructure assures efficiency and service levels SOFTWARE DEFINED SERVERS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK * Other names and brands may be claimed as the property of others.
Software Defined Infrastructure TCO 120% 100% 80% 60% 40% Relative cost per VM instance Up to 26% 1 Up to 66% 1 Orchestrated SDI lowering TCO of Data Center ORCHESTRATION: Reduces IT admin manual costs, and optimizes infrastructure SOFTWARE DEFINED NETWORK & SOFTWARE DEFINED STORAGE: Reduces infrastructure HW costs 20% 0% IT Datacenter Starting Virtualization Private Cloud Virtualized, automated, starting orchestration Private Cloud Full orchestration w/ SDI 2 Private cloud with no legacy burden in hybrid environment 1 Source: Intel Finance, 2014. Individual IT mileage may vary depending on workload, IT maturity and other factors. SDI assumes future benefits. Projections are extrapolations from Intel IT data. Private cloud model based on actual datacenter operations. IT DC based on Intel finance estimation for typical enterprise costs. Hybrid cloud model based on forward looking future benefits and market cost trends. Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance or cost.
Opportunity in the Journey to Hybrid Cloud SDI U Traditional DC Phase 1: Standardization Consolidate storage, network, compute Phase 2: Virtualization Resource pooling Server virtualization Basic automation Phase 3: Automation Patching / maintenance Provisioning Monitoring Reporting Upgrades Private Cloud Phase 4: Orchestration Policy-based workload movement SDN / SDS / NFV Automated resource pooling Hybrid Cloud Phase 5: Real-Time Enterprise Service managed SLA optimized Fully composable resources 1 2 3 Modernize & Load Balance Drive Cisco UCS, storage partners, 10 & 40GbE Automate and Optimize Drive UCS-director, Intelligent security and power policy automation, network virtualization, ACI/SDN Carefully Select Partners The hardware matters
1 Modernize: Aging Infrastructure is Inefficient and Costly 1* Source: Intel analysis, 2012 2* Source: Gartner, IT Metrics: Align IT Investment Levels With Strategy Using Run, Grow, Transform and Beyond (March 2012) Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance 32% of Servers are 32 >4 Years Old 1 Make up only 4% of 4 Total Performance Capabilities of Servers 1 % > 4 YEARS OLD % > 4 YEARS OLD > 4 YEARS OLD Use 65% of Total 65 % Energy Consumption 1 74 % OPE X 26 % CAPE X Business innovation throttled to 26% 2 Time to revenue Cost of lost time, effort, opportunity Unpredictable business cycles 74% captive in operations and maintenance 2 Rigid & aging infrastructure Application & information complexity Inflexible business processes Opportunity to help customers move to more efficient infrastructure on UCS
Re-Architecting the Data Center Software Defined Infrastructure Orchestration EMAIL ONLINE SALES CRM ERP Transform the Network ORCHESTRATION LAYER Infrastructure Attributes & Composable Architectures SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Unleash the Potential
Re-Architecting the Data Center Software Defined Infrastructure Orchestration EMAIL ONLINE SALES CRM ERP Transform the Network ORCHESTRATION LAYER Infrastructure Attributes & Composable Architectures SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Unleash the Potential
EMAIL ONLINE SALES CRM ERP. ORCHESTRATION LAYER Orchestration From Manual to Automated SOFTWARE DEFINED SOFTWARE DEFINED SERVERS STORAGE SOFTWARE DEFINED NETWORK * Other names and brands may be claimed as the property of others
Orchestration Focus : Broadest Enabled Ecosystem ONLINE EMAIL SALES CRM ERP. Policy driven workload and resource management to meet biz requirements Long history of enabling joint value SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Closing enterprise feature gaps High Availibility, Live Migration Capacity and Quota management Security and Compliance Ease of Install, Upgrade Delivered into OpenStack core code, working with partners Reference Architecture Juno Release, October 2014
Re-Architecting the Data Center Software Defined Infrastructure Orchestration EMAIL ONLINE SALES CRM ERP Transform the Network ORCHESTRATION LAYER Infrastructure Attributes & Composable Architectures SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Unleash the Potential
Real Time Intelligence, True Automation ONLINE SALES EMAIL CRM ERP ORCHESTRATION LAYER. The Hardware Matters Infrastructure Attributes Power Performance Security Thermals Utilization Location Latency SOFTWARE DEFINED SERVERS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK * Other names and brands may be claimed as the property of others.
The Hardware Matters 2005 2009 2011 2015 Paxville Nehalem Sandy Bridge Haswell Enable Virtualization Intel VT-x, VT-d Increased Memory Addressability Enable SDI Performance & Power Telemetry 40G Intel Ethernet NVMe SSDs
2 Automate : Intel Intelligent Power Node Manager Hardware-based engine to report & control server power Up to 40% more servers and performance per rack With enabled software, monitor & control server, rack and row power Up to 30% power optimization without performance impact Power Policy Limit Actual Usage Path to gradually eliminate IP power strips, saving ~$400 per rack Time 2015 Opportunity : Cisco UCS + Intel Intelligent Power Node Manager * Other names and brands may be claimed as the property of others.
2 Optimize: Disaggregated Server Architectures TODAY FUTURE RACK SCALE ARCHITECTURE Network Pooled and App App App App Compute disaggregated compute, network and storage App App resources. NVM NVM Storage Hardware attributes exposed to provisioning management layer Discreet Components Self Integration Composable Set of Pooled and Disaggregated Resources
2 Cisco UCS-M High density
3 Carefully select partners: Hardware matters in Cloud service delivery 70% 1 Of Enterprises Specifying HW In cloud Service Selection Application performance can vary up to 60% with up to 30% cost difference Security policy enforcement audit can be a challenge Sec 70 60 50 40 30 Job runtime in seconds 1 (lower is better) VM Instance with Intel Xeon E5 v2-2600 vs. Intel Xeon 5500 74% increase 1 Encryption performance Up to 5.2X Higher VM density 2 up to 65% Cost reduction 1 20 10 up to 52% Decrease in run-time 1 0 Lower Performing Premium Source: 1) IMR Research: Base: Total using Public Cloud US n=207; PRC n=241, Germany n=90, Brazil n=81 Q. What percent of these applications have you specified the hardware and/or other components to be used by the Cloud service provider as opposed to only specifying the level of service you expect? 2) Intellitrends Survey 2013 - (Comuware Report Measuring the Business Impact of Technology Performance) 3) HotCloud 2012 White Paper 1 NIBR, of Novartis Pharmaceuticals ran Next Gen Sequencing, Imaging & Modeling & Sim techniques (specifically Virtual Screening w/cpu bound, low mem, low IO, network, parallel Benchmarking SW ran same job many times (workload avg 32 secs on AWS cc2.8xlarge vs 1min 7 sec on AWS m1.large) revealed that best ROI was with cc2vspot instances Intel does not control or audit the design or implementation of third party benchmark data or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmark data are reported and confirm whether the referenced benchmark data are accurate and reflect performance of systems available for purchase. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance
Re-Architecting the Data Center Software Defined Infrastructure Orchestration EMAIL ONLINE SALES CRM ERP Transform the Network ORCHESTRATION LAYER Infrastructure Attributes & Composable Architectures SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Unleash the Potential
Network Transformation Stages 4:1 Workload Consolidation Network Virtualization Network Function Virtualization Software Defined Networking Speed development Maximize flexibility Overlays enable virtual networks on top of a physical network Flexible deployment models and placement Leverage virtualization Datacenter wide Automation & Flexibility Separates control and data plane
Cisco and Intel Driving Open Industry Collaboration Open Source Standards Open EcoSystem OPFLEX PROTOCOL + ECOSYSTEM APIC OPENSTACK & OPEN DAYLIGHT GROUP POLICY PROJECT OPEN VSWITCH & OPFLEX APIC OpFlex Policy Authority OPFLEX Opflex OpenFlow Affinity OpenFlow + OVSDB OpFlex Agent Hypervisor VM Open vswitch VM
Re-Architecting the Data Center Software Defined Infrastructure Orchestration EMAIL ONLINE SALES CRM ERP Transform the Network ORCHESTRATION LAYER Infrastructure Attributes & Composable Architectures SOFTWARE DEFINED SERVERS RESOURCE POOLS SOFTWARE DEFINED STORAGE SOFTWARE DEFINED NETWORK Unleash the Potential
50 Billion Devices 2020 35 Zettabytes The Era of Analytics *Sources: Cisco Internet Business Solutions Group, IDC: The Digital Universe
Intel Big Data Analytics Platform Strategy 2015 (CAGR%) $2.04B (56%) $394M (60%) $363M (60%) $1.25B (59%)
Era of Analytics: Solving problems that matter Personalized Healthcare With Michael J Fox Foundation, enabled research scientists to monitor progression of Parkinson s Disease by developing analytics platform for data collected from wearable sensors. Smart Building Enabled Daikin Applied to lower energy costs and predict maintenance schedule by analyzing data from HVAC rooftop units in the cloud and managing systems remotely. Responsive Store With Living Naturally, helped natural food retailers to optimize inventory in realtime by developing social media analytics solution on big data platform.
Your Opportunity in the Evolution to SDI Infonetics: New spending plans in datacenters 1) Sell UCS & UCS storage partners : Customers plan to grow DC server and storage capacity 2) Sell10&40GbE, Cisco Virtualized Network catalog: customers will upgrade their networking to support increasingly dense virtualization footprint 3) Start engaging on ACI and UCS trust and intelligent Power Management features: All respondents said they were exploring SDN & greater automation Datacenter & SDN strategies Infonetics October 2014
Summary The datacenter and the datacenter industry are transforming Underlying hardware matters in the transformation Modernize across the portfolio Sell UCS & UCS storage partners as your customers plan to grow DC server and storage capacity Sell10&40GbE, Cisco Virtualized Network catalog as customers increase virtualization footprint Explore high density form factors such as Cisco UCS-M Help customers automate and optimize their cloud footprint Start learning about the standards based approaches for SDN Deploy UCS trust and intelligent Power Management features - high interest in greater automation Carefully select cloud partners and premium instances - it offers your customers a better experience/roi Continuing Collaboration, Driving Datacenter Evolution 33