Sea Change Series: Scale in the Mega Data Center

Published: May, 2017 | Pages: 102 | Publisher: WinterGreen Research
Industry: ICT | Report Format: Electronic (PDF)

Scale is a vital part of the technology used to support next generation data centers. The study is targeted to C-level executives that need to move quickly and surely to improve IT. Automation of IT depends on understanding the business market opportunity from an independent perspective. Vendors are smart but they are committed to the technology they are pushing, the Sea Change Series from WinterGreen Research is able to provide a perspective not available anywhere else.

Extreme scale is what brings enough pathways inside a Mega data center to create a non-blocking (CLOS) networked server architecture. Non-blocking network architecture benefits the business because it permits launching thousands of virtual severs on demand at the application layer. In this manner, innovation can be made to happen quickly.

Using a mega data center, DevOps and/or automated processes can request and deploy additional resource without backing the Dell truck up to the data center every week to provide on-demand capacity. Automated deprovisioning handles freeing of surplus resources, while being virtual means not having stacks of surplus hardware to dispose of as underutilized capital assets.

Modern data centers are organized into processing nodes that manage different applications at a layer above infrastructure. Data is stored permanently and operated on in place. These are the two technologies to check for when choosing a data center. These architectural features provide economies of scale that greatly reduce the IT spend while offering better quality IT.

Once scale is in place, then the economies of scale kick in. When negotiating for cloud capability, managers need to check to see that sufficient multiple pathways are available to reach any node in a non-blocking manner. Non-blocking architecture is more efficient than other IT infrastructure and supports better innovation for apps and smart digitization. Not all cloud architectures offer this business benefit.

The Cloud 2.0 mega data center platform fabric technologies support the digital economy by creating scale with a network internal to the mega data center system image. Without scale, there are not enough pathways inside the data center to enable elimination of bottlenecks. Access to every node in the fabric, and multiple duplicate pathways to every node are needed to enable real time application connectivity. With sufficient scale, if one pathway is blocked, there are enough other pathways to get to the desired node resources.

The theme of this study is that scale matters.

Scale can be implemented by IT, but the executive needs to understand that there is a difference between different technologies. In a mega data center, the system is implemented as a fabric: servers are linked to top of rack switches, which are made from merchant silicon chips, mostly Broadcom, some switch ASICs. A pod of server racks are linked to each other through an edge aggregation switch. A pod of server racks is based on the same ASICs as in the top of rack switches.

The importance of nonblocking architecture is compelling. Aggregation switches are lashed together through a set of non-blocking spine switches. All switches are based on the same chip. This is precisely the way Facebook is building its own Wedge and 6-pack open switches– nine years after Google did it. The amazing thing is that Facebook had not done this already.

The four superstar companies that are able to leverage IT to achieve growth, Microsoft, Google, Facebook, and the leader AWS all use Clos architecture. What is significant is that systems have to hit a certain scale before Clos networks work Clos networks are what work now for flexibility and supporting innovation in an affordable manner. There is no dipping your toe in to try the system to see if it will work, it will not and then the IT says, “We tried that, we failed,” but what the executive needs to understand is that scale matters. A little mega data center does not exist. Only scale works.

Maybe scale is not the only answer, maybe in 20 years, Quantum computing will bring a new data center system, but for now, Clos architecture and scale dominate those IT centers that have the strongest growth engine.

Business leaders are challenged to move their enterprises to the next level of competition. An effective digital business player, transformer, and disruptor position depends on the effectiveness of employing digital technologies and leveraging connected digital systems. Organizational, operational, and business model innovation are needed to create ways of operating and growing the business using mega data center cloud technologies, systems are evolving. It is a journey to achieve the connected enterprise, ultimately connecting all employees and a trillion connected devices.

Many companies are using digital technology to create market disruption. Amazon, Uber, Google, IBM, and Microsoft represent companies using effective disruptive strategic positioning. As entire industries shift to the digital world, once buoyant companies are threatened with disappearing. A digital transformation represents an approach that enables organizations to drive changes in their business models and ecosystems leveraging cloud computing, and not just hyperscale systems but leveraging mega data centers. Just as robots make work more automated, so also cloud based communications systems implement the IoT digital connectivity transformation.

Disruption in the business markets represents major opportunity for vendors with cloud offerings. This is part of a larger digital transformation, a digital approach to interconnecting everything that enables organizations to drive changes in their business models and ecosystems.

Disruptive cloud systems are provided by Amazon (AWS), Microsoft, Google, and Facebook. Data centers are in a class by themselves, they have functioning fully automatic, self-healing, networked mega datacenters that operate at fiber optic speeds to create a fabric that can access any node in any particular data center because there are multiple pathways to every node. In this manner, they automate applications integration for any data in the mega data center.

By leveraging digital competencies, businesses can grow faster than they would otherwise. A digital strategy, in conjunction with the appropriate unified communications solution permits the implementation of innovative communications services. Digital connectivity with combined voice, video and file transfer can help organizations and their end users innovate and compete more effectively. It is imperative that organizations have a digital communications strategy in place.

This is an era where the distinction between the technologies and processes that businesses deploy is tightly linked. Digital technology directly impacts customers and markets. The boundary between internal operations of the enterprise and its external ecosystem is rapidly disappearing. Customers, markets, competitors, partners, and regulators are inextricably linked.

According to Susan Eustis, lead author of the team that prepared the study, “Mega data centers need to be understood by all senior executives whether they move in that direction or not. These are the IT used by the fastest growing organizations Google, AWS, Microsoft, and Facebook. There are 25 Sea Change Data Center study modules describing different aspects of the move to mega data centers. The Scale module describes that it is not sufficient just to try certain cloud techniques.

“Scale is an essential aspect of the data center positioning for these leading companies. These companies use Clos networks as their data center implementations. This module addresses how and why scale in the mega data center is important. The market shift to non-blocking network inside data center building means companies have to hit a certain scale before Clos networks work.”

WinterGreen Research is an independent research organization funded by the sale of market research studies all over the world and by the implementation of ROI models that are used to calculate the total cost of ownership of equipment, services, and software. The company has 35 distributors worldwide, including Global Information Info Shop, Market Research.com, Research and Markets, electronics.ca, and Thompson Financial. It conducts its business with integrity.

The increasingly global nature of science, technology and engineering is a reflection of the implementation of the globally integrated enterprise. Customers trust wintergreen research to work alongside them to ensure the success of the participation in a particular market segment.

WinterGreen Research supports various market segment programs; provides trusted technical services to the marketing departments. It carries out accurate market share and forecast analysis services for a range of commercial and government customers globally. These are all vital market research support solutions requiring trust and integrity.

Companies Profiled

Market Leaders
• Amazon
• Microsoft
• Google
• Facebook

Key Topics

• Scale In The Mega Data Center
• Realign IT Cost Structure
• Mega Datacenter Physical Infrastructure
• Automation of Mega Data Center
• Networking Fabric
• Exchange Of Data Between Servers
• Complex Automation Of Process
• Applications Customized For Each User
• Machine-To-Machine Management of Traffic Growth
• Fabric Network Topology
• Building-Wide Connectivity
• Highly Modular Data Cebter Design
• Scale Capacity
• Back-End Service Tiers
• Applications Scaling
• Mega Data Center Network
• Fabric Next-Generation Data Center Network Design
• Pod Unit of Network
• Mega Data Center Server Pods
• Non-Blocking Network Architecture
• Data Center Auto Discovery
• Large-Scale Network
• Rapid Deployment Architecture
• Expedites Provisioning And Changes
• Programmable Access To Network Stack
• Software Defined Networking (SDN)-Supports Scale and Automation
• Compute Engine Load Balancing
• Load Balanced Requests Architecture
• Scale-Out: Server And Storage Expansion
• Switches and Routers Deployed in Fabrics
• Mega Data Center Multipathing
• Routing Destinations
• Clos Topology Network
• Capacity Scalability
• Aggregation Switches
• Intelligent Cloud Platform
• Linux For Azure

 Table of Contents

Facebook Mega Datacenter Physical Infrastructure 13
Facebook Automation of Mega Data Center Process 14
Facebook Altoona Data Center Networking Fabric 15
Facebook Altoona Cloud Mega Data Center 16
Facebook Altoona Data Center Innovative Networking Fabric Depends on Scale 17
Facebook Fabric Operates Inside the Data Center 18
Facebook Fabric 19
Exchange Of Data Between Servers Represents A Complex Automation Of Process 20
Applications Customized For Each User 21
Machine-To-Machine Management of Traffic Growth 22
Facebook Data Center Fabric Network Topology 23
Building-Wide Connectivity 24
Highly Modular Design Allows Users To Quickly Scale Capacity In Any Dimension 25
Back-End Service Tiers And Applications 26
Scaling Up As a Basic Function Of The Mega Data Center Network 27
Facebook Fabric Next-Generation Data Center Network Design: Pod Unit of Network 28
Mega Data Center Server Pods 29
Facebook Sample Pod: Unit of Network 30
Non-Blocking Network Architecture 31
Data Center Auto Discovery 36
Facebook Large-Scale Network 37
Rapid Deployment Architecture 38
Facebook Expedites Provisioning And Changes 39
Google Douglas County Mega Data Center 40
Google Data Center Efficiency Measurements 41
Google Programmable Access To Network Stack 42
Google Software Defined Networking (SDN)-Supports Scale and Automation 43
Google Compute Engine Load Balancing 44
Google Compute Engine Load Balanced Requests Architecture 45
Google Compute Engine Load Balancing Scaling 46
Google Switches Provide Scale-Out: Server and Storage Expansion 47
Google Uses Switches and Routers Deployed in Fabrics 48
Google Mega Data Center Multi-pathing 49
Google Mega Data Center Multi-pathing: Routing Destinations 50
Google Clos Topology Network Capacity Scalability 51
Google Aggregation Switches Are Lashed Together Through a Set Of
Non-Blocking Spine Switches 52
Google Network Called Jupiter 53
Mega Data Center Scale: Non-Blocking Network Inside Building
Microsoft Cloud Data Center Multi-Tenant Containers 54
Microsoft Azure Running Docker Containers 55
Microsoft Data Center, Dublin, 550,000 Sf 56
Microsoft Builds Intelligent Cloud Platform 57
Microsoft Crafts Homegrown Linux For Azure Switches 59
Microsoft Azure Has Scale 60
Microsoft Azure Stack Hardware Foundation 62
Microsoft Azure Stack Key Systems Partners: Cisco Systems, Lenovo, Fujitsu,
and NEC 63
Microsoft Gradual Transformation From A Platform Cloud To A Broader Offering Leveraging Economies of Scale 64
Microsoft Contributing to Open Systems 65
Microsoft Mega Data Center Supply Chain 66
Microsoft Leverages Open Compute Project to Bring Benefit to
Enterprise Customers 67
Microsoft Assists Open Compute to Close The Loop On The Hardware Side 68
Microsoft Project Olympus Modular And Flexible 69
Microsoft Azure 70
Microsoft Azure Active Directory Has Synchronization 71
Microsoft Azure Has Scale 72
Mega Data Center Different from the Hyperscale Cloud 73
Mega Data Center Scaling 74
Mega Data Center Automatic Rules and Push-Button Actions 75
Amazon Capex for Cloud 2.0 Mega Data Centers 76
AWS Server Scale 77
Amazon North America 78
Innovation a Core Effort for Amazon 80
Amazon Offers the Richest Services Set 81
AWS Server Scale 81
On AWS, Customers Architect Their Applications 82
AWS Scale to Address Network Bottleneck 83
Networking A Concern for AWS Solved by Scale 84
AWS Regions and Network Scale 85
AWS Datacenter Bandwidth 88
Amazon (AWS) Regional Data Center 89
Map of Amazon Web Service Global Infrastructure 90
Rows of Servers Inside an Amazon (AWS) Data Center 91
Amazon Capex for Mega Data Centers 92
Mega Data Center Scale: Non-Blocking Network Inside Building
Amazon Addresses Enterprise Cloud Market, Partnering With VMware 92
Making Individual Circuits And Devices Unimportant Is A Primary Aim Of Fabric Architecture 93
Google Clos Network Architecture Topology Allows the Building a
Non-Blocking Network Using Small Switches 94
You Have To Hit A Certain Scale Before Clos Networks Work 95
Clos Network 96
Digital Data Expanding Exponentially, Global IP Traffic Passes
Zettabyte (1000 Exabytes) Threshold 99
Summary: Economies of Scale 100
WinterGreen Research, 101
WinterGreen Research Methodology 102

List of Figures

Enterprise Data Center as a Bottleneck: Scale Supports
Non Blocking Network Inside Building and More Efficient Processing
Figure 1. Slow Growth Companies Do Not Have Data Center Scale 2
Figure 2. Mega Data Center Fabric Implementation 3
Figure 3. Facebook Schematic Fabric-Optimized Datacenter Physical Topology 13
Figure 4. Facebook Automation of Mega Data Center Process 14
Figure 5. Facebook Altoona Positioning Of Global Infrastructure 15
Figure 6. FaceBook Equal Performance Paths Between Servers 16
Figure 7. FaceBook Data Center Fabric Depends on Scale 17
Figure 8. Facebook Fabric Operates Inside the Data Center,
Fabric Is The Whole Data Center 18
Figure 9. Fabric Switches and Top of Rack Switches, Facebook Took a Disaggregated Approach 19
Figure 10. Exchange Of Data Between Servers Represents A
Complex Automation Of Process 20
Figure 11. Samsung Galaxy J3 21
Figure 12. Facebook Back-End Service Tiers And Applications Account for Machine-To-Machine Traffic Growth 22
Figure 1. Facebook Data Center Fabric Network Topology 23
Figure 13. Implementing building-wide connectivity 24
Figure 14. Modular Design Allows Users To Quickly Scale Capacity In Any Dimension 25
Figure 15. Facebook Back-End Service Tiers And Applications Functions 26
Figure 16. Using Fabric to Scale Capacity 27
Figure 17. Facebook Fabric: Pod Unit of Network 28
Figure 18. Server Pods Permit An Architecture Able To
Implement Uniform High-Performance Connectivity 29
Figure 19. Non-Blocking Network Architecture 31
Figure 20. Facebook Automation of Cloud 2.0 Mega Data Center Process 32
Figure 21. Facebook Creating a Modular Cloud 2.0 mega data center Solution 33
Figure 22. Facebook Cloud 2.0 mega data center Fabric High-Level Settings Components 34
Figure 23. Facebook Mega Data Center Fabric Unattended Mode 35
Figure 24. Facebook Data Center Auto Discovery Functions 36
Figure 25. Facebook Automated Process Rapid Deployment Architecture 38
Figure 26. Google Douglas County Cloud 2.0 Mega Data Center 40
Figure 27. Google Data Center Efficiency Measurements 41
Figure 28. Google Andromeda Cloud High-Level Architecture 42
Figure 29. Google Andromeda Software Defined Networking (SDN)-Based Substrate Functions 43
Figure 30. Google Compute Engine Load Balancing Functions 44
Figure 31. Google Compute Engine Load Balanced Requests Architecture 45
Figure 32. Google Compute Engine Load Balancing Scaling 46
Figure 33. Google Traffic Generated by Data Center Servers 47
Figure 34. Google Mega Data Center Multi-pathing: Implementing Lots And Lots Of Paths Between Each Source And Destination 49
Figure 35. Google Mega Data Center Multi-pathing: Routing Destinations 50
Figure 36. Google Builds Own Network Switches And Software 50
Figure 37. Google Clos Topology Network Capacity Scalability 51
Figure 38. Schematic Fabric-Optimized Facebook Datacenter Physical Topology52
Figure 39. Google Jupiter Network Delivers 1.3 Pb/Sec Of Aggregate Bisection Bandwidth Across A Datacenter 53
Figure 40. Microsoft Azure Cloud Software Stack Hyper-V hypervisor 54
Figure 41. Microsoft Azure Running Docker Containers 55
Figure 42. Microsoft Data Center, Dublin, 550,000 Sf 56
Figure 43. Microsoft-Azure-Stack-Block-Diagram 60
Figure 44. Microsoft Azure Stack Architecture 62
Figure 45. Microsoft Data Centers 66
Figure 46. Microsoft Open Hardware Design: Project Olympus 67
Figure 47. Microsoft Open Compute Closes That Loop On The Hardware Side 68
Figure 48. Microsoft Olympus Product 69
Figure 49. Microsoft Azure Has Scale 72
Figure 50. Mega Data Center Cloud vs. Hyperscale Cloud 73
Figure 51. Amazon Web Services 76
Figure 52. Amazon North America Map 78
Figure 53. Amazon North America List of Locations 79
Figure 54. Woods Hole Bottleneck: Google Addresses Network Bottleneck with Scale 83
Figure 55. Example of AWS Region 85
Figure 56. Example of AWS Availability Zone 86
Figure 57. Example of AWS Data Center 87
Figure 58. AWS Network Latency and Variability 88
Figure 59. Amazon (AWS) Regional Data Center 89
Figure 60. A Map of Amazon Web Service Global Infrastructure 90
Figure 61. Rows of Servers Inside an Amazon (AWS) Data Center 91
Figure 62. Clos Network 96
Figure 63. Data Center Technology Shifting 97
Figure 64. Data Center Technology Shift 98