1. Cisco Quick Hit Briefing
Brian J. Avery, Territory Business Manager - Cisco Systems
Max out your Sales with UCS Mini
This session was recorded via Cisco WebEx! You can watch
the live session recording via the following URL:
https://new-webex.webex.com/new-
webex/lsr.php?RCID=6b77f2b9e20f4fc2ba881f3e144a9b04
Thanks for your interest and participation!
2. Cisco Quick Hit Briefing
Brian J. Avery, Territory Business Manager - Cisco Systems
Max out your Sales with UCS Mini
Welcome! Connect using the
audio conference box or by
calling into the meeting:
1. Toll-Free: (866) 432-9903
2. Enter Meeting ID: 205 383 827
3. Press “1” to join the conference.
7. 7
March 2009: Cisco Unified Computing System
"A year from now the difference will be UCS is dead
and we have had phenomenal market share growth in
the networking space" – Randy Seidl, HP
“Cisco deserves a lot of credit
for its industry chutzpah”
"The tough part here is that the server buyer has no relationship with Cisco.
And they don't know why they need one.” -Forrester Research
9. 9
Cisco UCS: The Right Solution at the Right Time
CISCO
UCS
Virtualization
Network&
StorageAccess
Compute
Application
Centricity
Operational
Simplicity
Platform for
IT Innovation
Reduce the Complexity that Drives OPEX
Get the Most Out of Virtualization
Automate and Move Faster
Get ready for cloud
Customer Asks in 2009
Help me:
10. 10
1IDC Worldwide Quarterly Server Tracker, Q1 2013 Revenue Share, May 2013
Real Innovation Channel Acceleration
Cisco UCS: After Five Short Years
Technology Partnerships
“Leader” in Gartner Magic
Quadrant for Blade Servers
#2 in World Wide x86
Blade Server Market1
23,000
UCS
Customers 5400
900 38,800
2010 2011 2012 2013 2014
2009
3850 Partners
90% of $2B+ UCS Business
Growing Partner Data Center
Practices
Simplify the Total Solution
Established and Emerging Players
Integrated and Converged
Infrastructures
Focused on Customer OPEX
Application Centric vs. HW Centric
ASIC-Level Virtualization Technology
Growth Drivers
21. 21
Edge-Scale Computing Needs
Central IT
Customer Needs
› Computing proximity for IoE and fog computing
› Comprehensive remote management at global scale
› Small footprint: space, power, and cooling
› Consistent configuration and policy enforcement
› Hardware separation (compliance and security)
Enabling Small-Scale IT
At the Edge of Large IT
Customer Needs
› “No assembly required” total computing solution
› Simplified systems management
› 1-15 servers
Small and Medium
Organizations Test/Dev
Remote Sites Customer Premise
Branch Offices
23. 23
Cisco UCS Mini: Enterprise Capabilities at Edge Scale
Unified computing in 6RU
› Chassis-integrated
fabric interconnects
› Cisco UCS Manager
› Standard Cisco UCS
blades, fans, and power
supplies
Connect up
to 7 Cisco UCS C-Series Rack
Servers for expanded capacity
Unified computing in 6RU:
› Chassis-integrated
fabric interconnects
› Cisco UCS Manager
› Standard Cisco UCS blades, fans,
and power supplies
New architectural entry point for
unified computing at 1 to 15 server
scale
Full-power Cisco UCS® in all-in-one
package:
› Computing › Networking
› Cisco UCS Manager
24. 24
Cisco UCS Mini: At the Edge of Large IT Configurations
Central IT
Cisco UCS® Mini
Cisco
UCS Mini
Cisco
UCS Mini
UCS Management
Cisco UCS Mini
25. 25
Cisco UCS Mini: Delivering the Power of Unified Computing
at the Edge
› Simple
− Allows for consolidation of multiple servers in a single managed
solution
− All hardware is managed from a single interface
− Cisco UCS® Central Software manages data centers and remote
sites as one
› Efficient
− Unified systems management
− Highly-availability system for the most uptime
− Reduced capital and operational expenses
with consolidation and virtualization
› Expandable
− Extensive scalability within a stable platform
− Support for future Cisco UCS blades
26. 26
Cisco UCS Mini: Enterprise Capabilities at the Edge
› Computing and storage integration
− 6RU rack-mount chassis
− Up to 8 blade servers
− Up to 7 rack servers
› Integrated networking
− SFP+ for 1- or 10-Gbps connectivity
− Hassle–free connectivity with end-host mode
› High availability
− Hot-pluggable redundant power (N+N, N+1, and N support)
− 100 to 120V, 200 to 240V, -48V support
− Redundant network connectivity
− Redundant fans
› Full systems management capabilities
− Full-featured Cisco UCS® Manager
− Cisco UCS Central Software for remote management
27. 27
Cisco UCS Mini: Server Support
› No-compromise computing capability
› Cisco UCS® B200 M3 Blade Server: feature-rich,
2-socket, half-width server
− Support for all CPU configurations
− Up to 768 GB of DDR memory
− Up to 40-Gbps bandwidth per blade in Cisco UCS Mini
28. 28
Cisco UCS Mini: Simplified Networking Support
› Easily connect to upstream networks with
end-host mode
› SFP+ supports 1 and 10 Gigabit Ethernet for
future proofing
› Redundant networking capability built in
› Simple expansion for additional computing power
29. 29
Cisco UCS Mini: Storage Connectivity Options
Direct-Attach Storage
ETH 1 ETH 2
iSCSI/NAS iSCSI/NAS
iSCSI/NAS/Fibre
Channel/FCoE
iSCSI/NAS/Fibre
Channel/FCoE
ETH 1 ETH 2
Network-Attached Storage
30. 30
Cisco UCS Mini: Management Simplicity
API API
Basic Management Functionality
API
API
Cisco UCS® Director API
Stand-Alone Cisco
UCS C-Series
Rack Server Cisco Unified Computing
System™
Cisco UCS Central Software
Cisco UCS Director
Cisco®
IMC
Cisco UCS Manager
Domain 1
Advanced Infrastructure Abstraction and Automation
Storage
Virtual Machines
Network
Devices
Servers
Third-Party
Infrastructure
API
Cisco UCS Manager
Domain 1
Cisco UCS Mini
Domain 1
› Single-office locations
− Cisco UCS Manager converges
management of servers, storage,
and networking
− Full-featured management with no
limitations
› Data centers
− Take advantage of the full Cisco
UCS management suite
− Integrated with third-party
ecosystems (Microsoft, VMware,
CA, IBM)
− Common management of Cisco
UCS in the data center and Cisco
UCS Mini at remote offices
31. 31
Cisco UCS or Cisco UCS Mini: Main Component
Differences
Cisco UCS® 5108 Blade Server
Chassis with new backplane
(UCSB-5108-AC2, etc.)
+ = Cisco
UCS
Cisco UCS 2204XP
Cisco UCS 2208XP
Cisco UCS 6248
Fabric Interconnect
Cisco UCS 6296
Fabric Interconnect
+
+
Cisco
UCS
Mini
Cisco UCS 5108 Blade
Server Chassis With
new backplane
(UCSB-5108-AC2, etc.)
+
New PSU at 200 to
240V (UCSB-PSU-
2500ACDV)
Cisco UCS
6324 Fabric
Interconnect for
Cisco UCS Mini
(UCS-FI-M-6324)
=
Dual-voltage PSU
(UCSB-PSU-
2500ACDV)
+
35. 35
Cisco UCS Mini Use Case : A Growing Coffee Roaster
Problem
› Need to reduce floor space
› Need an IT platform that will expand
› Need low-cost redundant storage expansion
› Require full-featured blades to run virtual
desktop infrastructure (VDI) for floor staff
› Want a service provider to remotely manage
hardware
Solution: Cisco UCS® Mini with Storage on Cisco UCS C240 M3
Rack Server with StorMagic SvSAN
› Servers consolidated into a single chassis, reducing floor space
› Easy computing and storage scalability: hot-pluggable in new blades
and drives
› Large, shared storage capacity and exceptional performance improve client
service, without the cost and complexity of a SAN
› Cisco UCS B200 M3 blades provide the computing power for VDI and order
management applications
› Cisco® Call Home and Cisco UCS Central Software allows remote monitoring
and trouble-ticket issuing
› Scenario: A growing coffee roaster continues to add new customers for its custom coffee roast and needs to add new
staff and a more flexible IT solution that supports the growing business. The company needs to improve communication
between the front office and the roasting team as well as be able to run its order management system more efficiently.
36. 36
Cisco UCS Mini Use Case: Retail Store Upgrade
Requirements
› Standardize on a single computing platform in
stores and data center
› Provide high availability to in-store
applications, reducing costly downtime
› Need local access to customer data for in-store
orders
› Manage store environment from central IT
without requiring local IT staff
› Reduce downtime from remote system issues
Solution: Cisco UCS Mini
› Standardize on Cisco UCS B200 M3 blades for data center and store
environments
› Direct connect existing in-store Fibre Channel storage to Cisco UCS Mini,
eliminating in-store Fibre Channel switch maintenance
› Standardize deployment of store configurations from Cisco UCS Central
Software
› Cisco UCS Central Software allows central IT to monitor and manage Cisco
UCS Mini remotely
› Common cisco UCS Manager, Cisco® IMC, and Cisco UCS Central Software
management of all server infrastructure
› Scenario: A retailer needs to standardize its in-store IT infrastructure to reduce costs. The store already has Cisco UCS®
installed in its data center and would like to expand it to the 200 stores, but it has not been cost effective to date.
Let’s take a look at the root causes of this maintenance predicament and some other recent cost trends you might be seeing in your data center.
Today, the data center makes up 44% of overall IT spend. The traditional model of data center investment was to build in silos—often with dedicated resources for lines of business—and to design them for peak environments. Unfortunately, this model had some inherent inefficiencies because resources couldn’t be shared and, for the majority of the time, applications weren’t maximizing the capability of the server platform, which drove up costs via underutilization of resources
A few other things jump out. First, more than 50% of the cost is the combination of people and software, with energy and facilities as the next largest chunk (contributing no new value-add to the IT department—it’s required to simply keep things running). One may come to the conclusion that infrastructure doesn’t matter that much, but when you actually analyze the root cause of the spiraling costs in people and software, one quickly realizes that they are related to the infrastructure – and the software to manage it.
For instance, if we double click on servers, while the cost of hardware has been flat or declining for the past 15 years, the cost of managing and operating servers has been growing steadily and now represents 2/3 of the total spend associated with servers. The problem has been exacerbated by virtualization because higher server utilization has come at the price of increasing complexity that shows in the rising Opex costs.
Before we do that let’s look at some of the trends going on within CapEx and OpEx at the bar chart. Note that server and power and cooling costs seem to be remaining steady. Even with the inherent efficiencies of server virtualization—companies continue to need to invest in servers because of the growth in project demand they continue to experience.
But where increase pain is really being felt is in the area of management and administration costs associated with your virtual servers. This stems from the problem of the traditional approach to management in so many IT organizations. Because a wide variety of traditional servers, software systems, storage have been utilized, and implemented in silos—most system management functions are complicated—designed to try to integrate solutions that were never designed to work together. Virtualization, of course, means there is increasing movement of virtual machines—and the need to ensure key information and adequate security moves with them. Without a simpler approach to management, administration costs will continue to escalate.
Key Elements:
· Containers: deploy multiple applications in a common resource pool
· 30% of Hadoop problems are configuration related: automate!
· Configure ACI network and services along with compute and storage
Our strategy for infrastructure management addresses the reality that IT professionals must operate the entire infrastructure – bare metal and virtual – to offer their services reliably and at scale. Solutions that address only virtualization place the onus on IT to figure out the rest on their own – often manually. Our approach delivers greater business agility, enabling IT to deploy applications faster, while reducing opex by making administrators more efficient, and operations more reliable.
UCS Director already manages compute, network, storage, and virtualization and supports both Cisco and third-party infrastructure products. Today, we are announcing several key advancements:
Application containers provide IT with the confidence and security to run multiple workloads on the same resource pool in an enterprise environment. Resource pooling is fundamental to cloud operating models and flexible resource allocation to increase data center utilization. It is about delivering the right resources, on-demand, via a programmable API, as required by the application. However, customers need a reliable method of ensuring that applications running within the resource pool do not conflict with each other. Our new containers capability provides both virtual and bare metal isolation of workloads, automated with UCS Director. Users simply request a container for a particular set of application components and UCS Director deploys the necessary infrastructure configuration required.
Data analytics and Hadoop in particular are huge interest areas for our customers. Cisco UCS has seen strong demand with Hadoop clusters in the 100’s of nodes. The challenge on infrastructure is how to get these large clusters to scale, be easily managed and at reasonable cost. Our enterprise customers care about simplicity of deployment and scale. UCS Director works directly with major Hadoop distributions and the integrated infrastructure to ensure consistent cluster configuration, relieving customers from manual operations.
UCS Director now supports Cisco’s new Application centric infrastructure and Nexus 9000 product family to seamlessly deploy and configure broader network and services along with compute and storage. UCS Director integrates the software programmable ACI fabric including L4-L7 services along with UCS, storage and virtualization. Users can deploy applications and ensure the ACI fabric adapts to meet the connectivity requirements including quality of service, bandwidth and failover.
With its focus on infrastructure automation and management, UCS Director is delivering a software development kit and open API to accelerate third party integration with Cisco’s technology partners. UCS has successfully followed an open API strategy, resulting in broad ecosystem integratoin and adoption. Public announcements of broader third-party device support leveraging UCS Director’s new API will occur later this year.
<<CLICK 1>> CORE DATA CENTER WORKLOADS
First, building on the existing UCS portfolio that powers core data center workloads, we are announcing the fourth generation of our workhorse two socket blade and rack servers as well as broad set of new functionality for UCS Management.
<<CLICK 2>> EDGE SCALE
Second, we are expanding the scope of UCS to the realm of Edge-scale computing, with UCS Mini
<<CLICK 3>> CLOUD SCALE
Third, we are taking the proven operational advantages of UCS, as well as a groundbreaking new approach that revolutionizes server architecture, into the Cloud-scale computing domain…..with UCS M-Series Modular servers and the new C3000 series server.
Crucially, all of these new capabilities are components of the Unified, extensible architecture that is the heart of UCS.
As we enter these new categories and segments we bring customer-proven operational advantages that are the hallmark of UCS: simplicity, automation and an application centric approach to data center infrastructure.
Now more than ever, with this expanded portfolio, our customers have the capability to deliver infrastructure at the right location, at the right scale and in the right configuration to meet the unique needs of applications across the edge, the core, and into cloud-scale
Key Elements:
Bringing Power of Core Data Center to Edge; IoT, bringing compute closer to data collection points
Power of UCS Manager, policy and control to the edge where people resources limited
Can scale compute, storage I/O to support demands of evolving applications at the edge
And put it into a new form-factor. [click]
Embedded in the chassis
Saves both CapEx and OpEx
And put it into a new form-factor. [click]
And embedded it directly in the chassis. This delivers all the power and functionality of Unified Computing for up to 15 blade and rack servers.
New Competitive Advantages to Highlight:
we can support high end systems, performance is a strength, system can be remote, but high performing
UCS Managed, UCS operating environment, programmatic, stateless computing, policy based management
We have the capacity to scale higher, greater amount of compute, bandwidth, given the way that we have allowed Mini to scale, connecting the two chassis and connecting rack server to the mini domain (For example Nvidia virtualized GPU, optimized for virtual desktop,)
30
1. Introduction Slide- PlaceHolder
See notes next slide
The scenarios are anecdotal – real world examples will follow after FCS
The scenarios are anecdotal – real world examples will follow after FCS
Confidence Monitor Prompts:
“Retail is Detail” – James Gulliver. Physical+digital experience.
VDI Tablets, Video streaming & signage, location services & log data
Repoint the boot target for hundreds of application servers
Script:
Length = 65 sec.
James Gulliver, one of the UK’s retail pioneers, coined the well-known dictum “Retail is Detail.” Today, detail and data are exploding in shopping malls and retail chains as physical and digital experiences converge, driving on-site demand for data center grade computing like UCS.
Traditional Point of Sale terminals are being surpassed by Tablet-based applications that prompt sales people to stock a dressing room with a customers online selections and greet them by name as location services sense their arrival. These VDI implementations and mobile application services require server performance delivered by the B-Series blades of UCS mini.
Meanwhile, video is being streamed in multiple directions as rich media signage is triggered by the presence of the customer and video data is being recorded and analyzed by security and theft detection systems. Rack mount servers attached to UCS Mini provide capacity and performance required by these applications.
Application servers power customer wayfinding, location-based app experiences and personalized offers. In turn massive amounts of log data are being collected and analyzed on customer traffic patterns, density and dwell time. Distilled information is then being pushed back to the data center core for global analytics.
UCS Management allows central IT to make changes to this type infrastructure, such as changing the boot target of a specific class of application servers, across dozens or hundreds of stores, simply by updating a global service profile setting.
The retail environment is a good example of how UCS Mini seamlessly extends data-center grade computing power and IT operations from Core to Edge. These types of local computing demands are being mirrored in hotels and convention centers, airports, hospitals, campuses and sites across many industry verticals.
Updated Script from Arnab:
Business Challenge
Lack of onsite IT staff
Standardization and consistency across 100s maybe 1000s of locations - compliance and troubleshooting
UCS Benefits would include
Datacenter class remote management that overcomes the lack of local IT staff
Cutting down time to provision and upgrade from hours for each location to just a few minutes
Policy driven standardization that improves operational consistency and compliance (Payment Card Industry, PCI compliance)
Use case script:
There are growing number of services that require local computing in a retail environment. These could be POS, location based services, inventory control, interactive sinage, local trend analysis. As more and more aspects of fast IT get deployed at the enterprise edge these remote sites need to become more agile, better monitored and their need for standardization and compliance grows. One of the main challenges in these remote environments is the lack of local IT staff to manage the infrastructure there.
UCS brings fully centralized datacenter class management, that is currently being used by 36,000 datacenter customers, to your remote sites. Even if you overcame the of lack of local IT staff, it usually takes hours to provision, upgrade, maintain and deploy new services, UCS brings the time required to do it to minutes. Finally UCS allows you to drive standardization and compliance by managing all the sites from a single policy engine.
Example:
A leading consumer electronic company that deploys standalone rack servers in its stores spends avg $2000 per store when they need to upgrade the firmware of the in-store computing infrastructure. At more than 400 stores worldwide, that’s almost a $1M every time they need to do something simple like “upgrade”. That amount doesn’t take into account the logistical challenge of sending skilled admins into all those stores, the hours of downtime, the possibility of human errors and the challenge for standardizing. Bottom line, their enterprise edge which is one of their primary interface points to their own customer is not very dynamic.
With UCS management these upgrades would happened in a matter of minutes without requiring local IT staff and according to policy that would be consistent across all the stores. Initial provisioning, deployment of new services and continuous monitoring would also be as quick. In other words, your enterprise edge would become as dynamic, agile and standardized as your datacenter.
CDW-G AM’s have great resources to help close a UCS deal. Use them.