Document Sample
data_center_ecsp_for_transcription Powered By Docstoc
					                         Module: <Presentation Name>

Slide 2 of 23

   Extreme Networks has developed a network infrastructure solution that will
   accommodate the evolving technology landscape in the data center. This solution will
   allow users to migrate from a traditional or physical infrastructure to a virtual one,
   without forcing a certain technology or operating methodology on the user. This
   approach applies to enterprise and hosting centers of any scale.

Slide 3 of 23

   First, we will go through the common requirements and trends that we are seeing in
   the data center.

Slide 4 of 23

   The data center is changing rapidly. Enterprise and service providers alike are looking
   to provision applications on the fly, scale applications seamlessly, and simply submit
   jobs without worrying about server capacity, memory, or even location.

   Blade servers, rack servers, multi-core processors, and virtualization are enabling a
   high degree of consolidation in the data center. These technologies have led to a much
   greater computational density in the server rack and data center. As a result,
   enterprises have been able to greatly reduce the number and physical footprint of their
   data center locations.

   It is very common for enterprises today to look to host their data centers in two or
   three locations as opposed to ten or greater. Efficient power management is becoming
   key. Cost of power is increasing, as new high density data centers draw more power
   and cooling. Governments around the world are putting caps in place as they look to
   reduce carbon emissions. Data center administrators are looking to optimize power
   consumption by automating power management. An example of this is being able to
   follow the moon due to lower power cost savings at night.

   There are many similarities between enterprise and managed hosted data centers with
   enterprise data centers. Enterprise data centers are looking more like service providers
   having application level SLAs with their internal departments.

Slide 5 of 23

   Now we will talk about some of the issues that arise in the network as you move
   towards highly virtualized environments.

   With the advent of blade switches and soft switches and the hypervisors, we are
   seeing a blurring of the network edge in the data center. The traditional switching
   model of two or three tier networks, were switching at the edge begins, has been
   replaced by the addition of one or two tiers of switching in the server.

                                       Page 1 of 7
                         Module: <Presentation Name>

   This model has greatly increased the complexity of switching in the data center, both
   in terms of number of switches in the network, but also in terms of complexity of
   switching in the data center.

   Ability to configure and manage different virtual switches is an issue. Today no
   standards exist for virtual switches. Each vendor has their own set of functionalities
   and capabilities. The challenge is how to have all these network layers and different
   vendors coexist, not just in the data center, but even within a single server.

Slide 6 of 23

   [No Audio]

Slide 7 of 23

   Another issue created by the inclusion of switches in the server, whether virtual switch
   or blade switch, is which functional group has responsibility for configuration and
   management of the switch and the server.

   Traditionally the networking group has been responsible for switches, but with
   virtualization, there has been a shift in the control and management of networking
   level functions from the network to the server group.

   The blurring of the line between the server and the networking group can increase the
   likelihood of errors, and can lead to more complicated maintenance and
   troubleshooting between the server and networking groups.

   This blurring of the line between the server and the networking groups'
   responsibilities has created a shift in the control and management of networking level

Slide 8 of 23

   ViaMobility is a benefit of virtualization. Disaster and site recovery processes, as well
   as load balancing applications, count on the ability to dynamically move workloads
   across the data center infrastructure. Resiliency and fault tolerance depend heavily on
   the dynamic capability as well.

   The ability to dynamically provision new workloads relies on being able to move
   workloads around, yet networks are static. Traditionally network configurations were
   set up and rarely changed, but the need to provision workloads dynamically means
   that network configurations must also be dynamic. ViaMobility requires the network
   configuration, QoS, ACLs, rate limiting, etcetera, to dynamically follow the virtual
   machine as it moves through the network.

                                       Page 2 of 7
                          Module: <Presentation Name>

Slide 9 of 23

   Extreme Networks has developed a network infrastructure solution that will
   accommodate the evolving technology landscape in the data center. This solution will
   allow users to migrate from a traditional or physical infrastructure to a virtual one,
   without forcing a certain technology or operating methodology on the users. This
   approach applies to enterprise and hosting centers of any scale.

Slide 10 of 23

   The Extreme Networks solution for migrating from physical to virtual to cloud is
   represented by Four Pillars. Each pillar represents an advanced data center
   implementation phase.

   The first phase is called Physical. This represents the physical network that most data
   centers are using today. As this physical network is the foundation for all
   virtualization, it is important that this infrastructure be robust, easy to manage, and

   The second pillar is called Efficient. This is where the efficiencies of virtualization are

   The third pillar is called Scalable. This pillar provides support for dynamic scaling of
   switches and virtual machines. It also includes the highly scalable Extreme Networks
   Direct Attach Architecture, that supports switching of the virtual machines in the
   network rather than on the server.

   The last pillar on the far right is called Automated & Customized. This provides a
   model for data center managers to customize their environment and automate routine
   tasks, which could otherwise be very time consuming and prone to errors at scale. We
   will address these in more detail, but we wanted to put this framework up so that
   things will fit into context.

Slide 11 of 23

   The physical data center network infrastructure is the foundation for the migration
   from traditional network designs to highly efficient virtualized environments and the
   cloud. As the data center evolves, the physical infrastructure must have the capability
   to meet the needs of evolving virtualized and cloud models. The infrastructure must
   be able to handle the increased bandwidth requirements, address virtualization
   efficiency in the network, and deploy newer scalable network architectures.

   The physical infrastructure for both stackable and module or chassis-based platforms
   needs to be able to address these evolving technologies without requiring the network
   manager to rip and replace their existing infrastructure.

                                        Page 3 of 7
                         Module: <Presentation Name>

Slide 12 of 23

   Today we have in production a broad range of data center specific products. We have
   stackables with 1 Gig connectivity, stackables with 10 Gig connectivity, and all our
   stackables really do stack, simplifying management in the data center.

   We have a complete series of chassis-based system, BlackDiamond 8800 and 8900
   Series modules. The BlackDiamond 8900 has more density and larger table size for
   large deployments. All of these systems run on the same version of the Operating
   System, ExtremeXOS. All of these systems will support our virtualization

Slide 13 of 23

    One way we have enabled the simplification of data center network is to reduce the
    number of tiers required to support these highly virtualized environments. With our
    Direct Attach model, we were able to reduce the complexity in number of network
    tiers by one or two tiers based on configuration deployed. This architecture can
    greatly reduce cost, power, and cabling, and as well increase performance of the

    The architecture leverages our BlackDiamond 8900 Series modules, that utilize high
    fan out service modules and the MRJ21 Cable consolidation connector for high
    density data center environments.

Slide 14 of 23

    Our chassis-based products also are very power efficient. In the data center
    environment, our chassis-based products; 8800, the 8805, and 8809, use about one-
    third the power of the Cisco 6509 and half the power of the Brocade chassis. This
    provides tremendous cost in cost savings due to power and cooling cost.

Slide 15 of 23

    Now, for the Efficient pillar. Today’s virtualized data centers are highly inefficient
    with respect to server level virtualization, because the network and server are

    The next step in the Four Pillar strategy is to efficiently manage a highly virtualized
    network environment, with tight integration between the server environment and the
    network. This integration is required to enable mobility in the network, as well as to
    provide network administrator level insight and control into virtualization.

    Also critical to the success of achieving efficiencies are, tight integration with
    virtualization management platforms, support for heterogeneous hypervisor
    environments, meaning VMware, Microsoft, Citrix, KVM, etcetera, management of

                                       Page 4 of 7
                         Module: <Presentation Name>

    virtualization mobility through virtual port profiles, and full lifecycle management of
    virtual machines, from creation to end of life.

Slide 16 of 23

    One of the key components to enabling efficient VM deployments is making the
    network VM aware. VM aware capability enables administrators to tightly and
    dynamically manage virtualization moves and the associated virtual port profile
    moves within the network. Enabling a highly efficient and automated virtualization
    environment, where the network and the virtual machines are tightly integrated. A
    critical element is to make the VM aware capability open, allowing support of all
    hypervisor environments.

Slide 17 of 23

    Extreme Networks has recently introduced a product that enables a tight coupling of
    the network and the server environment. This product is XNV (ExtremeXOS
    Network Virtualization). The product enables full integration into the virtualization
    environment from an inventory and lifecycle management perspective.

Slide 18 of 23

    The third pillar in the Four Pillar solution is around scalability. Large data center
    operators are quickly running into scale issues in their facilities. These scale issues
    manifest themselves in many distinct areas, one of which is being able to handle
    thousands of virtual machines across thousands of switches. This provides unique
    scale issues within the data center.

    A second area where scale becomes very important is the need for higher capacity
    links in the network, as more and more devices get connected via 10 Gigabit
    Ethernet. This is driving demand for 40 Gigabit Ethernet in the aggregation part of
    the network.

Slide 19 of 23

     Today’s architecture includes switching at the virtual switch level. This architecture
     enables virtualization today, but not without issues. There is a movement within the
     industry to move the networking functionality back to the network. The IEEE
     standard for this is 802.1Qbg. The industry is working on ratifying this standard as
     we speak.

     This standard allows switching to be moved back to the network. This has multiple
     benefits, including reducing complexity, taking the switching burden off of the
     server, increasing security, and removing the organizational complexity introduced
     by the virtual switch.

                                       Page 5 of 7
                        Module: <Presentation Name>

Slide 20 of 23

     Extreme Networks also offers highly scalable edge and core products in the data
     center. Recently we announced 40 Gigabit Ethernet solution for both the
     BlackDiamond chassis-based products and the Summit Stackable Products.

     In the Summit product line, 40 Gig is offered via the VIM module, which provides a
     4×40 Gig VIM interface to the network.

     On the BlackDiamond 8806 and 8810, there is a 6 port 40 Gig blade for the chassis-
     based products.

     As data centers deploy at scale, with increased complexity and computational
     density, it becomes increasingly important to be able to automate redundant tasks
     and simplify the add-change-delete process within the data center.

     In addition, as integration of data center storage networking and server resources
     increases, the network administrator needs the ability to easily integrate and
     customize deployments via common APIs. The ability to tightly couple all
     components to the infrastructure and orchestrate the management of the pieces via a
     common platform becomes essential to realizing the benefits of highly virtualized
     and cloud environments.

Slide 22 of 23

     The Extreme Networks product family is designed from the ground up to support
     automation and customization in the data center. We have a single Operating
     System that runs across the product family. The single Operating System has a
     standard XML API and SDK for easy integration, as well as scripting capability,
     which we call Widgets, that enable easy automation of routine tasks.

     The product family also includes EPICenter, a management portfolio for managing
     and configuring the entire product portfolio. The platform also supports the ability
     to add loadable modules to the network to enable various functionalities and

Slide 23 of 23

     Given the big push towards green and the escalating cost of power, the network
     infrastructure is an important part of the power equation; for every dollar spent on
     power, the network needs $2 in cooling. Extreme Networks’ BlackDiamond 8800
     Series consumes a third the power of similar competing products.

     Also, Extreme Networks solution offer the ability to reduce power during off-peak
     hours, using customizable profiles. Profiles are essentially scripts that based on

                                      Page 6 of 7
                   Module: <Presentation Name>

time of day can turn off power to blade reports. This allows data centers to truly
follow the moon and be more efficient with power management.

                                Page 7 of 7

Shared By: