Data Center Solutions
Data Center Virtualization – from Physical to Virtual to Cloud
Extreme Networks has developed a network infrastructure solution to accommodate the evolving technology landscape in the data center. This solution allows users to migrate from a traditional or “physical” infrastructure to a virtualized one without forcing a certain technology, or operating methodology on the user. The goal of this approach is to preserve the tools, security appliances and policies and to not strand the network assets. This approach – that we call the “Four Pillars” – applies to Enterprises and hosting centers of any scale.
The Four Pillar Strategy For Next Generation Data Centers
Our solution for migrating from physical to virtual and on to the cloud has been divided into four pillars. Simply put, the four pillars can be considered as phases of implementation for the migrating data center. A user can move from pillar to pillar in a stepwise fashion, or immediately implement the complete solution based on their business imperatives.
Following are descriptions of each pillar that examine where the pillar stands within your network and how it can affect your network and business.
Pillar 1: Physical Network Infrastructure
The physical data center networking infrastructure is the foundation on which the migration from traditional network designs (to highly efficient virtualized environments) is built. It is critical that this infrastructure be highly robust and deliver the highest levels of predictable performance. The physical infrastructure includes the data center core, aggregation (if needed) and access tiers of the network.
Pillar 2: Efficient Virtualization Lifecycle Management
Data center networks today are very inefficient in dealing with server virtualization. From the time a Virtual Machine (VM ) is created to the time it is activated, moved around or deactivated, the network has no visibility into the virtual machine lifecycle. Network administrators have had few tools if any, when it comes to troubleshooting and managing VMs in the network. Additionally, due to the ability of VMs to dynamically move from server to server, provisioning the network for VM security and application performance has proven to be a networking challenge. The next phase in the evolution of the data center makes the network very “efficient” when it comes to managing, troubleshooting, provisioning and securing virtual machines in the network. Key elements to making the network efficient in this regard are bringing network level insight and visibility to the virtual machine lifecycle, the ability to configure network and port level capabilities at the individual VM level, as well as dynamically tracking VMs as they move across the data center and enforcing the network attributes of the VMs wherever they migrate in the data center.
Pillar 3: Scalable Virtualized Environments
Large data center operators are quickly running into scale issues in the data center. The issues manifest themselves in several distinct areas:
- Explosion in the number of virtual machines on the network
- Rapid growth of Layer 2 and Layer 3 scale in the network required to handle the growth of virtual machines
- Increased demand for higher speed connectivity in the aggregation and networking core
- Higher performance direct server connections
Extreme Networks solves the scale issue above by:
- Moving switching back into the network with Direct Attach architecture
- Providing upgrades to existing chassis and stackable products to 40 Gigabit Ethernet
- Being prepared for the move to 10 gigabit iSCSI and 10 gigabit connected servers
- Delivering remarkable Layer 2 and Layer 3 scale to handle the rapid propagation of virtual machines
Pillar 4: Automated And Customized Data Center Networks
As data centers deploy with increased complexity and server density, it becomes increasingly important to be able to automate repetitive tasks and simplify add / change / delete processes. In addition, as the data center integration of storage, networking and server resources increases - network managers need to integrate and customize deployments via common APIs. Being able to tightly couple all components of the infrastructure and orchestrate the management of these components through a platform becomes essential to realizing the power of highly virtualized and cloud environments.