What Role Does Network play in Data Center?

Data centers are generally where critical business applications reside and where critical business logic occurs, both for internal and external consumers. There are many levels of communications that must happen internally and externally to data centers. Ensuring that these communications are carried out without problems, efficiently and safely is a fundamental role of the network that unites all these components.

Let's take a look at the simplified figure below. Each block has a dependency with multiple blocks that establishes the workload patterns that the network must carry. These dependencies between the modules perform specific commercial functions. These commercial functions could carry different workloads such as:

• Run complex critical business applications at multiple levels and locations
• Shared load and grouping of applications in different geographies.
• Cloud computing: automation and orchestration workloads
• Disaster recovery and business continuity (DR / BC) - availability workloads
• Data replication and backup workloads
• Security and law enforcement
• Development and testing of workloads.
• Daily maintenance
• Workload management

One thing in common in all these functions is the network and its ability to unite these components!

It is essential now that we never have an intelligent, reliable and functional network that provides next-generation innovations for companies to evolve from a traditional network to a "cloud-enabled" network. What is a "cloud-enabled" network? A network that is compatible with VM, a network that can grow and shrink according to consumer demands, a network that can recalculate routes dynamically during failures, a network that can guarantee different kinds of service based on predefined parameters and postures, a network that ensures that there are no blocked routes, a network that can track changing workloads and react accordingly (VM mobility), we can go on and on. Simply put, networks are becoming programmable (API) and flexible to accommodate the paradigm of changing applications that are required in various Cloud models.

There are many network-based innovations that have been widely discussed in Cisco and other forums, such as Virtual Port-Channels (vPC), Overlay Transport Virtualization (OTV), Locator / ID Separation Protocol (LISP), FabricPath, FiberChannel-over-Ethernet (FCoE), Virtual Security Gateway (VSG), etc. These innovations with next-generation HW / SW combinations, such as Cisco Nexus series products, help create a path to a unified fabric, network and computing approach to Cloud Computing. This is further proof that we are trying to address business and technical challenges with smarter networking tools. I am not saying that this level of smart grids is required in all scenarios, but based on commercial and technological requirements, next-generation data center networks are making application decisions that they never had to make before!

For any given data center, its capabilities are finite. Then, immediately, we have an agotable resource to start, normally would be the facilities: energy, rack space, available ports, etc. Or they could be other physical assets within the Data Center such as network, computing or storage. Since we are talking about networks, let's accept that even network resources are finite from various perspectives, for example, scale; number of MAC addresses, VLAN, Layer 3 pairs, performance, over-subscription relationships, to name a few. I will cover some of these aspects of the network in a future topic when discussing data center consolidation and migration planning.

For now, the next time someone claims that networks do not play an important role in Cloud Computing, they will have something to say about it!

This article was originally published on ------- Read More