Software defined networking (SDN) and big data go together like peanut butter and jelly. Big data ROI comes from the ability to mine both structured data and unstructured data, such as video content and social media, as part of analytics. Software defined networking makes it possible to build huge intelligent adaptive networks capable of handling both structured and unstructured data as part of big data analytics. What SDN offers is a programmable interface to program software intelligence into networks so they are customizable, scalable, and agile; all essential ingredients for delivering big data on demand. Like peanut butter and jelly, SDN and big data each can work by itself, but bringing them together makes something even better.
According to a recent report from Research and Markets, 35 to 40 percent of the overall network spend by 2018 and 2019 will be on software defined networking, which means SDN offers a huge revenue potential for VARs. What’s driving those sales is more demand for network virtualization, data center consolidation, and new strategies that enable network automation as enterprises become more complex in order to support things like big data.
For big data ROI, today’s enterprises require highly distributed networks with globally dispersed resources. However, creating a distributed infrastructure using the cloud means a drop in data transfer rates. The only way to support cheap, efficient, diverse network resources is with software defined networking. Or thought of another way, demand for high performance computing to power big data analytics is creating demand for more agile, elastic networks with resources available as needed; a demand that can only be met with software defined networking.
Software Defined Networking Powers Big Data
Provisioning and managing networks has always been a labor-intensive process. Once the enterprise environment is stable maintenance is a nuisance but not much of a hardship, but what happens when network resources are continually shifting? Virtualization, for example, adds a layer of complexity that makes network administration that much more complex; too complex for manual processes.
The best way to manage large volumes of highly distributed data used for big data, along with compute-intensive applications, is by creating a more efficient software framework using virtualization. Big data applications and the operating system are partitioned into separate available resources; virtual machines are isolated to prevent system failures; and each virtual machine can be represented as a single file to facilitate resource sharing. The result is scalability and operating efficiency.
Software defined networks eliminate the pain of manual administration by using virtual resources. Control and forwarding are separated and the network is treated as a unified whole so the SDN controller can use the entire network infrastructure to service application workloads as needed.
SDN controllers can also serve as good platforms to run virtual applications. By virtualizing applications in this fashion and using the entire network the applications can streamline provisioning by using top-down intelligence based on controller input. The result is consolidated and scalable administration; perfect for big data applications.
The Symbiosis of Software Defined Networking and Big Data
Now let’s take the relationship between SDN and big data once step further.
The intelligence to automate highly distributed networks resides in analytics; assessing data traffic to program responses to traffic behavior, such as optimizing data paths. Then the question becomes, how many endpoints do you include in your analysis and what impact does virtualization have on those endpoints?
The more endpoints you can add the better your intelligence and the more efficient the network automation. The more traffic you have to process, the more it looks like big data. Suddenly you find yourself using big data to manage software defined networks that in turn enable big data analytics. Software defined networking and big data form a symbiotic relationship:
- Mining big data intelligence is a three-stage process: 1) split the data into multiple server nodes; 2) analyze each data block; and 3) merge the results.
- Software defined networking separates the control and data planes to provide customizable scalability and agility.
- SDN configures the network on demand using endpoint analysis so the network has the capacity to support big data applications like real-time analytics.
Software defined networking also helps ensure big data ROI by not only streamlining data access but facilitating interoperability. SDN uses OpenFlow and OpenStack protocols, which not only makes network management and automation easier, but also ensures interoperability with any other vendor’s OpenFlow-enabled devices.
The scalability and agility of automating network resources using software defined networks is the key to long-term big data ROI. Big data will provide the patterns to support network programmability, and SDN will use big data intelligence to automate the network. An important byproduct of this interrelationship is more efficient big data processing.
How far have you progressed in your plans for offering SDN and big data?