Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Optimizing Network Performance For The Cloud

There's something about cloud services that makes some companies forget the fundamentals of IT, things like capacity management, change control, and network management. In the rush to move to cloud services, many businesses become so focused on building their cloud platform, that they lose focus on other critical aspects of their enterprise.

One of the most critical elements that enterprises need to retain control of is their network infrastructure. Indeed, in the new cloud-based world, the network is more important than ever to IT operations. And with critical servers and data now hosted in the cloud, network traffic will only increase, as users access that data and computing resources from both remote locations and corporate headquarters. In the cloud-based world, the network in essence becomes part of the cloud.

For many businesses, it's likewise critical to focus on ways to maximize the efficiency of their existing network and networking budget. That's especially true for small businesses, which must add network capacity to support increasing network demands as their existing users access the cloud, as well as new locations and offices as the company grows and takes on more staff. And while large businesses may have much greater resources than smaller businesses, they still face immense pressure to cut costs and do more with less.

As companies look for way to optimize their networks to work with cloud platforms, a few strategies come to the forefront, including:

WAN Optimization - WAN optimization, or WAN acceleration, is really a collection of techniques designed to manage your network traffic by reducing inefficiencies and redundancies in your traffic, thus reducing your overall traffic demands. Some WAN optimization techniques include:

  • Caching - Caching involves the use of locally stored, pre-populated data such as images and other elements that are commonly requested by your user's browsers. Caching sends a given image, such as a corporate logo for example, to a given remote site, and that image is cached by a local device, which can then save that image and serve it up locally, thus saving on bandwidth usage.
  • De-duplication - De-duplication removes network traffic associated with data backups by performing a single complete backup of a given file(s), and then only sends changed portions of those files over time, rather than the entire file.
  • Compression - Compression utilizes a variety of algorithms and techniques to reduce the size of the data being sent across the network, removing unnecessary portions of the file(s) in question, and thus reducing the amount of traffic being transmitted.

Broadband bonding - Broadband networks bring a lot of promise for both SMB and large enterprises, offering the option of relatively large bandwidth at a very cost-effective rate. But broadband circuits are typically limited in terms of both size and reliability, which in turn limits the effectiveness of using broadband connections. However, by utilizing broadband bonding, you can combine several broadband and Internet connections into a single, unified connection that provides scalability and fault tolerance.

WAN traffic management - Many companies don't manage their network traffic effectively, wasting their network resources and limiting the effectiveness of their networks, and moving into the cloud, typically, only exasperates this problem. WAN traffic management utilizes techniques such as traffic shaping and traffic monitoring to identify and prioritize critical traffic, increasing the efficiency of your network.

The key thing to remember is, as the cloud evolves, it's clear that the network and the cloud are quickly becoming one and the same. In essence, the cloud is becoming a function of the network, and the end goal of utility computing -- when you think about it -- is that computing will eventually become a network function.