Network performance and bandwidth critical to cloud success

For the most part, cloud computing’s hype has quieted down and reality has set in. U.K. enterprises now have cloud high in their list of IT priorities and are moving ahead with budgets, pilots and implementations. But cloud-based applications–while promising to deliver business requirements such as flexibility, on-demand compute power, improved agility and faster time-to-market—are only as good as the network connections that support them.

Those connections, especially those that link enterprise branch offices to cloud-based servers, are vital. They have to be capable of handling the cloud computing growth that’s already occurring. According to market research firm TechMarketView, the U.K. cloud computing market is predicted to reach £6.1 billion by 2014.In the report titled 'Information Communications Technology (ICT) in the UK: investment opportunities,' issued by the U.K. government in February 2014, 81% of established cloud users in the United Kingdom plan to increase cloud usage over the next two years. Already, 18% of small and medium enterprises (SMEs) in the United Kingdom use cloud and another 30% plan to use cloud computing in the next 12 months.

> See also: 4 steps IT should take to prepare for the UK's new flexible-working law

Any disruptions, delays or bottlenecks in the networks that carry cloud traffic are simply unacceptable, when so many businesses use cloud. They can be debilitating to productivity, customer service, product development and delivery, revenue and innovation.

How can the performance of cloud-based applications that are hosted in private or public data centers be improved to provide better user experience for the enterprise branch offices? More importantly, can it be done cost-effectively, without MPLS or other costly off-network solutions?

The answer to both questions is yes. Various technologies are available alone or in combination to solve network issues without massive capital expenditure or infrastructure upheaval.

WAN optimisation is used to streamline payloads via content caching, de-duplication and/or compression to enable traffic and content to flow more effectively in existing pipes.

Broadband Bonding merges several Internet lines into a single, faster and more reliable IP pipe. Inherently more reliable and about 70% less costly per month compared to a single transport, Broadband Bonding provides a remarkably efficient way to boost overall performance. Because it uses existing infrastructure, it’s a practical way for firms with branch offices to beef up their networks quickly.

WAN traffic management techniques, such a traffic shaping, utilize bandwidth reservations and traffic monitoring to improve performance. Traffic shaping prioritizes protocols and applications to deliver traffic to application servers in a more efficient manner. It also anticipates upcoming requests and bundles them to reduce chatty back and forth.

> See also: Most UK businesses not BYOD- ready but take netwirk investments seriously- report 

Of course, don’t neglect basic traffic monitoring as a way to find hidden bandwidth consumption that might be eliminated. In most cases, Broadband Bonding devices will have built-in traffic management features and can be deployed in conjunction with WAN optimization devices.

Truth is, single-line Internet access links are not reliable enough to support today’s business-critical imperative on WAN connectivity. But getting to a desired service level agreement (SLA) doesn’t need to be cost-prohibitive. The three options—WAN optimization, Broadband Bonding, and WAN traffic management—are all worth considering to keep bandwidth issues from compromising business success.

Sourced from Dr. Cahit Jay Akin, co-founder and CEO of broadband specialists Mushroom Networks

Avatar photo

Ben Rossi

Ben was Vitesse Media's editorial director, leading content creation and editorial strategy across all Vitesse products, including its market-leading B2B and consumer magazines, websites, research and...

Related Topics

Data
Networks