Businesses today are highly dependent on distributed applications to support every aspect of operations. If these applications under-perform for remote users or fail, losses of productivity, revenue and opportunity inevitably result. It is thus critical to ensure the consistent performance of applications across the network.
One of the gating factors controlling application performance is bandwidth. As more applications and services are activated on the network, they contend for the finite available bandwidth. Bandwidth can be an especially critical factor for companies with small or overseas locations that may not have high-capacity network connections to the data center.
Typically, IT organizations approach this critical relationship between application performance and bandwidth by managing supply. This supply-side management approach is characterized by adding more bandwidth or implementing technologies that prioritize use of the bandwidth that's currently available.
But IT organizations can no longer depend on supply-side bandwidth management alone. Demand -- driven by more applications, higher volumes of data and increasing intensity of use -- is just growing too fast. Funding for technology infrastructure is growing too slowly. And the consequences of service interruptions are too great.
In fact, supply-side management alone fails to address a variety of issues. Some applications aren't very well designed for deployment on the network, so they won't perform well, regardless of how much bandwidth you throw at them. Some applications will perform a bit better with more bandwidth, but those incremental performance gains aren't worth the cost of the additional infrastructure. In some cases, management needs to consider retiring an application altogether. In other cases, steps must be taken to reduce end-user demand.
Simply put, network managers have to do more than just manage bandwidth supply. They have to apply best governance practices to the consumption of bandwidth, so that utilization of network resources is closely aligned with business drivers. Only by exercising this kind of governance can IT use its infrastructure dollars in the most effective possible way.
The governance lifecycle
Good bandwidth governance actually begins well before an application is deployed on the network. With the right technologies, developers can start assessing the behavior of their applications over the network early in the design and development stages. That way, they can resolve excessive bandwidth consumption or poor performance issues as soon as they arise, rather than later in the game, when such problems can be very costly to fix.
This kind of testing should continue right up to deployment, so that there are no surprises when the application is rolled out onto the production network. It should also be done every time the application is upgraded or modified, because subtle changes in code often have unexpected impact on the behavior of applications on the network.
IT can apply these bandwidth governance best practices to applications that are already in production, too. For example, before throwing bandwidth at an application performance problem, network managers should first model potential solutions to find out if the additional bandwidth will, in fact, deliver expected improvements. What-if scenarios should also be run to answer key governance questions such as "Will current bandwidth levels support the addition of 20 users in our Atlanta office?" and "How will night shift users be affected if we start backing up remote servers over the network at 2:00 AM?"
Only by answering these kinds of questions in advance can network managers ensure that bandwidth is being used for the best possible business purposes.
Bandwidth governance best practices
To achieve best practices bandwidth governance, IT organizations require technology capable of replicating the production network environment as it exists today and as it might look tomorrow. This "virtual enterprise" should be capable of assimilating all the factors that impact application performance in the real world: live applications, the data center that supports them, the topology and bandwidth constraints of the network, the number of distribution of end users, etc.
By leveraging this virtual environment, everyone involved with bandwidth governance -- from application designers and QA staff to network managers and architects -- can more effectively control bandwidth utilization and preempt potential consumption and performance problems. They can also verify the effectiveness of any planned supply-side measures, such as QoS and bandwidth grooming, they plan to implement in production.
Unfortunately, most IT organizations rely only on development LANs (which don't reflect conditions on real-world enterprise networks) or mathematical simulations to assess the behavior of applications. These resources are useful, but don't provide the precision or flexibility necessary for the kind of true bandwidth governance IT will have to implement if it is going to maximize returns on development and infrastructure investments.
That's why it's essential that IT organizations re-evaluate their bandwidth management strategies and their technology portfolios. Those that continue to manage application network performance in one silo and application development in another won't be able to govern bandwidth effectively across the application lifecycle. Only with an accurate, flexible and proactive approach can IT bridge the gap between development and production, and thereby meet its goals of reliable performance, cost-efficient service delivery, and tight alignment of expenditures with business priorities.
About the author:
David Hochhauser is Vice President of Marketing at Shunra Software, which delivers application and network optimization solutions for distributed environments. He is responsible for developing the company's marketing strategy and operations including product, sales and channel marketing efforts. For more information on Shunra, visit the Web site at www.shunra.com.
Dig deeper on Bandwidth and capacity planning