WAN optimisation versus more bandwidth
Regardless of system size or area covered, companies are relying more and more on wide-area networks (WANs) to provide their employees with access to data.
However, poor application performance can be a common problem experienced by users accessing data over the WAN, resulting in reduced productivity and frustration for employees.
When users complain of poor application performance, organisations often want to upgrade the bandwidth of WAN links. But instead of solving the problem, CIOs often discover that upgrading bandwidth to remote sites has little or no effect on application performance. This is because the problem is often the result of latency and application protocol inefficiencies across the WAN, rather than constrained bandwidth.
Bottlenecks affecting application performance over WAN connections typically have lower bandwidth and higher latency than LAN links, but how do those constraints actually affect application performance? There are four distinct bottlenecks – one relating to bandwidth and three relating to latency. The bandwidth bottleneck is straightforward – no application can send more data than the available bandwidth. The three latency bottlenecks are more subtle and tend to be evident only when there is no bandwidth bottleneck. Due to latency bottlenecks, applications may not be able to take advantage of available bandwidth even when the bandwidth appears to be plentiful.
The first latency bottleneck is caused by the end-to-end acknowledgement behaviour of transmission control protocol (TCP). TCP has a window of packets that can be in flight from one end to the other (i.e. between client and server). After the window is full, the sender cannot send additional packets until the destination acknowledges receipt of at least some of what has already been sent. If the maximum window is too small, the throughput of the link will be limited by the rate at which each full window can be sent to the other side and acknowledged.
The second latency bottleneck is caused by the slow start and congestion-control behaviours of TCP. The first latency bottleneck is a limit based on the maximum window possible. This second latency bottleneck is caused by TCP not even running at that (probably inadequate) maximum window size all the time. Instead, TCP gradually ramps up its window size when transmission appears to be successful and sharply cuts back its window size when transmission appears to be unsuccessful. In networks with both high bandwidth and high latency, this behaviour leads to extended periods in which available bandwidth is unused. However, this bottleneck is primarily an issue for users trying to fill long fat networks.
The third latency bottleneck is caused by application protocols that are running on top of TCP. Recall that with the first latency bottleneck, the availability of bandwidth didn’t matter if TCP was limited by the size of a window of data and the need to acknowledge that data. Analogously, the availability of bandwidth and the avoidance of the first and second latency bottlenecks (at the TCP layer) don’t matter if the application is limited by the size of application messages and the need to acknowledge or respond to that data at the application layer.
Application protocols that were originally designed for a wide-area environment, such as HTTP and FTP, generally don’t encounter this third latency bottleneck. However, application protocols originally designed for use on LANs such as Microsoft Windows file sharing via CIFS, are often severely affected by the third latency bottleneck.
Reducing bottlenecks with WAN optimisation solutions can utilise a variety of approaches to reduce bottlenecks over the WAN. These can include data reduction, caching of data, files and email, block replication, TCP optimisation, quality of service, network compression and SSL acceleration. However, each of these approaches only offers improvement for a narrow set of protocols. For example, caching will overcome application latency, but not TCP latency.
Enterprises should therefore consider a WAN optimisation solution that combines a number of approaches to target multiple bottlenecks simultaneously.
With a multi-protocol, multi-configuration and multi-application WAN optimisation solution in place, businesses can take advantage of their networks, infrastructure and applications in ways they never imagined possible. This multiple layered approach not only improves the performance of applications running over TCP, but also addresses chatty application protocols with application specific modules. Enterprises can improve application performance across the network typically by five to 50 times and in some cases up to 100 times, and can simultaneously reduce bandwidth utilisation by 65 to 95%.
This can mean an existing WAN can support many more users. New applications can be rolled out and an expensive bandwidth upgrade can be delayed or avoided, in some cases by up to five years. This is made possible by a thorough understanding of, and the ability to address the four distinct WAN bottlenecks outlined above. In addition, investment in a WAN optimisation solution can enable enterprises to realise substantial cost savings in other ways, including:
Consolidating infrastructure into the data centre – enterprises can remove much of the IT infrastructure such as file and e-mail servers, SMS servers, SharePoint servers, tape autoloaders, network attached storage and remote office backup systems that sit in branch offices without impacting application performance.
Optimising disaster recovery – performance of a disaster recovery site can be improved, resulting in cost savings for organisations and data backups can be performed in a more frequent and reliable manner.
Enabling greater collaboration – employees can share large files regardless of where they are located, resulting in more productive users.
Reducing RPO – backups and replication can be performed over long-distance WAN links and completed during backup windows that were previously unachievable.
WAN optimisation is a key tool for distributed organisations. WAN traffic can be reduced, application performance can be significantly improved and IT infrastructure consolidation and backup and recovery projects can be implemented.
WAN optimisation has enabled many companies to get more out of their existing infrastructure and at the same time enabled them to avoid or delay costly bandwidth upgrades.
Increasing bandwidth versus WAN optimisation << discussion
EngineerIT
Loading ...