Subscribe
  • Home
  • /
  • TechForum
  • /
  • Visibility, optimisation, control: how to win the yellow jersey

Visibility, optimisation, control: how to win the yellow jersey

Wimpie van Rensburg, Country Manager - South Africa, at Riverbed Technology


Johannesburg, 28 Jul 2015

The Tour de France is undoubtedly the largest and most famous bicycle race in the world, with millions of fans following and analysing the performance of the riders. If you look closely, there are quite a few similarities between professional cycling teams and modern, hybrid IT networks, says Wimpie van Rensburg, Country Manager - South Africa, at Riverbed Technology.

A cycling team is a hybrid organisation. In addition to the riders who come from all corners of the world, there are mechanics, caregivers, doctors, team leaders and nutritional experts, who all contribute to the success of the riders. While most teams have a central office, many of them are working from different locations.

All employees - and riders - are obviously dependent on optimal information. The same applies to companies that work with branches and offices spread across remote locations. This could be a sales office, production facility or a country residence. In all these locations, it is important that the overall IT landscape - from infrastructure to applications - is working at an optimal level and provides the highest possible availability. This requires visibility, optimisation and control.

Visibility

Often, an IT department doesn't know when a network performance problem occurs and only realises when an end-user mentions it. This suggests a major lack of understanding of the status of the network. Ideally, the IT department needs to be aware of any performance issues before the end-user is affected, and with the right tools in place, it is possible to gain that visibility into the network.

For example, such a tool may indicate that the server is taxed on Monday more than average because employees check their e-mail from the weekend, or on the last day of each month, and may use additional capacity with the SAP environment due to month-end closing. Without the right tools to detect such patterns and without a good understanding of the condition of the infrastructure, the consequences can be quite substantial: unhappy customers, lost sales and lower employee satisfaction.

Compare it with the coaching staff of the cycling team, who wants to know as much about the condition of the riders to develop the best exercise routine, and as the leader of a cycling team, to be aware of the stages within the ride. For example, a mountain stage may require protective clothing; poor road surfaces may require special tyres; and a sunny day may require additional hydration. Mitigating these circumstances ensures riders can focus on optimum performance.

Optimisation

The next step is to optimise the performance of the infrastructure and applications. This is not a luxury - while performance of common applications such as Microsoft Office, Exchange and SharePoint are usually fine on a local area network (LAN) - the same is not true when applied to WANs. Possible causes include a limited bandwidth in the WAN, application interference (aka "chatty apps") and high latency.

This can be prevented in various ways by a number of WAN optimisation technologies:

1. Streamlining data
By reduplicating the data, the WAN is less loaded with redundant data that is being transported. Commonly used data is not continuously sent back and forth between locations, but rather stored locally in an intelligent way and locally merged with new data when needed.

2. Streamlining transportation
With this technology, the aim is to make transportation via TCP more efficient, with data sent back and forth less frequently. The emphasis is on finding the ideal size of packages and then sending them in that size. This can reduce WAN traffic by up to 98%.

3. Streamlining the application
Tools can be used to streamline applications aimed at reducing chattiness and latency caused by application protocols like CIFS, HTTP, HTTPS, MAPI, NFS and SQL.

Control

Control is about finding ways to simply increase the efficiency of the network, so an organisation can reduce the total cost of ownership. It is important to bring the performance of applications in line with the business goals. Additionally, it is of interest to reduce or avoid bottlenecks in the network. This is particularly true for the finance and sales departments, which must have sufficient bandwidth to make the final sprint.

It is important to note that applications can be grouped depending on the type of application or the specific business use case. As a result, having service level agreements (SLAs) for different types of automation applications can be put in place, so that business-critical applications can run on their own, faster and more secure network, whereas other applications that are less critical and inexpensive can run over the public Internet. As a result, applications can be delivered to end-users depending on the priorities that have been put in place to meet business needs.

This is made possible by cost-effectively differentiating the network types. In cycling terms, the management team knows the route of each stage and therefore has the right tools at hand for optimal performance cyclists. They get food before hunger strikes, or the rainwear is waiting in the trailers, by following the weather forecasts closely.

As a cycling network wins the coveted yellow jersey through visibility, optimisation and control, so too does the modern, hybrid enterprise.

Share

Editorial contacts

Jeannie Erasmus
Red Ribbon Communications
(022) 433 3684
jeannie@redribboncommunications.co.za