| ||
|
|
|
Posted: Thu, March 22, 2007
Fault-tolerant Linux - expanding the choice of operating system for availabilityby Sri Sridharan
Choosing the operating system that an organisation uses to run its critical applications also remains a tough decision. Linux is growing in popularity compared with other operating systems and using Linux offers a route for organisations to achieve high availability at a potentially lower cost. As a free operating system, the level of cost would be much lower than other approaches and this is contributing to its growing popularity from a business continuity perspective. An example of this is that Linux has entered the top three operating systems chart by the volume of servers sold for the first time, according to recent research by IDC. Demand for Linux-based, high availability servers is growing at a rate of 40% year-on-year, compared to a growth rate of only 15% for the overall business continuity server market. This level of demand has two main drivers: the first is that organisations are seeking to replace their existing UNIX boxes with newer, more powerful machines. As the physical machines reach their end of life and the manufacturers' warranties run out, replacing them becomes a necessity. Secondly, organisations are seeking to move off their proprietary UNIX systems and move to an open systems platform that is cheaper to license and maintain. While UNIX remains a very reliable OS and there is not a lack of technical staff to support it, there are two other OSes that could be used instead: Microsoft and Linux. From the perspective of someone who is used to running UNIX, the shift over to Linux should seem pretty simple from the viewpoint of managing the system in the long term. Another advantage over taking a Windows-based approach is that there are substantially less operating system patches and Service Packs to put in place, with the high levels of testing that this requires. Windows has also been subjected to a barrage of attacks and viruses over recent years, so avoiding this headache has also been a reason to evaluate Linux. Where Microsoft has had an advantage in the past over Linux is that it has had ways of guaranteeing the running of its operating system - whether it is a cluster-based system or a fault-tolerant server, Windows has been capable of running in a high availability environment for several years, while this market has remained untapped for Linux-based systems. This situation is now changing, and fault-tolerant platform support for Linux is now available. In the telecommunications market, this is particularly important as servers and operating systems have to comply with strict regulations around availability and working alongside other electrical devices. The servers running in this kind of environment have to be reliable, but the operating system also has to be aware of the environment it is running in and be hardened against possible failure. The choice of OS will therefore be governed by what application you are looking to support, the level of ongoing maintenance and costs that the organisation can afford and the skills that are required of its staff. Where an application can run across multiple platforms, evaluating the level to which availability is required compared to the cost that it will take to maintain this availability is now a much wider choice than it was before. The benchmark for availability is still the "five nines" of uptime, or 99.999% availability of a service per year - when this works out at about five minutes of downtime per year, you can start to see the extent to which the operating system chosen to support an application can be an important choice when designing an organisation's IT infrastructure. Sri Sridharan is Vice President, Solutions Services Group, Stratus. The company will be exhibiting at the Business Continuity Expo and Conference held at EXCEL Docklands from 28 - 29 March 2007 - the UK's definitive event for managing risk, resilience and recovery. This event will explore the solutions and best practice to ensure operational continuity and protect a company's interests before during and after an incident. For further information visit www.businesscontinuityexpo.co.uk. Send a comment about this article to editor@itwales.com. |
If you ask any IT Manager if they would want an operating system that was
'guaranteed' to be available and supporting applications running continuously twenty four hours a day, 365 days a year,
then the answer would probably be very simple. However, if you want to measure how much this level of availability is worth to the organisation, then you'll probably get a very different response.
Balancing the level of costs involved in supporting a continuously available application against how much any incident of downtime might cost the organisation remains a real concern. |
|
Subscribe to ITWales Updates Click Here! |