5th, December 2007
When application failure is not an option Posted by Dan Kusnetzky Here are some thoughts from a conversation I had with the good folks at Datasynapse a while ago. I came across my notes and thought that it might be interesting reading. It has become clear that most organizations simply have no time for application failure. Their IT infrastructure simply must be up and available when it is needed. Datasynapse would assert that application virtualization can address this issue while keeping hardware, software and staff-related costs in line. I think that if the technology is properly utilized, theyre right. Whats the pain Organizations live on Internet time now. This means that their information systems and the supported applications must be up and available 24 hours a day, 7 days a week, 365 days a year. There simply is no time for unscheduled downtime. Theres little time for scheduled downtime for system maintenance or backup of critical data. Needing to keep the store online in all time zones all of the time means that there simply isnt a good time to take the system down. IT executives face a great deal of pressure from customers, staff, partners, and now, regulatory organizations to keep their applications available at all time. The imperative they face is that their systems must always perform. These executives know that if their application systems become unavailable, the organization is likely to lose revenues, customers and, in some cases, face penalties and fines. Customers often wont wait. If they cant order the desired products or services, theyll just hop down the net searching for other suppliers. Staff members and partners face tremendous pressure to be highly productive and they view application failure as simply unacceptable. Regulatory organizations have no patience for late or inaccurate reports. How can an organization create a plan that provides an environment in which application failure can be avoided? According to DataSynapse, application virtualization makes it possible for key applications to remain available during planned system downtime and during unplanned outages. Application virtualization Unleashes High Availability DataSynapse would tell IT executives that application virtualization can address all of the issues mentioned above by unleashing higher availability without unduly increasing costs for hardware, software or administration. This technology can also make it possible for organizations to make the best use of their resources (see the Kusnetzky Group paper
Application Virtualization and Utopia: Proving the Value of Virtualizing Applications) and actually improve application performance as well (see Unparalleled Performance: Harvesting the Power of Application Virtualization). While there are other ways to achieve these ends, the conversations Ive had with users of DataSynapses products lead me to conclude that theyre telling the truth. Redundancy In the past, application availability was made possible placing by each application system on its own set of redundant servers. Failure of any component could be hidden by having backup systems take over necessary tasks before anyone would notice a failure had occurred. Organizations are now challenged to reduce costs of hardware, software and administration by utilizing virtualization technology without also introducing potential application failure points. Monitoring To accomplish this feat, IT executives know that they must adopt technology that allows application components as well as the underlying physical and virtual systems to be carefully monitored. Monitoring only the health of the physical systems that support multiple application components, applications and/or virtual systems just isnt enough. Optimization It is not enough to gather the details concerning the state of all of the application components, applications, virtual systems and physical systems. There is just too much information for the IT administrative staff to monitor in real time. This information must be integrated, decisions must be made on the appropriate actions and these decisions must be put into immediate action. Unfortunately no human being or group of human beings knows enough about whats happening inside of a complex computing solution or can do this fast enough to make any necessary changes to the environment before an application slow down or failure is seen by those accessing the solution. Automation Making these key decisions isnt enough. People simply can not act fast enough unaided. So, other tools must be deployed to act based upon the decisions made by optimization technology. This means giving high priority tasks more resources (processing time, memory, storage and the like) when it appears that their performance is not going to mee...