|By Jake Sorofman||
|December 10, 2008 07:00 PM EST||
Recessions are about as appealing as a root canal; but they do force us to think differently. Now that the recession is official, it's an ideal time to explore how virtualization and cloud computing can help "recession-proof" IT by transforming yesterday’s costly and rigid computing model to one that puts costs under control and sets applications free.
The National Bureau of Economic Research recently declared that the
1. Virtualize infrastructure to increase capacity utilization.
Traditional server infrastructure tightly couples applications to hardware, wasting computing capacity whenever applications utilize less than 100 percent of system resources. Virtualized infrastructure decouples applications from hardware, freeing excess capacity for use by other applications. A single virtualized server can often support 5X the workload of a non-virtualized server. This allows IT to consolidate server infrastructure, which reduces capital costs associated with server acquisition and datacenter infrastructure, as well as operating costs associated with management, maintenance, and energy consumption.
2. Use external clouds to offset capital infrastructure expense.
While virtualized infrastructure can reduce capital expenses, IT may have the opportunity to eliminate those expenses altogether by using the variable compute model of external clouds like Amazon’s Elastic Compute Cloud (Amazon EC2). In this model, compute capacity becomes elastic, allowing lines of business to align the cost of application consumption to actual demand. Swapping traditional datacenter for external cloud provides infinitely scalable capacity and the ability to align cost to value received.
3. Virtualize applications to accelerate and simplify deployment.
Packaging and deploying application workloads as virtual images can close the “deployment gap” which adds cost and delay to the deployment of enterprise applications. The virtualized application is separated from its operating infrastructure and a self-contained unit that includes just enough operating system (JeOS), databases, and middleware required to run the software in production. These bits travel with the application package and allow it to run as an image in any virtualized or cloud-based execution environment without any manual setup, tuning, configuration, or certification. Suddenly, applications are set free and deployment cycles are compressed from months to minutes. This equates to cost savings and improved business agility.
4. Construct virtual applications for simplified management, automated maintenance.
The reality is that this new approach to application delivery can create new costs and risks. Taking the friction out of application deployment will lead to an onslaught of volume and demand, resulting in what is often called “VM sprawl.” What organizations must recognize is that they may be exchanging one cost and management burden for another, as physical machines become virtual machines. In fact, virtual sprawl is likely to far outstrip any physical sprawl you’ve witnessed heretofore. As such, organizations need a scalable approach for managing and maintaining application images. Adding headcount isn’t an option, so the answer is finding ways to do more with less. In this case, this means architecting application images for management and control, trading manual one-at-a-time updates for seamless changes that are implemented en masse. It also means complete lifecycle control and transparency wherever the application is being run — datacenter or cloud, internal or external.
Recessions are about as appealing as a root canal. But they do force us to think differently — to take an inventory of costs, retool, reinvent. The reality is that this recession is coincident with a fundamental inflection point in IT. The friction and the economics of traditional computing models no longer work. This is why organizations must embrace virtualization and cloud — both to weather the storm of a down economy and to transform yesterday’s costly and rigid computing model to one that puts costs under control and sets applications free.
Jul. 29, 2015 03:15 PM EDT Reads: 232
Jul. 29, 2015 03:00 PM EDT Reads: 1,251
Jul. 29, 2015 02:00 PM EDT Reads: 1,163
Jul. 29, 2015 01:45 PM EDT Reads: 417
Jul. 29, 2015 07:30 AM EDT Reads: 278
Jul. 28, 2015 06:30 PM EDT Reads: 1,374
Jul. 28, 2015 04:30 PM EDT Reads: 1,753
Jul. 28, 2015 11:00 AM EDT Reads: 2,031
Jul. 27, 2015 09:00 PM EDT Reads: 2,038
Jul. 27, 2015 10:00 AM EDT Reads: 2,020
Jul. 27, 2015 09:00 AM EDT Reads: 311
Jul. 27, 2015 08:00 AM EDT Reads: 1,892
Jul. 26, 2015 09:00 PM EDT Reads: 1,559
Jul. 26, 2015 08:00 AM EDT Reads: 2,141
Jul. 25, 2015 02:00 PM EDT Reads: 381
Jul. 25, 2015 01:00 PM EDT Reads: 1,945
Jul. 25, 2015 12:15 PM EDT Reads: 446
Jul. 25, 2015 12:00 PM EDT Reads: 1,521
Jul. 25, 2015 09:00 AM EDT Reads: 1,480
Jul. 24, 2015 11:00 PM EDT Reads: 2,046