In fact, there are three main approaches to virtualization:
- Consolidating servers to improve utilization level
- GRID-computing for complex, heavy applications
- Virtualization jobs
Server Consolidation – Have To Build A Cluster
From the point of view of server consolidation, we want to reduce our infrastructure costs, energy consumption, the system administrator in banks. Collect a bunch of unnecessary resources, release it, move applications to a virtual environment, and they’re great. In principle, it is most commonly found in my practice approach when collecting applications, virtualize and it is wonderful with this when goes live. The only drawback approach, not a disadvantage, but, well, the one thing that we need to consider is reliability. If we are virtualizing the dedicated servers, you have to put more than one machine, and cluster. For the simple reason that we have not hosted one application in this virtual environment, but they are a few dozen. And, if the machine we have dies, not just one service goes down, but a whole group. Therefore it is necessary to provide disaster recovery, because the business does not forgive unexpected downtime.
Cloud Computing Solutions Can Reduce Downtime
The Main Task: To provide a large pool of resources. It’s no secret that there are complex, resource-intensive applications, servers, databases, and they are constantly demanding for resources. Continually buy more servers and migrate to them, business is not always ready to go at it. The more we obtain, the higher is the cost of migration from one equipment to another. It is much easier to build a hardware stack, which will allow you to add to the application or database resources as much as necessary. It may be noted that despite the implementation of this idea “resources as long as necessary”, we get an incomplete use of resources, for the simple reason that any application in its life cycle has a peak load. As applied to the banking industry, the end of the year – it’s just devouring huge resources, because of the annual reports to shareholders, the regulator, the calculation of financial results, it eats up a lot of resources. And, despite the wonderful possibilities of balancing resources, usually in practice, if we gave someone a particular application resources, then we back them and not take away, because that really is not necessary. Yet another aspect that there can be, which is present in server consolidation – is moving from one set of hardware to another. Heavy set of applications, this problem is more relevant than in the case of consolidation.
I can give you a practical example: We needed to upgrade the equipment for one of our databases, in principle, except for the time spent by administrators to deploy infrastructure for testing, some test runs, the transition itself took a backup and restore all only in 6 hours. There was only one problem: on this basis, we had tied a customer service that we provide 24 x 7. And 6 hours of downtime is not small enough for the loss of business and despite the fact that we did it and put in the estimated time, went smoothly. In the case of GRID-computing, in our down time, it would amount to a maximum of 10 minutes. And it would take this time to shut down a virtual machine on the same equipment and transfer-switch on the other. Agree that the time is not comparable.
The third area, which is interesting and important – is desktop virtualization. On the one hand, in terms of infrastructure support, the decision itself is wonderful. Unhappy support specialists workstations stop constantly that are running around in the office, deal with applications in the field, with the problems and go in the console. Again, gathering in a virtual environment workstations, we reduce the risks of information leakage. It’s no secret that there are many ways to carry out a local station. Here, we can more clearly and more tightly control the resources available, to restrict the user’s ability to put a device in the workstation and something to try to merge… and so on and so forth. Here, one of the limitations is that the cost of server resources allocated for virtualization, while more expensive than buying the workstation. In my opinion, the price difference is still not large enough to interest the business. Because we still do a miscalculation of the project, and that is what comes out, unfortunately.
Limitations Of Virtualization – Licensing
I also want to address the issue of the limitations of virtualization, because it is actually having a lot of solutions and solutions are great, but they do have some of the shortcomings in my mind. In one of the projects we faced with the fact that since the beginning of something we virtualized, and then carefully re-read the license agreement with the vendor had to just buy the equipment, software and put back to use. Although there are various partnerships that avoid such conflicts. The next thing that we faced in their practice, this means all sorts of application security: hardware tokens and so on. Have to buy new hardware, additional software licenses that it was supported in a virtual environment, and this should be taken into account in the calculation of the project.
Pay Attention To The Order To Launch Applications
Pre-loaded application, I have already said, it is the another aspect that also need to pay attention – IT infrastructure services. The first urge: administrators sometimes put them in first place in a virtual environment, but then it turns out that it is not always a good thing. In case of any outages, failure make the things out of order and we have to start again to start our infrastructure.