Database management systems (DBMS) are one of the core components of almost every company today. However, the license and support costs for these systems can be considerable and exacerbate the already tense budget situation in IT departments, while the vendor lock-in that accompanies some license agreements has also become a real inhibitor to business change. With these factors driving many organisations to look for more flexible alternatives, TmaxSoft, explains what IT decision-makers should look for when switching databases.
Almost imperceptibly, many companies have entered a price spiral in which the support and maintenance prices of their core database systems are increasing year by year. Existing DBMS licenses are often based on the maximum number of server cores and not on the actual usage.
With the trend towards Software-Defined Data Centres (SDDC), virtualisation, and hyper-convergence, where multiple virtualisation solutions such as VMware, Hyper V, and Xen are consolidated on the same hardware, many companies in practice only use part of the computing power available from their server, but still pay the full server capacity and the extra hardware needed to run workloads. This inevitably leads to database licensing costs that are disproportionate to the actual capacity used and, in doing so, consume valuable budgets that are urgently needed for IT innovation.
By re-examining their DBMS, businesses can reduce the cost of these systems by as much as 60 per cent, which can then be used to fuel IT innovation. A flexible and, above all, user-friendly licensing model can be a strategic lever to divert funds from software licensing costs to digital transformation. However, not all license models for databases and DBMS have the same structure and it can be difficult to determine the best way forward.
With this in mind, TmaxSoft has set out the key criteria that IT leaders should consider when deciding to buy or change:
1 – Beware of potentially hidden hardware costs for running workloads. Some solutions appear to be “open”, but on closer inspection turn out to be restrictive, which is always a disadvantage for the user.
2 – Be sure to anticipate the actual cost of running a custom application or workload. Here, the real costs are usually much higher than what a company originally projected.
3 – Check the ratio of your licensed and actually used resources. Companies should pay only for the resources actually used by their database – not for 100 percent of their potential processing power, as with many licensing models.
4 – Virtualisation. Certifications such as “VMware Ready ™” show that the database supports building virtual data centres through appropriate open cloud infrastructures. In particular, check which vendor-independent virtualisation solutions the DBMS vendor recognises as “hard-partitioning” to ensure that only the processors assigned to them need to be licensed, not the entire server / server network.
5 – Intelligent business analytics based on big data are increasingly becoming core requirements for database systems. This requires particularly fast access times, which are realised by in-memory databases (IMDB), since they keep their data permanently in the working memory. However, many manufacturers of such solutions only offer them as an asset package of hardware and software and thus tie their customers to years of hardware support. Therefore, ask for purely software-based alternatives that run manufacturer-independent on standard hardware with sufficient performance and storage capacity.
6 – When changing the database system, the compatibility with the previous solution should be checked – especially in software clustering, commands, data types and SQL extensions. The more compatible the old system and the new system are, the easier and quicker the system change.
7 – Open Source is not the right solution for every business. Although the appropriate DBMS solutions are often well suited to a specific niche, other users experience issues that did not need to be addressed by the original target audience. It is therefore important to consider the support model of a database solution, as Open Source databases often require a lot of time and resources.