At this stage, the deadline is too far off to feel its dark shadow. However, that said it will come around sooner than I can imagine, especially when I will be juggling the workload of all the other courses as well. So I think it is time to define the research question and move forwards. In the end, this project is about going through the proposal process, not necessarily about coming up with a world-beating project idea 🙂 Knowing me, it will be good to have a ‘dry run’ at it anyway before I do it for real. My ideas and thoughts are very likely to bend and change as I go.
Application of Database Services in the Cloud
Cloud services are now well established as a viable solution for many tasks. There are a whole lot of platforms, Paas, SaaS, IaaS etc. etc. etc. and there seems to be a real growth area in the area of cloud databasing. While I think the advantages / disadvantages of cloud computing are well understood, there still seems to be a lot of work to do in applying this to databases, so this could be well worth looking into.
Questions to consider:
1. Who are the major vendors?
- Microsoft Azure
- Oracle Cloud
- Amazon Web Services
2. Do any of these providers offer a significant point of difference?
3. In brief – describe the advantages of cloud computing
4. What are the current limitations of cloud databases?
5. Integration with systems.
- What systems (or user cases) would benefit from working alongside a database in the cloud?
Why is it interesting to me?
I am definitely keen to learn more about databases. From a business (or employability) perspective it is probably well worth being familiar with the current trends and movements in databasing. However, this will not necessarily make me any more familiar with a particular DBMS, though I am not sure I could ever do enough in a research project that 6 months on the job wouldn’t surpass.
Three things I know about it
- Database technology is a proven performer. Relational databases in particular are based on solid foundation of algebra and calculus. (e.g. Sql Services have a 12 Billion row db here in Nelson, that they can query and return a result from in less than 1 sec!)
- There are multiple data storage mechanisms out there, of which relation databases hold the market share in the commercial / enterprise domain
- There are jobs in Nelson
Three things I believe about it
- I believe that working with databases takes an analytical mind-set.
- I believe it is a growth industry that underpins the rest of the IT world
- I believe that databases in the cloud have a very long way to go before they are a viable service for enterprises
Three things I don’t know about it
- The rest of the IT world is constantly changing. Mobile, flexible and scalable systems are now the norm, and yet the database industry seems static in comparison. Why?
- What do organisations REALLY use their databases for?
- And, do the data requirements define the storage mechanism (this is getting closer to the heart of my real project idea!)
My gut feeling is that cloud databases have an awfully long way to go before they can meet the storage needs of enterprises. Microsoft Azures 150 GB limit is a prime example of the current limitations. This is a critical factor when I begin to think seriously about my ultimate project choice and the ability to see the project through to a meaningful end. But saying that, investigating the possibilities in the cloud will be great background for my ultimate project.
In the end, I am still leaning towards investigating the different data storage methods and their characteristics that lend them to specific application domains. For example, an organisation that needs to a high level of structure and the ability to access all parts of its data would be silly not to use a relational database. But what if you don’t need to perform complex queries, or just need to store blobs of data (i.e. images, video…)? For now, I will file these thoughts safely away and let them mature. It is time to start the RES wheel turning.