Big Data – Big Help!!
I am trying to reach out to my tweep and fellow Big Data experts via this blog post. This is for my buddy @ki_run who is looking for little help starting off with Big Data. Details are below. Really appreciate your effort in reading and providing any help you can.
Background
[icon_list style=”star”]
- He plans to implement REST API – Meaning a server side application with which clients will interact using REST calls. He has freedom of choosing the implementation stack. And he opted to choose J2EE with JAX-RS as the application layer (After all, my friend 🙂 )
- He has experience using traditional databases like MySQL, DB2 (what a surprise! 😀 ) etc. But since this particular application may need to scale up – even to cloud-like numbers; He is considering going radical and experimenting with NoSQL databases instead.
- He has fair amount of knowledge about CouchDB, Mongo (with its JSON-like API) and has read about Hadoop.
[/icon_list]
Questions
[icon_list style=”star”]
- We see there are several names like NoSQL, Hadoop, Big Data being thrown around. There are many blogs being written about the advantages they bring in. But there is only so much we know (technically) about how to use these technologies to our advantage. What are the first steps (A Hello World approach??) we need to take to get started with Big Data ?
- In traditional SQL world, we have ORMs – Object-Relational Mappings. These API’s allow us to keep the actual underlying database hidden from the application layer so that tomorrow we can replace the database product (say, throw out Oracle and put in MySQL). Do you know about equivalents in NoSQL world? Concern here is that if the NoSQL does not work fine, tomorrow we should be able to revert back to traditional SQL database without having to re-code the application layer. (Fair enough expectation I guess!)
[/icon_list]
Do reach out to me or my friend directly if you have answers to any or all of the questions above. Thanks for your time.
Note: Extracted information from the original email I received by @ki_run. Lazy me!
COMMENTS
Leave A Comment
RECENT POSTS
Composable Applications Explained: What They Are and Why They Matter
Composable applications are customized solutions created using modular services as the building blocks. Like how...
Is ChatGPT a Preview to the Future of Astounding AI Innovations?
By now, you’ve probably heard about ChatGPT. If you haven’t kept up all the latest...
How MDM Can Help Find Jobs, Provide Better Care, and Deliver Unique Shopping Experiences
Industrial data is doubling roughly every two years. In 2021, industries created, captured, copied, and...
Thanks for reaching out to the community Prashant. After I sent you the mail, I did more research and came across Kundera, which is a JPA compliant ORM for both RDBMS and NoSQL databases (Cassandra, MongoDB and HBase).
The question remains though: in the RDBMS world, things are well-defined. SQL is a standard. Various database server vendors differ only slightly in the interface they present to the developer. In the NoSQL world, things are significantly different.As far as I can tell, every product has its own unique mechanism for the developer to interact with the data (some are just object stores themselves!). So, I’m confused as to how does an object-mapping figure in the picture?
And not to forget performance. The whole point of going the NoSQL path is performance. Take MongoDB for instance. It has a JSON based API for storing/retrieving data. I’m wondering whether the overhead of an object-mapper on top of this is going to be feasible. It looks to me like the overhead in this case will be more than that we see in case of traditional ORM’s over SQL databases.
Lots of confusion. Lot of doubt 🙂
More research, more confusion 🙂
So far I have been using JPA. I had heard about JDO but I was always under the impression that JDO has been superseded by JPA. Turns out it is not so. JPA is specific to RDBMS while JDO can handle any data store (RDBMS, Excel, HBase .. whatever). So, JPA is an RDBMS-specific subset of JDO!
DataNuclues seems to provide an implementation of both JDO and JPA, and it seems to fulfil my needs. Will look into it and let you know how it went.
Good stuff Kiran. I know how it is to be working on a “current & happening” technology with least amount of help available. Reminds me of a day in lab.. isn’t it? 🙂
Keep this post updated on your findings. And I’m sorry I couldn’t be of much help here at this point. I am still hoping someone reaches out to you and answers your doubts. I will keep re-posting links on twitter.