 |  |  |  |  |  |  |  |  |  |  |  |
Can we do away with RDBM’s ?
Abhijit brings up an interesting thought on the possibility of scaling without a database:
"Again, SQL is wonderful for queries, especially dynamic queries. This could be countered with using either XPath style queries or a search engine like Lucene, or even a combination of the two. I consider a database is necessary today because of two reasons - along with the main content we serve a lot of related content and data integrity…..However, I am not able to find replacements for relating content and data integrity. Of course of this can be moved to the software, by building a domain model, but not all."
Here are some other reasons I can think of:
-RDBMS's have a huge legacy base built over the last couple of decades. OK OK. Mainframes were the legacy when RDBMS's came along. Right ? The crucial difference is that RDBMS's came along at pretty much the time PC's caught on. PC's were a huge inflexion point and RDBMS's played the ideal foil and piggybacked along just fine. By the time the web came and scalability got a whole new angle attached to it, RDBMS's had a good ten years to mature and were ready for the challenge. Takeway: I do not see another inflexion point that would expose an RDBMS weakness to the point that warrants another change.
- Google has proved that you can scale with a commodity database with the right kind of infrastructural support and architecture. Yes. They use MySql.
- "Nobody ever got fired for buying IBM" and so the adage goes. The same holds true for Oracle as well.
-Storage as a category has got too many interesting innovations happening already. The dollars have plenty of other more exciting options with a clearer ROI payoff.Innovation Products
"Again, SQL is wonderful for queries, especially dynamic queries. This could be countered with using either XPath style queries or a search engine like Lucene, or even a combination of the two. I consider a database is necessary today because of two reasons - along with the main content we serve a lot of related content and data integrity…..However, I am not able to find replacements for relating content and data integrity. Of course of this can be moved to the software, by building a domain model, but not all."
Here are some other reasons I can think of:
-RDBMS's have a huge legacy base built over the last couple of decades. OK OK. Mainframes were the legacy when RDBMS's came along. Right ? The crucial difference is that RDBMS's came along at pretty much the time PC's caught on. PC's were a huge inflexion point and RDBMS's played the ideal foil and piggybacked along just fine. By the time the web came and scalability got a whole new angle attached to it, RDBMS's had a good ten years to mature and were ready for the challenge. Takeway: I do not see another inflexion point that would expose an RDBMS weakness to the point that warrants another change.
- Google has proved that you can scale with a commodity database with the right kind of infrastructural support and architecture. Yes. They use MySql.
- "Nobody ever got fired for buying IBM" and so the adage goes. The same holds true for Oracle as well.
-Storage as a category has got too many interesting innovations happening already. The dollars have plenty of other more exciting options with a clearer ROI payoff.Innovation Products
Linking and Sharing
If you found this page useful, consider linking to it.
Simply copy and paste the code below into your web site (Ctrl+C to copy)
It will look like this: Can we do away with RDBM's ?
Trackback
Tell a friend
Good points there Prakash.
I have been thinking about this for quite some time. I wonder if there are scenarios where RDBMSs do not help in content modeling or retrieval, but primarily for performance. Database operations are faster than the traditional disk I/O. But if Robert McIntosh’s experiments can prove that scalability and performance is possible without databases it can open new doors.
Comment by Abhijit Nadgouda — April 17, 2007 @ 7:19 am
Thanks Abhijit for your comments.
Comment by Prakash — April 17, 2007 @ 5:02 pm