|Facebook, Google, LinkedIn, and Twitter Collaborate Over WebScaleSQL For Massive Databases|
| 6:09 pm on Mar 27, 2014 (gmt 0)|
|WebScaleSQL currently includes contributions from MySQL engineering teams at Facebook, Google, LinkedIn, and Twitter. Together, we are working to share a common base of code changes to the upstream MySQL branch that we can all use and that will be made available via open source. |
| 10:10 pm on Mar 28, 2014 (gmt 0)|
No software engineer here, but MySql acts and feels so very "outdated" considering today's web realities, even at a comparatively small scale. This initiative should have taken place already ages ago.
| 1:08 am on Mar 29, 2014 (gmt 0)|
I run a site with a large (>250GB) frontend (MySQL) database. Some of the tables have tens of millions of rows. Queries are fast but this development looks interesting.
| 9:22 am on Mar 29, 2014 (gmt 0)|
Heisje, I think I know, what you mean - yet MySQL is a fantastic tool and database to work with. It has a huge user base, is stable and has adopted to a lot of changes since I first saw it 1995 (where it already set standards).
I love stable and flexible technology. For example: I jumped on the GFS (file system) very early after it was called "stable" and went back to NFS a few days later. Why? Because NFS is stone old and has absolutely no surprises, no lockups, no sudden deaths. I got one internal company server which runs for 4 years unresetted.
Long story short: MySQL is the AK-47 of Internet machine guns. It may not be very pretty, but it works underwater, too.
The scalability is the only thing that bothers a lot of developers and with that workgroup they address it. Perfect!
| 10:24 am on Mar 29, 2014 (gmt 0)|
I doubt it'll make MySQL any more scalable for "the rest of us", though.
| 2:22 pm on Mar 29, 2014 (gmt 0)|
Growth can be key for any website that operates on an already large mysql database. Your application may work just fine today, but this may change looking towards the future. What if your database grows by 25% per year?
It will be worth keeping an aye on WebscaleSQL to work out if and when it is worth testing.