The proliferation of specialized appliances for MySQL applications raises some interesting questions when viewed in light of the rest of the database market. Many of these appliances do not solve a general database problem, they are using better hardware to cover for deficiencies in the MySQL software design. For example, it is well-established that for OLTP workloads the other major Open Source database PostgreSQL will run rings around MySQL for throughput on a given piece of server hardware. Spending a lot of money coaxing extra performance out of MySQL that could be had by installing free alternatives is not a recipe for long-term success or viability, and one of the reasons there are so many MySQL accelerator startups out there is that it is so easy to find deficiencies that can be addressed with modest effort.
Specialized appliances and accelerated hardware have a place, in the database world I would argue that high-end analytics is one such place, but I have a hard time coming up with an economic argument for fixing software deficiencies with hardware solutions that would lead me to believe such companies are good venture investments.
No comments:
Post a Comment