The MariaDB open source database team and software optimized storage company Fusion-io have partnered on atomic writes capabilities. Atomic writes refers to an operation in which a processor can simultaneously write multiple independent storage sectors as a single transaction.
White PapersMore >>
- Step Up Your Game in Loan Operations in 2014
- IT and LOB Win When Your Business Adopts Flexible Social Cloud Collaboration Tools
This new function claims to achieve performance and cost savings through a "flash-aware architecture", further differentiating MariaDB from more expensive proprietary databases.
NOTE: Steve Wozniack is chief scientist at Fusion-io.
Atomic writes prevent data losses from MariaDB (and related) databases in certain infrequent events, such as an unexpected power outage midway through a database transaction, without the need for mechanisms like the "double write buffer" that reduces speed and the life of flash media.
MariaDB team blogger Axel Schwenke explained that currently, in order to use atomic writes, it is necessary to use the DirectFS file system, which is a part of the Fusion IO SDK. "Wlad Vaintroub from Monty Program AB, in cooperation with FusionIO developers, implemented the necessary changes in InnoDB/XtraDB to use the new feature," writes Schwenke.
The MariaDB engineering team used Fusion-io's flash-aware APIs to optimize the InnoDB and XtraDB storage engines for atomic writes. The resulting scalable solution uses the underlying flash translation layer for atomicity of database updates.
Monty Widenius, MariaDB creator said, "Increasingly our customers expect MariaDB products to not just compete with, but to exceed what they can get from rival database technologies. The highly innovative solutions we have worked on with Fusion-io are a great example of how both companies are bringing the best thinking to the best database in the world."
Fusion-io's Nisha Talagala said that with a software-centric approach, her team can leverage the potential of flash to develop new features that accelerate databases and other applications in ways previously not possible.