​MySQL: Percona plugs in TokuDB storage engine for big datasets

By using Tokutek's TokuDB storage engine instead of InnoDB, firms using Percona's MySQL fork are being told to expect to see a number of benefits

Peter Zaitsev: TokuDB instead of the InnoDB. Image: Percona

MySQL services firm Percona says the integration of Tokutek's database storage engine into its popular Percona Server MySQL fork will give users substantially better performance and compression.

The two firms have joined forces to offer Percona Server 5.6 with the option of TokuDB as a drop-in replacement for the standard InnoDB storage engine.

The resulting read-write performance shows a 20 times improvement over InnoDB, while data compression is up between five and 25 times, according to Percona.

"TokuDB is the only other MySQL open-source transactional storage engine, besides InnoDB, which I consider production-ready. It has truly innovative fractal trees technology and is also a very solid piece of engineering," Percona CEO Peter Zaitsev said.

"There are many demanding workloads that MySQL developers can use to evaluate the TokuDB storage engine as an alternative to InnoDB. Having TokuDB available with Percona Server makes this comparison easier."

Using a master-system with Percona Server, TokuDB and a PCIe card for the beta release of its Percona Cloud Tools, Percona said it achieved 500 million data-point insertions per day and hundreds of monitored instances.

It said those figures, together with others collected since the beta release, show the potential of TokuDB to help Percona Server users with large datasets and high read-write workloads, such as those found in applications running real-time advertising analytics, social media, and e-commerce.

Percona Server is an open-source MySQL alternative, which the company says has been downloaded 1.4 million times and is optimised for cloud computing, NoSQL access, and hardware such as SSD and Flash storage.

Tokutek ascribes the speed of open-source TokuDB to its use of a fractal-tree index data structure to aggregate database access requests at base nodes intelligently. The result is greater compression and reduced memory use.

MongoDB CTO: How our new WiredTiger storage engine will earn its stripes

With general availability of the next version of MongoDB looming as early as next month, CTO Eliot Horowitz believes it represents an important step for the open-source database.

Read More

"Many MySQL applications are ingesting very large volumes of data at ingestion rates never imagined 40 years ago, when B-tree indexing was conceived," Tokutek CEO John Partridge said in a statement.

"With a modern fractal-tree indexing engine, MySQL applications can scale up to meet the needs of these applications much more cost-effectively than alternative approaches."

Percona is also offering support for Percona Server with TokuDB, together with consulting services for companies that want to use the TokuDB storage engine in their MySQL environments.

More on databases

Show Comments