The second limitation of traditional automated tiering involves how both the system itself and the data migration process are managed. Most systems require the IT department to determine which data sets will be migrated at what times between the tiers. This might be part of a regular schedule (e.g. a certain day of the week or month), or on an ad-hoc basis as needed (e.g. an event or project scheduled for the following day). Either way, a sub-optimal determination inevitably results in minimal improvement in performance.
SSD tiering extends cache performance
For years, flash technology has been used as cache to improve storage array performance. Performance gains from cache are limited by the practical size of a cache memory. Overcoming these limitations will extend the performance gains afforded by caching into the SSD storage tier as shown in the diagram below. Note how the practical size of a cache limits its ability to achieve gains beyond a certain point.
The SSD's far larger capacities (in the hundreds of Gigabytes range), which are also available at a significantly lower cost-per-byte than cache memory, make it possible to scale the performance gains considerably. But this will only be the case if the migration of data between tiers can be made sufficiently intelligent and dynamic to keep pace with the real-time changes in hot data.
How real-time tiering works
Real-time tiering utilizes three separate processes, all of which operate in an autonomic manner in real-time, including:
- Scoring to maintain a current page ranking on each and every I/O using an efficient process that adds less than one microsecond of overhead. The algorithm takes into account both the frequency and recency of access. For example, a page that has been accessed five times in the last 100 seconds would get a high score.
- Sorting of all high-scoring pages occurs every five seconds, utilizing less than 1 percent of the system's CPU. Those pages with the highest scores then become candidates for promotion to the higher-performing SSD tier.
- Migration is the process that actually moves, or migrates, the pages: high scoring pages from hard disk drive to SSD and low scoring pages from SSD back to hard disk. Less than 80MB of data are moved during any five-second sort, so the impact on overall system performance is minimal.
Measuring performance advantages of real-time tiering
Real-time tiering solutions can deliver up to 100,000 random read and 32,000 random write I/Os per second. The chart below shows the potential performance gains achievable with real-time tiered storage. Naturally, the higher the "hit rate" in the SSD tier, the higher the gains. But even a conservative hit rate of 70% (easily attainable in most environments) can deliver a 3x improvement in application performance. Far greater performance gains can be realized when the SSD tier handles 80% or more of the I/O load, even though that tier represents only 5% to 10% of the system capacity.
Sign up for CIO Asia eNewsletters.