Home > Storage > PowerScale (Isilon) > Product Documentation > Storage (general) > PowerScale OneFS SmartFlash > L3 cache sizing
A large-scale storage system must provide the performance required for various workflows and datasets. Figuring out the size of the active data, or working set, for your environment is the first step in an L3 cache SSD sizing exercise.
L3 cache uses all available SSD space over time. As a rule, L3 cache benefits more with more available SSD space. However, sometimes losing spindle count hurts more than adding cache helps a workflow. If possible, add a larger capacity SSD rather than multiple smaller SSDs.
An L3 cache sizing exercise involves calculating the correct amount of SSD space to fit the working dataset. This calculation can be done by using the isi_cache_stats command to periodically capture L2 cache statistics on an existing cluster.
Run the following commands based on the workload activity cycle, at job start and job end. Initially run isi_cache_stats –c to reset, or zero out, the counters. Then run isi_cache_stats –v at workload activity completion and save the output. Observing the L2 cache miss rates for both data and metadata on a single node will help determine the working dataset size and the correct amount of SSD space.
Note: These cache miss counters are displayed as 8 KB blocks. As such, an L2_data_read.miss value of 1024 blocks represents 8 MB of missed data.
The formula for calculating the working set size is:
(L2_data_read.miss + L2_meta_read.miss) = working_set size
Once the working set size has been calculated, a good guideline is to size L3 cache SSD capacity per node according to the following formula:
L2 capacity + L3 capacity >= 150% of working set size.
There are diminishing returns for L3 cache after a certain point. When an SSD-to-working-set size ratio becomes too high, the cache hits decrease and fail to add greater benefit. Conversely, when compared to SmartPools SSD strategies, another benefit of using SSDs for L3 cache is that performance degrades more gracefully if metadata happens to exceed the available SSD capacity.
Note: L3 cache is not applicable for nodes containing 16 or more SSDs.