To allow better control the cache eviction process.
Introduce MINIO_CACHE_WATERMARK_LOW and
MINIO_CACHE_WATERMARK_HIGH env. variables to specify
when to stop/start cache eviction process.
Deprecate MINIO_CACHE_EXPIRY environment variable. Cache
gc sweeps at 30 minute intervals whenever high watermark is
reached to clear least recently accessed entries in the cache
until sufficient space is cleared to reach the low watermark.
Garbage collection uses an adaptive file scoring approach based
on last access time, with greater weights assigned to larger
objects and those with more hits to find the candidates for eviction.
Thanks to @klauspost for this file scoring algorithm
Co-authored-by: Klaus Post <klauspost@minio.io>
MINIO_CACHE_DRIVES: List of mounted cache drives or directories delimited by ","
MINIO_CACHE_EXCLUDE: List of cache exclusion patterns delimited by ","
MINIO_CACHE_EXPIRY: Cache expiry duration in days
MINIO_CACHE_QUOTA: Maximum permitted usage of the cache in percentage (0-100).
MINIO_CACHE_AFTER: Minimum number of access before caching an object.
MINIO_CACHE_WATERMARK_LOW: % of cache quota at which cache eviction stops
MINIO_CACHE_WATERMARK_HIGH: % of cache quota at which cache eviction starts
...
...
7. Start MinIO gateway to s3 with edge caching enabled on '/mnt/drive1', '/mnt/drive2' and '/mnt/export1 ... /mnt/export24',
exclude all objects under 'mybucket', exclude all objects with '.pdf' as extension
with expiry up to 40 days. Cache only those objects accessed atleast 3 times.
exclude all objects under 'mybucket', exclude all objects with '.pdf' as extension. Cache only those objects accessed atleast 3 times. Garbage collection triggers in at high water mark (i.e. cache disk usage reaches 90% of cache quota) or at 72% and evicts oldest objects by access time until low watermark is reached ( 70% of cache quota) , i.e. 63% of disk usage.
- Disk cache size defaults to 80% of your drive capacity.
- Disk cache quota defaults to 80% of your drive capacity.
- The cache drives are required to be a filesystem mount point with [`atime`](http://kerolasa.github.io/filetimes.html) support to be enabled on the drive. Alternatively writable directories with atime support can be specified in MINIO_CACHE_DRIVES
- Expiration of each cached entry takes user provided expiry as a hint, and defaults to 90 days if not provided.
- Garbage collection sweep of the expired cache entries happens whenever cache usage is > 80% of drive capacity, GC continues until sufficient disk space is reclaimed.
- Garbage collection sweep happens whenever cache disk usage reaches high watermark with respect to the configured cache quota , GC evicts least recently accessed objects until cache low watermark is reached with respect to the configured cache quota. Garbage collection runs a cache eviction sweep at 30 minute intervals.
- An object is only cached when drive has sufficient disk space.
Disk caching can be enabled by setting the `cache` environment variables for MinIO gateway . `cache` environment variables takes the mounted drive(s) or directory paths, cache expiry duration (in days) and any wildcard patterns to exclude from being cached.
Disk caching can be enabled by setting the `cache` environment variables for MinIO gateway . `cache` environment variables takes the mounted drive(s) or directory paths, any wildcard patterns to exclude from being cached,low and high watermarks for garbage collection and the minimum accesses before caching an object.
Following example uses `/mnt/drive1`, `/mnt/drive2` ,`/mnt/cache1` ... `/mnt/cache3` for caching, with expiry up to 90 days while excluding all objects under bucket `mybucket` and all objects with '.pdf' as extension while starting a s3 gateway setup. Objects are cached if they have been accessed three times or more.Cache max usage is restricted to 80% of disk capacity in this example.
Following example uses `/mnt/drive1`, `/mnt/drive2` ,`/mnt/cache1` ... `/mnt/cache3` for caching, while excluding all objects under bucket `mybucket` and all objects with '.pdf' as extension on a s3 gateway setup. Objects are cached if they have been accessed three times or more.Cache max usage is restricted to 80% of disk capacity in this example. Garbage collection is triggered when high watermark is reached - i.e. at 72% of cache disk usage and clears least recently accessed entries until the disk usage drops to low watermark - i.e. cache disk usage drops to 56% (70% of 80% quota)