Was this page helpful?
Caution
You're viewing documentation for a previous version of ScyllaDB Open Source. Switch to the latest stable version.
This document describes the compaction strategy options available when creating a table. For more information about creating a table in Scylla, refer to the CQL Reference.
By default, Scylla starts a compaction task whenever a new SSTable is written. Compaction merges several SSTables into a new SSTable, which contains only the live data from the input SSTables. Merging several sorted files to get a sorted result is an efficient process, and this is the main reason why SSTables are kept sorted.
The following compaction strategies are supported by Scylla:
Size-tiered Compaction Strategy (STCS)
Leveled Compaction Strategy (LCS)
Incremental Compaction Strategy (ICS)
Time-window Compaction Strategy (TWCS)
This page concentrates on the parameters to use when creating a table with a compaction strategy. If you are unsure which strategy to use or want general information on the compaction strategies which are available to Scylla, refer to Compaction Strategies.
The following options are available for all compaction strategies.
compaction = {
'class' : 'compaction_strategy_name',
'enabled' : (true | false),
'tombstone_threshold' : ratio,
'tombstone_compaction_interval' : sec,
'unchecked_tombstone_compaction' : (true | false)}
class (default: SizeTieredCompactionStrategy)Selects the compaction strategy.
It can be one of the following. If you are unsure which one to choose, refer to Compaction Strategies :
SizeTieredCompactionStrategy
TimeWindowCompactionStrategy
LeveledCompactionStrategy
enabled (default: true)Runs background compaction (known as “minor compaction”). Can be one of the following:
true - runs minor compaction
false - disable minor compaction
tombstone_threshold (default: 0.2)The ratio (expressed as a decimal) of garbage-collectable tombstones compared to the data. When this threshold is exceeded on a specific table, a single SSTable compaction begins. Acceptable values are numbers in the range 0 -1.
tombstone_compaction_interval (default: 86400s (1 day))tombstone_compaction_interval is lower-bound for when a new tombstone compaction can start. If an SSTable was compacted at a time X, the earliest time it will be considered for tombstone compaction again is X + tombstone_compaction_interval. This does not guarantee that sstables will be considered for compaction immediately after tombstone_compaction_interval time has elapsed after the last compaction.
unchecked_tombstone_compaction (default: false)If unchecked_tombstone_compaction is set to true, tombstone_threshold will be ignored, and an SSTable will be compacted according to tombstone_compaction_interval.
When using STCS, SSTables are put in different buckets depending on their size. When an SSTable is bucketed, the average size of the tables is compared to the new table as well as the high and low threshold levels.
The database compares each SSTable size to the average of all SSTable sizes on the node. It calculates bucket_low * avg_bucket_size and bucket_high * avg_bucket_size and then compares the result with the average SSTable size. The conditions set for bucket_high and bucket_low dictate if successive tables will be added to the same bucket. When compaction begins it merges SSTables whose size in KB are within [average-size * bucket_low] and [average-size * bucket_high].
Once the min_threashold is reached, minor compaction begins.
The following options only apply to SizeTieredCompactionStrategy:
compaction = {
'class' : 'SizeTieredCompactionStrategy',
'bucket_high' : factor,
'bucket_low' : factor,
'min_sstable_size' : int,
'min_threshold' : num_sstables,
'max_threshold' : num_sstables}
bucket_high (default: 1.5)A new SSTable is added to the bucket if the SSTable size is less than bucket_high * the average size of that bucket (and if the bucket_low condition also holds).
For example, if ‘bucket_high = 1.5’ and the SSTable size = 14MB, does it belong to a bucket with an average size of 10MB?
Yes, because the SSTable size = 14`, which is less than ‘bucket_high’ * average bucket size = 15.
So, the SSTable will be added to the bucket, and the bucket’s average size will be recalculated.
bucket_low (default: 0.5)A new SSTable is added to the bucket if the SSTable size is more than bucket_low* the average size of that bucket (and if the bucket_high condition also holds).
For example, if ‘bucket_high = 0.5’ and the SSTable size is 10MB, does it belong to a bucket with an average size is 15MB?
Yes, because the SSTable size = 10 which is more than ‘bucket_low’ * average bucket size = 7.5.
So, the SSTable will be added to the bucket, and the bucket’s average size will be recalculated.
min_sstable_size (default: 52,428,800)All SSTables smaller than this number of bytes are put into the same bucket.
min_threshold (default: 4)Minimum number of SSTables that need to belong to the same size bucket before compaction is triggered on that bucket. If your SSTables are small, use min_sstable_size to define a size threshold (in bytes) below which all SSTables belong to one unique bucket.
Note
Enforcement of min_threshold is controlled by the compaction_enforce_min_threshold configuration option in the scylla.yaml configuration settings.
By default, compaction_enforce_min_threshold: false, meaning the Size-Tiered Compaction Strategy will compact any bucket containing 2 or more SSTables.
Otherwise, if compaction_enforce_min_threshold: true, the value of min_threshold is considered and only those buckets that contain at
least min_threshold SSTables will be compacted.
max_threshold (default: 32)Maximum number of SSTables that will be compacted together in one compaction step.
The compaction class LeveledCompactionStrategy (LCS) creates SSTables of a fixed, relatively small size (160 MB by default) that are grouped into levels. Within each level, SSTables are guaranteed to be non-overlapping. Each level (L0, L1, L2 and so on) is ten times as large as the previous level.
compaction = {
'class' : 'LeveledCompactionStrategy',
'sstable_size_in_mb' : int}
sstable_size_in_mb (default: 160)This is the target size in megabytes, that will be used as the goal for an SSTable size following a compression. Although SSTable sizes should be less or equal to sstable_size_in_mb, it is possible that compaction could produce a larger SSTable during compaction. This occurs when data for a given partition key is exceptionally large.
ICS is only available in ScyllaDB Enterprise. See the ScyllaDB Enetrpise documentation for details.
The basic concept is that TimeWindowCompactionStrategy will create 1 SSTable per file for a given time window.
Caution
We strongly recommend using a single TTL value for any given table.
This means sticking to the default time to live as specified in the table’s schema.
Using multiple TTL values for a given table may lead to inefficiency when purging expired data, because an SSTable will remain until all of its data is expired.
Tombstone compaction can be enabled to remove data from partially expired SSTables, but this creates additional WA (write amplification).
Caution
Avoid overwriting data and deleting data explicitly at all costs, as this can potentially block an expired SSTable from being purged, due to the checks that are performed to avoid data resurrection.
compaction = {
'class' : 'TimeWindowCompactionStrategy',
'compaction_window_unit' : string,
'compaction_window_size' : int,
'expired_sstable_check_frequency_seconds' : int,
'min_threshold' : num_sstables,
'max_threshold' : num_sstables}
compaction_window_unit (default: DAYS)A time unit used to determine the window size which can be one of the following:
'MINUTES'
'HOURS'
'DAYS'
compaction_window_size (default: 1)The number of units which will make up a window.
expired_sstable_check_frequency_seconds (default: 600)Specifies (in seconds) how often Scylla will check for fully expired SSTables, which can be immediately dropped.
min_threshold (default: 4)Minimum number of SSTables that need to belong to the same size bucket before compaction is triggered on that bucket.
max_threshold (default: 32)Maximum number of SSTables that will be compacted together in one compaction step.
Copyright
© 2016, The Apache Software Foundation.
Apache®, Apache Cassandra®, Cassandra®, the Apache feather logo and the Apache Cassandra® Eye logo are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries. No endorsement by The Apache Software Foundation is implied by the use of these marks.
Was this page helpful?