diff options
| author | Henning Baldersheim <balder@yahoo-inc.com> | 2024-02-02 22:17:59 +0000 |
|---|---|---|
| committer | Henning Baldersheim <balder@yahoo-inc.com> | 2024-02-03 11:19:23 +0000 |
| commit | 62bee4e38718e3fd670908f09cbf14ca0acc69cf (patch) | |
| tree | 9748d39952e08de123634add41760da75bbc89c1 /configdefinitions | |
| parent | 25e2abf250c7aa1439b4bf7a0e6bd2d4a8a991f9 (diff) | |
throttle_individual_merge_feed_ops has long been enabled, cleaning up
Diffstat (limited to 'configdefinitions')
| -rw-r--r-- | configdefinitions/src/vespa/stor-filestor.def | 7 |
1 files changed, 2 insertions, 5 deletions
diff --git a/configdefinitions/src/vespa/stor-filestor.def b/configdefinitions/src/vespa/stor-filestor.def index cefce5fc648..93650d9a328 100644 --- a/configdefinitions/src/vespa/stor-filestor.def +++ b/configdefinitions/src/vespa/stor-filestor.def @@ -27,6 +27,7 @@ response_sequencer_type enum {LATENCY, THROUGHPUT, ADAPTIVE} default=ADAPTIVE re ## When merging, if we find more than this number of documents that exist on all ## of the same copies, send a separate apply bucket diff with these entries ## to an optimized merge chain that guarantuees minimum data transfer. +## TODO GC and hardcode common_merge_chain_optimalization_minimum_size int default=64 restart ## Chunksize to use while merging buckets between nodes. @@ -37,6 +38,7 @@ bucket_merge_chunk_size int default=16772216 restart ## Whether or not to enable the multibit split optimalization. This is useful ## if splitting is expensive, but listing document identifiers is fairly cheap. ## This is true for memfile persistence layer, but not for vespa search. +## TODO verify its usage enable_multibit_split_optimalization bool default=true restart ## Whether or not to use async message handling when scheduling storage messages from FileStorManager. @@ -71,8 +73,3 @@ async_operation_throttler.window_size_backoff double default=0.95 async_operation_throttler.min_window_size int default=20 async_operation_throttler.max_window_size int default=-1 # < 0 implies INT_MAX async_operation_throttler.resize_rate double default=3.0 -## If true, each put/remove contained within a merge is individually throttled as if it -## were a put/remove from a client. If false, merges are throttled at a persistence thread -## level, i.e. per ApplyBucketDiff message, regardless of how many document operations -## are contained within. -async_operation_throttler.throttle_individual_merge_feed_ops bool default=true |