issue: #38142
pr: #38143
current balance channel policy only consider current collection's
distribution, so if all collections has 1 channel, and all channels has
been loaded on same querynode, after querynode num increase, balance
channel won't be triggered.
This PR enable score based balance channel policy, to achieve:
1. distribute all channels evenly across multiple querynodes
2. distribute each collection's channel evenly across multiple
querynodes.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
sparse vectors may have arbitrary number of non zeros and it is hard to
optimize without knowing the actual distribution of nnz. this PR adds a
metric for analyzing that.
pr: #38329
also fixed a bug of sparse when searching by pk
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
issue: #37764
pr: #37815
also see: #38259
- add a local client to call local server directly for
querycoord/rootcoord/datacoord.
- enable local client if milvus is running mixcoord or standalone mode.
- after removing rpc layer from mixcoord, the querycoord at standby mode
will be blocked forever of deployment rolling
---------
Signed-off-by: chyezh <chyezh@outlook.com>
enhance :
alterindex delete properties
We have introduced a new parameter deleteKeys to the alterindex
functionality, which allows for the deletion of properties within an
index. This enhancement provides users with the flexibility to manage
index properties more effectively by removing specific keys as needed.
altercollection delete properties
We have introduced a new parameter deleteKeys to the altercollection
functionality, which allows for the deletion of properties within an
collection. This enhancement provides users with the flexibility to
manage collection properties more effectively by removing specific keys
as needed.
3.support altercollectionfield
We currently support modifying the fieldparams of a field in a
collection using altercollectionfield, which only allows changes to the
max-length attribute.
Key Points:
New Parameter - deleteKeys: This new parameter enables the deletion of
specified properties from an index. By passing a list of keys to
deleteKeys, users can remove the corresponding properties from the
index.
Mutual Exclusivity: The deleteKeys parameter cannot be used in
conjunction with the extraParams parameter. Users must choose one
parameter to pass based on their requirement. If deleteKeys is provided,
it indicates an intent to delete properties; if extraParams is provided,
it signifies the addition or update of properties.
issue: https://github.com/milvus-io/milvus/issues/37436
pr: https://github.com/milvus-io/milvus/pull/37437
---------
Signed-off-by: Xianhui.Lin <xianhui.lin@zilliz.com>
1. taskQueueCapacity 256 is too small for production when we want to
re-write the entire collection
2. tasks should be cleaned when unable to recover, or the meta will
remain in etcd forever later.
pr: #37896
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
issue: #33550
pr: #37850
balance segment and balance segment execute at same time, which will
cause bounch of corner case.
This PR disable simultaneous balance of segments and channels
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Cherry-pick from master
pr: #37416
See also #37404#37402
IP address in paramtable need validation and fail fast with reasonable
error message
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Timeout is a bad design for long running tasks, especially using a
static timeout config. We should monitor execution progress and fail the
task if the progress has been stale for a long time.
This pr is a small patch to stop DC from marking compaction tasks
timeout, while still waiting for DN to finish. The design is
self-conflicted. After this pr, mix and L0 compaction are no longer
controlled by DC timeout, but clustering is still under timeout control.
The compaction queue capacity grows larger for priority calc, hence
timeout compactions appears more often, and when timeout, the queuing
tasks will be timeout too, no compaction will success after.
See also: #37108, #37015
pr: #37118
---------
---------
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
5 percent of free memory is too less for l0 compaction. This pr will
raise it to 50 percent.
See also: #36614
pr: #36690
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
issue: #35859
pr: #36181
This PR introduce two new param: toleranceFactor and checkRequestNum,
after every checkRequestNum request has been assigned, try to compute
querynode's workload score.
if the diff is less than the toleranceFactor, replica selection policy
will fallback to round_robin, which reduce the average cost to about
500ns.
if the diff is larger than the toleranceFactor, replica selection policy
will compute querynode's score to select the target node with smallest
score in every assigment.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Cherry-pick from master
pr: #36189
Related to #35303
This PR add a param item to support change l0 forward behavior from bf
filtering and forward to remote load.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
