astelm/minio
1
0
Fork 0
mirror of https://github.com/pgsty/minio.git synced 2026-10-06 17:55:59 +03:00
Code Issues Packages Projects Releases Wiki Activity

add data update tracking using bloom filter (#9208)

Browse Source 
By monitoring PUT/DELETE and heal operations it is possible
to track changed paths and keep a bloom filter for this data. 

This can help prioritize paths to scan. The bloom filter can identify
paths that have not changed, and the few collisions will only result
in a marginal extra workload. This can be implemented on either a
bucket+(1 prefix level) with reasonable performance.

The bloom filter is set to have a false positive rate at 1% at 1M 
entries. A bloom table of this size is about ~2500 bytes when serialized.

To not force a full scan of all paths that have changed cycle bloom
filters would need to be kept, so we guarantee that dirty paths have
been scanned within cycle runs. Until cycle bloom filters have been
collected all paths are considered dirty.
This commit is contained in:
Klaus Post
2020-04-27 19:06:21 +02:00
committed by GitHub
parent eff4127efd
commit 073aac3d92
24 changed files with 1270 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files
cmd/xl-v1-object.go
+6
View File
@@ -69,6 +69,8 @@ func (xl xlObjects) CopyObject(ctx context.Context, srcBucket, srcObject, dstBuc
// Check if this request is only metadata update.
if cpSrcDstSame {
defer ObjectPathUpdated(path.Join(dstBucket, dstObject))
// Read metadata associated with the object from all disks.
storageDisks := xl.getDisks()
@@ -481,6 +483,7 @@ func (xl xlObjects) PutObject(ctx context.Context, bucket string, object string,
// putObject wrapper for xl PutObject
func (xl xlObjects) putObject(ctx context.Context, bucket string, object string, r *PutObjReader, opts ObjectOptions) (objInfo ObjectInfo, err error) {
defer ObjectPathUpdated(path.Join(bucket, object))
data := r.Reader
uniqueID := mustGetUUID()
@@ -695,6 +698,7 @@ func (xl xlObjects) putObject(ctx context.Context, bucket string, object string,
func (xl xlObjects) deleteObject(ctx context.Context, bucket, object string, writeQuorum int, isDir bool) error {
var disks []StorageAPI
var err error
defer ObjectPathUpdated(path.Join(bucket, object))
tmpObj := mustGetUUID()
if bucket == minioMetaTmpBucket {
@@ -755,6 +759,7 @@ func (xl xlObjects) doDeleteObjects(ctx context.Context, bucket string, objects
if errs[idx] != nil {
continue
}
tmpObjs[idx] = mustGetUUID()
var err error
// Rename the current object while requiring
@@ -774,6 +779,7 @@ func (xl xlObjects) doDeleteObjects(ctx context.Context, bucket string, objects
if err != nil {
errs[idx] = err
}
ObjectPathUpdated(path.Join(bucket, objects[idx]))
}
}