Last active
July 14, 2019 16:34
-
-
Save amuraru/b750f54e55bb7208c42d to your computer and use it in GitHub Desktop.
HBase Merge Empty regions
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/python | |
| ## hbase shell <<<'status "detailed"' > /tmp/hbase_regions | |
| ## python merge_adjacent_regions.py <table_name> <target_region_size_mb> | |
| import collections | |
| import sys | |
| TABLE=sys.argv[1] | |
| target_size_mb = int(sys.argv[2]) | |
| LOCKED_REGION='07d9c26d0dec29cac5fc9c193ebe1889' | |
| r=open('/tmp/hbase_regions') | |
| lines=r.readlines() | |
| regions_parsed=[] | |
| Region=collections.namedtuple('region', 'fqn start_key id size info') | |
| for i in range(len(lines)): | |
| if "%s," % TABLE not in lines[i]: | |
| continue; | |
| fqn=lines[i].strip(' "') | |
| start_key = fqn.split(',')[1] | |
| try: | |
| id = fqn.split('.')[1] | |
| except: | |
| print fqn | |
| sys.exit(1) | |
| info=dict(item.strip().split("=") for item in lines[i+1].split(",")) | |
| regions_parsed.append( | |
| Region( fqn=fqn, | |
| start_key=start_key, | |
| id=id, | |
| size=int(info['storefileSizeMB']), | |
| info=info)) | |
| i+=2 | |
| sorted_regions=sorted(regions_parsed, key=lambda r: r.start_key) | |
| size = 0 | |
| merged_regions = [] | |
| adjacent_regions = [] | |
| idx = 0 | |
| while idx < len(sorted_regions): | |
| region = sorted_regions[idx] | |
| size += region.size | |
| if size > target_size_mb or region.id == LOCKED_REGION: | |
| # group size over target, roll-over | |
| if len(adjacent_regions) > 1: | |
| merged_regions.append(adjacent_regions) | |
| adjacent_regions = [] | |
| size = 0 | |
| else: | |
| adjacent_regions.append(region) | |
| idx += 1 | |
| print "Mergeable regions" | |
| for s in merged_regions: | |
| print "New merged region: %s MB" % sum([r.size for r in s]) | |
| print "#merge_region %s" % ','.join(['"%s"'%r.id for r in s]) | |
| for r in s: | |
| print "%s %20s %7d" % (r.start_key.ljust(40), r.id, r.size) | |
| print "Results:" | |
| no_total_regions = len(sorted_regions) | |
| no_mergeable_regions = sum([len(s) for s in merged_regions]) | |
| no_merged_regions = len(merged_regions) | |
| print "Total regions: %s" % no_total_regions | |
| print "Mergeable regions: %s" % no_mergeable_regions | |
| print "Merged regions: %s" % no_merged_regions | |
| print "Resulted regions: %s" % (no_total_regions - no_mergeable_regions + no_merged_regions) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment