percolatorTrackSpec = track.Track(
name="percolator",
description=
"This test indexes 2M AOL queries and use the percolator query to match",
source_root_url="http://benchmarks.elastic.co/corpora/percolator",
indices=[
track.Index(
name="queries",
types=[
# The type for the percolator queries:
track.Type(name=".percolator",
mapping_file_name="queries-mapping.json",
document_file_name="queries.json.bz2",
number_of_documents=2000000,
compressed_size_in_bytes=123502,
uncompressed_size_in_bytes=148039748),
# The used for documents being percolated:
track.Type(name="content",
mapping_file_name="document-mapping.json")
])
],
estimated_benchmark_time_in_minutes=5,
# Queries to use in the search benchmark
queries=[
PercolatorQuery(content="president bush"),
PercolatorQuery(content="saddam hussein"),
PercolatorQuery(content="hurricane katrina"),
PercolatorQuery(content="google"),
PercolatorQueryNoScoring(content="google"),
PercolatorQueryWithHighlighting(),
PercolatorQuery(content="ignore me"),
PercolatorQueryNoScoring(content="ignore me")
],
track_setups=[
track.TrackSetup(
name="4gheap",
description=
"same as Defaults except using a 4 GB heap (ES_HEAP_SIZE), because the ES default (-Xmx1g) sometimes hits OOMEs.",
candidate=track.CandidateSettings(
index_settings=percolatorIndexSettings, heap="4g"),
benchmark=track.BenchmarkSettings(benchmark_search=True))
])
def test_default_challenge_none_if_no_challenges(self):
self.assertIsNone(
track.Track(name="unittest",
description="unittest track",
challenges=[]).default_challenge)
if self.scroll_id:
es.clear_scroll(scroll_id=self.scroll_id)
self.scroll_id = None
geonamesTrackSpec = track.Track(
name="geonames",
description=
"This test indexes 8.6M documents (POIs from Geonames, total 2.8 GB json) using 8 client threads and 5000 docs per bulk "
"request against Elasticsearch",
source_root_url="http://benchmarks.elastic.co/corpora/geonames",
index_name=GEO_NAMES_INDEX_NAME,
type_name=GEO_NAMES_TYPE_NAME,
number_of_documents=8647880,
compressed_size_in_bytes=197857614,
uncompressed_size_in_bytes=2790927196,
document_file_name="documents.json.bz2",
mapping_file_name="mappings.json",
# for defaults alone, it's just around 20 minutes, for all it's about 60
estimated_benchmark_time_in_minutes=20,
# Queries to use in the search benchmark
queries=[
DefaultQuery(),
TermQuery(),
PhraseQuery(),
CountryAggQuery(use_request_cache=False),
CountryAggQuery(suffix="_cached", use_request_cache=True),
ScrollQuery()
],
track_setups=track.track_setups)