tsworkload_smw1t

# Copyright (c) 2020 YCSB contributors. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License"); you
# may not use this file except in compliance with the License. You
# may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
# implied. See the License for the specific language governing
# permissions and limitations under the License. See accompanying
# LICENSE file.

# Yahoo! Cloud System Benchmark
#
# Workload Defintion: Smart Metering Workload 1 (SMW1T)
#
# This workload was created as part of a bachelor thesis
# at the Technical University of Berlin, by Michael Smart
# https://github.com/smartygus
#
# Data to load as basis:
#
# There are 50000 time series, each with one year of
# data at half-hourly (1800 seconds) interval.
# We are using a fixed starting time stamp representing
# 1/1/2019 at midnight.
# This amounts to a total of:
# 50000 (time series) * 2 (point per hour) * 24 (hours in a day) * 365 (days in 2019) = 876 Million records
#
# Each time series has three tags:
#
# - se_segment: representing some socio-economic category defined by the provider
#   - cardinality = 5 (ie. there are 5 different categories)
# - pricing_strategy: some customers have different pricing strategies
#   - cardinality = 2 (ie. there are 2 different strategies, eg. fixed, and dynamic)
# - household_id = an id representing the household. The ID is unique within the context of the other two tags
#   - cardinality = 5000
#
# Run phase of the workload:
#
# There are 12 SCAN queries that make up the run phase of 
# the workload, one for each month of the year.
# The idea behind this workload is from the perspective of
# an energy provider that wants to know what the average
# monthly usage is grouped by socio-ecomonic segment.
#
# The SCAN queries in this workload pull *a lot* of
# data from the database at once (eg. ~74.5 Million
# points for the Jan 2019 query), but that's the
# point of this workload, to test how the database
# handles pulling large ranges of data out of the
# database.
#
# The suggested way to run this workload is with
# just one thread, so each of the SCAN queries
# are just carried out one behind the other.
# Being an "offline" style workload where data
# is just pre-loaded for the purposes of analysis,
# it's unlikely that queries such as those in this
# workload would be run more than once, so just
# running them all once in a row is a reasonable
# representation of such an analysis workflow.
workload=site.ycsb.workloads.TimeSeriesWorkload

recordcount=876000000
operationcount=24
insertcount=876000000
insertstart=1546300800
# This workload is "offline", so during the run phase
# there are no INSERTs, just SCANs.
# Therefore we do not need to specify
# an inserttransactionstart timestamp
#
# We only have a single metric, so we are
# dividing up the write load into groups of
# individual time series instead of groups
# of metrics
threadedwritedistribution=timeseries

fieldlength=3
# single metric: 'kwh'
fieldcount=1
# three tags, "real world" names: se_segment, pricing_strategy, household_id
tagcount=3
tagcardinality=5,2,5000

sparsity=0.0
delayedseries=0.0
delayedintervals=0

timestampunits=SECONDS
# one measurement every half hour
timestampinterval=1800
randomtimeseriesorder=false

# Query 1: January 2019
scan1proportion=0.08333
scan1tags=-1,-1,-1
scan1startts=1546300800
scan1querytimespan=2678400
scan1queryrandomtimespan=false
scan1groupbyfunction=AVERAGE
scan1groupbytagkeys=1,0,0
scan1downsamplingfunction=SUM
scan1downsamplinginterval=0

# Query 2: February 2019
scan2proportion=0.08333
scan2tags=-1,-1,-1
scan2startts=1548979200
scan2querytimespan=2419200
scan2queryrandomtimespan=false
scan2groupbyfunction=AVERAGE
scan2groupbytagkeys=1,0,0
scan2downsamplingfunction=SUM
scan2downsamplinginterval=0

# Query 3: March 2019
scan3proportion=0.08333
scan3tags=-1,-1,-1
scan3startts=1551398400
scan3querytimespan=2678400
scan3queryrandomtimespan=false
scan3groupbyfunction=AVERAGE
scan3groupbytagkeys=1,0,0
scan3downsamplingfunction=SUM
scan3downsamplinginterval=0

# Query 4: April 2019
scan4proportion=0.08333
scan4tags=-1,-1,-1
scan4startts=1554076800
scan4querytimespan=2592000
scan4queryrandomtimespan=false
scan4groupbyfunction=AVERAGE
scan4groupbytagkeys=1,0,0
scan4downsamplingfunction=SUM
scan4downsamplinginterval=0

# Query 5: May 2019
scan5proportion=0.08333
scan5tags=-1,-1,-1
scan5startts=1556668800
scan5querytimespan=2678400
scan5queryrandomtimespan=false
scan5groupbyfunction=AVERAGE
scan5groupbytagkeys=1,0,0
scan5downsamplingfunction=SUM
scan5downsamplinginterval=0

# Query 6: June 2019
scan6proportion=0.08333
scan6tags=-1,-1,-1
scan6startts=1559347200
scan6querytimespan=2592000
scan6queryrandomtimespan=false
scan6groupbyfunction=AVERAGE
scan6groupbytagkeys=1,0,0
scan6downsamplingfunction=SUM
scan6downsamplinginterval=0

# Query 7: July 2019
scan7proportion=0.08333
scan7tags=-1,-1,-1
scan7startts=1561939200
scan7querytimespan=2678400
scan7queryrandomtimespan=false
scan7groupbyfunction=AVERAGE
scan7groupbytagkeys=1,0,0
scan7downsamplingfunction=SUM
scan7downsamplinginterval=0

# Query 8: August 2019
scan8proportion=0.08333
scan8tags=-1,-1,-1
scan8startts=1564617600
scan8querytimespan=2678400
scan8queryrandomtimespan=false
scan8groupbyfunction=AVERAGE
scan8groupbytagkeys=1,0,0
scan8downsamplingfunction=SUM
scan8downsamplinginterval=0

# Query 9: September 2019
scan9proportion=0.08333
scan9tags=-1,-1,-1
scan9startts=1567296000
scan9querytimespan=2592000
scan9queryrandomtimespan=false
scan9groupbyfunction=AVERAGE
scan9groupbytagkeys=1,0,0
scan9downsamplingfunction=SUM
scan9downsamplinginterval=0

# Query 10: October 2019
scan10proportion=0.08333
scan10tags=-1,-1,-1
scan10startts=1569888000
scan10querytimespan=2678400
scan10queryrandomtimespan=false
scan10groupbyfunction=AVERAGE
scan10groupbytagkeys=1,0,0
scan10downsamplingfunction=SUM
scan10downsamplinginterval=0

# Query 11: November 2019
scan11proportion=0.08333
scan11tags=-1,-1,-1
scan11startts=1572566400
scan11querytimespan=2592000
scan11queryrandomtimespan=false
scan11groupbyfunction=AVERAGE
scan11groupbytagkeys=1,0,0
scan11downsamplingfunction=SUM
scan11downsamplinginterval=0

# Query 12: December 2019
scan12proportion=0.08333
scan12tags=-1,-1,-1
scan12startts=1575158400
scan12querytimespan=2678400
scan12queryrandomtimespan=false
scan12groupbyfunction=AVERAGE
scan12groupbytagkeys=1,0,0
scan12downsamplingfunction=SUM
scan12downsamplinginterval=0


readproportion=0.00
insertproportion=0.00
updateproportion=0.00