Validated test results and cleaned-up bench code.

Author: tinybit

tembo-olap/README.md

@@ -1,11 +1,7 @@
 # Instantiation
 
-1- Create Account and Login to https://cloud.tembo.io
-
-2- Select OLAP Stack after clicking Create Instance button.
-
-![Instance for benchmark](olap-stack.png "OLAP Stack")
-
+1) Create Account and Login to https://cloud.tembo.io
+2) Select OLAP Stack after clicking Create Instance button
 
 ## Choose System Configuration:
 
@@ -15,62 +11,21 @@
 
 **RAM** : 32GB
 
-**Storage**: 300GB
-
+**Storage**: 500GB 
 
-![Instance Type](olap-instance-type.png "OLAP Stack")
+Note: instance creation takes *20-25* seconds.
 
-Instance creation takes *20-25* seconds.
-
-## Connection String
+## Specify Connection params
 Tembo provides connection-string through instance **Home** tab.
 
+Use hostname and password taken from Tembo console and specify them in benchmark.sh
 ```
-export PGPASSWORD=TEMBO_GENERATED_PGPASSWORD
-export PGHOST=TEMBO_GENERATED_PGHOST
-```
-
-## Data loading
-Data loaded remotely using psql client.
-
-**psql** installation
-
-```
-sudo apt-get update
-sudo apt-get install -y postgresql-client
-
+HOSTNAME="<hostname>"
+PASSWORD="<password>"
 ```
-**Data** download
 
-```
-wget --no-verbose --continue 'https://datasets.clickhouse.com/hits_compatible/hits.tsv.gz'
+## Start Benchmark
 
-gzip -d hits.tsv.gz
 ```
-
-**Data** loading
-
-```
-psql postgresql://postgres:$PGPASSWORD@$PGHOST:5432 -t -c 'CREATE DATABASE test'
-
-psql postgresql://postgres:$PGPASSWORD@$PGHOST:5432/test -t < create.sql
-
-psql postgres://postgres:$PGPASSWORD@$PGHOST:5432/test -t -c '\timing' -c "\\copy hits FROM 'hits.tsv'"
-
-psql postgresql://postgres:$PGPASSWORD@$PGHOST:5432/test -t < index.sql
-
-psql postgres://postgres:$PGPASSWORD@$PGHOST:5432/test -t -c "select pg_total_relation_size('hits');"
-            33864704000
-
-
-```
-
-## Benchmark Start
-
-```
-./run.sh 2>&1 | tee log.txt
-
-cat log.txt | grep -oP 'Time: \d+\.\d+ ms' | sed -r -e 's/Time: ([0-9]+\.[0-9]+) ms/\1/' |
-    awk '{ if (i % 3 == 0) { printf "[" }; printf $1 / 1000; if (i % 3 != 2) { printf "," } else { print "]," }; ++i; }'
-
+./benchmark.sh
 ```

tembo-olap/benchmark.sh

@@ -0,0 +1,35 @@
+#!/bin/bash
+
+HOSTNAME="<hostname>"
+PASSWORD="<password>"
+
+sudo apt-get update
+sudo apt-get install -y postgresql-client
+
+wget --no-verbose --continue 'https://datasets.clickhouse.com/hits_compatible/hits.tsv.gz'
+gzip -d hits.tsv.gz
+chmod 777 ~ hits.tsv
+
+psql postgresql://postgres:$PASSWORD@$HOSTNAME:5432 -t -c 'CREATE DATABASE test'
+psql "host=$HOSTNAME port=5432 dbname=test user=postgres password=$PASSWORD sslmode=require" < create.sql
+psql "host=$HOSTNAME port=5432 dbname=test user=postgres password=$PASSWORD sslmode=require" -t -c '\timing' -c "\\copy hits FROM 'hits.tsv'"
+psql "host=$HOSTNAME port=5432 dbname=test user=postgres password=$PASSWORD sslmode=require" < index.sql
+psql "host=$HOSTNAME port=5432 dbname=test user=postgres password=$PASSWORD sslmode=require" -t -c '\timing' -c "select pg_total_relation_size('hits');"
+
+./run.sh "${HOSTNAME}" "${PASSWORD}" 2>&1 | tee log.txt
+
+cat log.txt |
+grep -oP 'Time: \d+\.\d+ ms' |
+sed -r -e 's/Time: ([0-9]+\.[0-9]+) ms/\1/' |
+awk '{
+    if (i % 3 == 0) {
+        printf "[";
+    }
+    printf $1 / 1000;
+    if (i % 3 != 2) {
+        printf ",";
+    } else {
+        print "],";
+    }
+    ++i;
+}'

tembo-olap/create.sql

@@ -1,5 +1,5 @@
 CREATE EXTENSION IF NOT EXISTS COLUMNAR;
-DROP TABLE hits CASCADE;
+DROP TABLE IF EXISTS hits CASCADE;
 CREATE TABLE hits
 (
     WatchID BIGINT NOT NULL,

tembo-olap/olap-instance-type.png

@@ -40,4 +40,4 @@ SELECT URL, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >
 SELECT TraficSourceID, SearchEngineID, AdvEngineID, CASE WHEN (SearchEngineID = 0 AND AdvEngineID = 0) THEN Referer ELSE '' END AS Src, URL AS Dst, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >= '2013-07-01' AND EventDate <= '2013-07-31' AND IsRefresh = 0 GROUP BY TraficSourceID, SearchEngineID, AdvEngineID, Src, Dst ORDER BY PageViews DESC LIMIT 10 OFFSET 1000;
 SELECT URLHash, EventDate, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >= '2013-07-01' AND EventDate <= '2013-07-31' AND IsRefresh = 0 AND TraficSourceID IN (-1, 6) AND RefererHash = 3594120000172545465 GROUP BY URLHash, EventDate ORDER BY PageViews DESC LIMIT 10 OFFSET 100;
 SELECT WindowClientWidth, WindowClientHeight, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >= '2013-07-01' AND EventDate <= '2013-07-31' AND IsRefresh = 0 AND DontCountHits = 0 AND URLHash = 2868770270353813622 GROUP BY WindowClientWidth, WindowClientHeight ORDER BY PageViews DESC LIMIT 10 OFFSET 10000;
-SELECT DATE_TRUNC('minute', EventTime) AS M, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >= '2013-07-14' AND EventDate <= '2013-07-15' AND IsRefresh = 0 AND DontCountHits = 0 GROUP BY DATE_TRUNC('minute', EventTime) ORDER BY DATE_TRUNC('minute', EventTime) LIMIT 10 OFFSET 1000;
+SELECT DATE_TRUNC('minute', EventTime) AS M, COUNT(*) AS PageViews FROM hits WHERE CounterID = 62 AND EventDate >= '2013-07-14' AND EventDate <= '2013-07-15' AND IsRefresh = 0 AND DontCountHits = 0 GROUP BY DATE_TRUNC('minute', EventTime) ORDER BY DATE_TRUNC('minute', EventTime) LIMIT 10 OFFSET 1000;

tembo-olap/olap-stack.png

@@ -1,58 +1,58 @@
 {
-    "system": "Tembo OLAP(Columnar)",
+    "system": "Tembo OLAP (columnar)",
     "date": "2024-02-09",
-    "machine": "c6a.4xlarge, 300gb gp3",
+    "machine": "c6a.4xlarge, 500gb gp3",
     "cluster_size": 1,
     "comment": "",
 
     "tags": ["C", "PostgreSQL compatible", "column-oriented"],
 
-    "load_time": 14998.247,
+    "load_time": 4903,
     "data_size": 33864704000,
 
     "result": [
-[0.322266,0.279012,0.278818],
-[0.262802,0.263067,0.263524],
-[1.03503,1.02851,1.01109],
-[0.694882,0.69194,0.684567],
-[24.1366,23.8622,24.2121],
-[37.8125,37.3546,37.4614],
-[0.088267,0.071204,0.072592],
-[0.27134,0.272216,0.270787],
-[25.1755,24.8466,24.944],
-[28.1199,28.1654,28.4494],
-[2.20623,2.17468,2.14656],
-[2.4656,2.4581,2.45926],
-[7.42677,7.57167,7.69995],
-[10.5288,10.3381,10.4471],
-[8.29726,8.19628,8.19725],
-[16.6094,16.9222,16.7514],
-[38.8504,38.4172,38.3965],
-[0.388963,0.358219,0.36013],
-[75.809,75.2281,76.712],
-[0.302681,0.303397,0.302277],
-[2.62566,2.61241,2.61167],
-[2.8029,2.81523,2.84338],
-[3.70515,3.68054,3.66689],
-[21.0447,20.8168,21.0621],
-[1.44812,1.34319,1.33672],
-[0.34479,0.318921,0.317873],
-[1.21699,1.20804,1.20271],
-[4.83795,4.79006,4.79372],
-[41.5645,41.008,40.9968],
-[7.96046,8.03584,7.98607],
-[8.77845,9.01301,8.9083],
-[14.8007,14.662,14.7183],
-[69.205,69.6116,70.0597],
-[32.5098,31.3654,32.3301],
-[33.5434,33.3958,33.0693],
-[14.3637,14.5175,14.395],
-[0.373303,0.373503,0.376101],
-[0.193498,0.190686,0.194189],
-[0.103831,0.111741,0.119202],
-[0.754824,0.760665,0.798245],
-[0.110722,0.110677,0.114796],
-[0.101987,0.10852,0.101449],
-[0.148887,0.12897,0.146846]
+[0.335424,0.288406,0.291833],
+[0.273387,0.276144,0.280566],
+[1.04303,1.03311,1.02745],
+[0.699428,0.698998,0.701829],
+[23.4004,23.0228,23.3303],
+[35.6514,35.9655,36.1087],
+[0.091153,0.074667,0.076713],
+[0.27855,0.276916,0.278651],
+[27.9524,27.7416,27.1891],
+[30.495,31.5809,30.8035],
+[2.98169,3.01209,2.96031],
+[3.36352,3.38749,3.32161],
+[9.25541,9.33663,9.30109],
+[10.6757,10.6452,10.352],
+[10.0414,9.99654,9.95628],
+[23.6013,23.7551,24.0158],
+[27.5649,26.7495,27.0685],
+[0.422969,0.37809,0.382737],
+[46.2609,46.2509,46.5879],
+[0.312248,0.312461,0.319174],
+[2.71278,2.74919,2.66242],
+[2.85679,2.8941,2.91063],
+[3.79428,3.82795,3.76534],
+[21.6777,21.8457,22.4842],
+[1.55266,1.4268,1.42082],
+[0.3739,0.336904,0.343475],
+[1.23959,1.22557,1.23188],
+[4.89249,4.94013,4.92233],
+[42.7291,42.4905,42.5116],
+[8.16955,8.15319,8.1022],
+[11.0438,10.8213,10.9278],
+[18.9006,18.1182,17.7459],
+[72.6787,71.4905,72.13],
+[44.2683,42.1593,42.9159],
+[44.1782,44.6741,44.1548],
+[17.2926,18.1069,17.8896],
+[0.393598,0.395635,0.394274],
+[0.234447,0.237086,0.196967],
+[0.108035,0.125385,0.117379],
+[0.78023,0.826582,0.806038],
+[0.106995,0.113052,0.118778],
+[0.120202,0.11803,0.119093],
+[0.153295,0.150266,0.151742]
 ]
 }

tembo-olap/queries.sql

@@ -1,10 +1,13 @@
 #!/bin/bash
 
+HOSTNAME=$1
+PASSWORD=$2
+
 TRIES=3
 
 cat queries.sql | while read query; do
     echo "$query";
     for i in $(seq 1 $TRIES); do
-	    psql postgres://postgres:$PGPASSWORD@$PGHOST:5432/test -t -c '\timing' -c "$query" | grep 'Time'
+	    psql "host=$HOSTNAME port=5432 dbname=test user=postgres password=$PASSWORD sslmode=require" -t -c '\timing' -c "$query" | grep 'Time'
     done;
 done;