1.1 piliona malaga taxi: 108-core ClickHouse cluster

O le faaliliuga o le tusiga na saunia faapitoa mo tamaiti aoga o le vasega Inisinia Fa'amaumauga.

1.1 piliona malaga taxi: 108-core ClickHouse cluster

Kiliki Fale ose fa'amaumauga tu'ufa'atasi puna matala. Ose si'osi'omaga lelei lea e mafai ai e le faitau selau o tagata su'esu'e ona vave fesiligia fa'amatalaga au'ili'ili, tusa lava pe fia sefulu piliona o fa'amaumauga fou e fa'aulu i le aso. O tau o galuega tetele mo le lagolagoina o sea faiga e mafai ona maualuga i le $100 i le tausaga, ma atonu o le afa e faalagolago i le faʻaaogaina. I se tasi taimi, o le ClickHouse faʻapipiʻi mai Yandex Metrics o loʻo i ai 10 trillion faamaumauga. I le faaopoopo atu ia Yandex, ua maua foi e ClickHouse le manuia ma Bloomberg ma Cloudflare.

Lua tausaga talu ai sa ou faaaluina iloiloga fa'atusatusa faʻamaumauga e faʻaaoga ai le masini e tasi, ma na avea le televave free database software ua ou vaai i ai. Talu mai lena taimi, e leʻi taofia e le au atinaʻe le faʻaopoopoina o foliga, e aofia ai le lagolago mo Kafka, HDFS ma le ZStandard compression. I le tausaga talu ai na latou faʻaopoopoina le lagolago mo auala faʻapipiʻi faʻamalosi, ma delta-mai-delta coding ua mafai. Pe a faʻapipiʻi faʻamaumauga faʻasologa o taimi, e mafai ona faʻapipiʻi lelei fua faʻatatau i le faʻaogaina o le delta encoding, ae mo faʻatau e sili atu le faʻaogaina o le delta-by-delta encoding. O le faʻamalosi lelei ua avea ma ki i le faʻatinoga a ClickHouse.

ClickHouse e aofia ai le 170 afe laina o le C++ code, e le aofia ai faletusi lona tolu, ma o se tasi lea o faʻamaumauga faʻamaumauga tuʻufaʻatasia laʻititi. I le faʻatusatusaina, e le lagolagoina e le SQLite le tufatufaina atu ma e aofia ai le 235 afe laina o le code C. E pei o lenei tusitusiga, 207 inisinia ua saofagā i le ClickHouse, ma o le malosi o tautinoga ua faʻateleina talu ai nei.

Ia Mati 2017, na amata ai ona faʻatautaia le ClickHouse sui ogalaau o se auala faigofie e siaki ai le atinaʻe. Na latou talepeina foʻi le faila faʻamaumauga monolithic i se faʻasologa o faila faila a Markdown. O fa'afitauli ma foliga o lo'o su'esu'eina e ala i le GitHub, ma i se tulaga lautele ua sili atu ona faigofie le polokalama i nai tausaga talu ai.

I lenei tusiga, o le a ou vaʻavaʻai i le faʻatinoga o le ClickHouse cluster i luga o le AWS EC2 faʻaaogaina 36-core processors ma NVMe storage.

FUAFUAGA: I le vaiaso talu ona faʻasalalau muamua lenei pou, na ou toe faʻataʻitaʻiina le suʻega ma se faʻaleleia atili o faʻatulagaga ma maua ai ni faʻaiʻuga sili atu. Ua toe faafou lenei pou ina ia atagia ai nei suiga.

Tatalaina le AWS EC2 Cluster

O le a ou faʻaaogaina tolu c5d.9xlarge EC2 taimi mo lenei pou. O i latou taʻitasi o loʻo i ai 36 virtual CPUs, 72 GB o le RAM, 900 GB o le NVMe SSD teuina ma lagolagoina le 10 Gigabit network. Latou te tau $1,962/itula taʻitasi i le eu-sisifo-1 itulagi pe a taʻavale pe a manaʻomia. O le a ou faʻaaogaina le Ubuntu Server 16.04 LTS e fai ma faiga faʻaoga.

O le firewall e faʻapipiʻiina ina ia mafai e masini taʻitasi ona fesoʻotaʻi ma le tasi ma le isi e aunoa ma ni faʻatapulaʻaina, ma naʻo laʻu tuatusi IPv4 o loʻo paʻepaʻe e le SSH i le fuifui.

NVMe ta'avale i le tulaga sauni galue

Mo le ClickHouse e galue, o le a ou fatuina se faila faila i le EXT4 format i luga o le NVMe drive i luga o sapalai taʻitasi.

$ sudo mkfs -t ext4 /dev/nvme1n1
$ sudo mkdir /ch
$ sudo mount /dev/nvme1n1 /ch

A maeʻa ona faʻatulagaina mea uma, e mafai ona e vaʻai i le mauga ma le 783 GB avanoa avanoa i luga o faiga taʻitasi.

$ lsblk

NAME        MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT
loop0         7:0    0  87.9M  1 loop /snap/core/5742
loop1         7:1    0  16.5M  1 loop /snap/amazon-ssm-agent/784
nvme0n1     259:1    0     8G  0 disk
└─nvme0n1p1 259:2    0     8G  0 part /
nvme1n1     259:0    0 838.2G  0 disk /ch

$ df -h

Filesystem      Size  Used Avail Use% Mounted on
udev             35G     0   35G   0% /dev
tmpfs           6.9G  8.8M  6.9G   1% /run
/dev/nvme0n1p1  7.7G  967M  6.8G  13% /
tmpfs            35G     0   35G   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs            35G     0   35G   0% /sys/fs/cgroup
/dev/loop0       88M   88M     0 100% /snap/core/5742
/dev/loop1       17M   17M     0 100% /snap/amazon-ssm-agent/784
tmpfs           6.9G     0  6.9G   0% /run/user/1000
/dev/nvme1n1    825G   73M  783G   1% /ch

O le fa'amaumauga o le a ou fa'aogaina i lenei su'ega ose fa'amaumauga tu'ufa'atasi na ou gaosia mai le 1.1 piliona ta'avale taxi na ave i le Aai o Niu Ioka mo le ono tausaga. I luga o le blog Tasi Piliona malaga a le taxi i Redshift fa'amatalaga pe na faapefea ona ou aoina lenei seti fa'amaumauga. O loʻo teuina i le AWS S3, o lea o le a ou faʻapipiʻiina le AWS CLI ma loʻu avanoa ma ki faalilolilo.

$ sudo apt update
$ sudo apt install awscli
$ aws configure

O le a ou setiina le fa'atapula'a o talosaga fa'atasi a le kalani i le 100 ina ia vave ona la'u faila nai lo tulaga fa'aletonu.

$ aws configure set 
    default.s3.max_concurrent_requests 
    100

O le a ou sii maia le faamaumauga o taʻavale taxi mai le AWS S3 ma teu i luga o le NVMe drive i luga o le server muamua. Ole fa'amaumauga lenei e ~104GB ile GZIP-compressed CSV format.

$ sudo mkdir -p /ch/csv
$ sudo chown -R ubuntu /ch/csv
$ aws s3 sync s3://<bucket>/csv /ch/csv

Fa'apipi'i ClickHouse

O le a ou faʻapipiʻiina le OpenJDK tufatufaga mo Java 8 e pei ona manaʻomia e faʻatautaia ai Apache ZooKeeper, lea e manaʻomia mo le faʻapipiʻiina o le ClickHouse i luga o masini uma e tolu.

$ sudo apt update
$ sudo apt install 
    openjdk-8-jre 
    openjdk-8-jdk-headless

Ona ou setiina lea o le fesuiaiga o le siosiomaga JAVA_HOME.

$ sudo vi /etc/profile
 
export JAVA_HOME=/usr
 
$ source /etc/profile

Ona ou faʻaogaina lea o le polokalama faʻapipiʻi a le Ubuntu e faʻapipiʻi ai ClickHouse 18.16.1, tilotilo ma ZooKeeper i luga o masini uma e tolu.

$ sudo apt-key adv 
    --keyserver hkp://keyserver.ubuntu.com:80 
    --recv E0C56BD4
$ echo "deb http://repo.yandex.ru/clickhouse/deb/stable/ main/" | 
    sudo tee /etc/apt/sources.list.d/clickhouse.list
$ sudo apt-get update

$ sudo apt install 
    clickhouse-client 
    clickhouse-server 
    glances 
    zookeeperd

O le a ou faia se lisi mo ClickHouse ma faia foi nisi faʻatonuga faʻasologa i luga o sapalai uma e tolu.

$ sudo mkdir /ch/clickhouse
$ sudo chown -R clickhouse /ch/clickhouse

$ sudo mkdir -p /etc/clickhouse-server/conf.d
$ sudo vi /etc/clickhouse-server/conf.d/taxis.conf

O fa'atonuga ia o le a ou fa'aaogaina.

<?xml version="1.0"?>
<yandex>
    <listen_host>0.0.0.0</listen_host>
    <path>/ch/clickhouse/</path>

 <remote_servers>
        <perftest_3shards>
            <shard>
                <replica>
                    <host>172.30.2.192</host>
                    <port>9000</port>
                 </replica>
            </shard>
            <shard>
                 <replica>
                    <host>172.30.2.162</host>
                    <port>9000</port>
                 </replica>
            </shard>
            <shard>
                 <replica>
                    <host>172.30.2.36</host>
                    <port>9000</port>
                 </replica>
            </shard>
        </perftest_3shards>
    </remote_servers>

  <zookeeper-servers>
        <node>
            <host>172.30.2.192</host>
            <port>2181</port>
        </node>
        <node>
            <host>172.30.2.162</host>
            <port>2181</port>
        </node>
        <node>
            <host>172.30.2.36</host>
            <port>2181</port>
        </node>
    </zookeeper-servers>

 <macros>
        <shard>03</shard>
        <replica>01</replica>
    </macros>
</yandex>

Ona ou faʻatautaia lea o le ZooKeeper ma le ClickHouse server i luga o masini uma e tolu.

$ sudo /etc/init.d/zookeeper start
$ sudo service clickhouse-server start

Tuuina atu o faʻamaumauga ile ClickHouse

I luga o le server muamua o le a ou fatuina se laulau malaga (trips), lea o le a teuina ai se faʻamaumauga o malaga taʻavale e faʻaaoga ai le masini Log.

$ clickhouse-client --host=0.0.0.0
 
CREATE TABLE trips (
    trip_id                 UInt32,
    vendor_id               String,

    pickup_datetime         DateTime,
    dropoff_datetime        Nullable(DateTime),

    store_and_fwd_flag      Nullable(FixedString(1)),
    rate_code_id            Nullable(UInt8),
    pickup_longitude        Nullable(Float64),
    pickup_latitude         Nullable(Float64),
    dropoff_longitude       Nullable(Float64),
    dropoff_latitude        Nullable(Float64),
    passenger_count         Nullable(UInt8),
    trip_distance           Nullable(Float64),
    fare_amount             Nullable(Float32),
    extra                   Nullable(Float32),
    mta_tax                 Nullable(Float32),
    tip_amount              Nullable(Float32),
    tolls_amount            Nullable(Float32),
    ehail_fee               Nullable(Float32),
    improvement_surcharge   Nullable(Float32),
    total_amount            Nullable(Float32),
    payment_type            Nullable(String),
    trip_type               Nullable(UInt8),
    pickup                  Nullable(String),
    dropoff                 Nullable(String),

    cab_type                Nullable(String),

    precipitation           Nullable(Int8),
    snow_depth              Nullable(Int8),
    snowfall                Nullable(Int8),
    max_temperature         Nullable(Int8),
    min_temperature         Nullable(Int8),
    average_wind_speed      Nullable(Int8),

    pickup_nyct2010_gid     Nullable(Int8),
    pickup_ctlabel          Nullable(String),
    pickup_borocode         Nullable(Int8),
    pickup_boroname         Nullable(String),
    pickup_ct2010           Nullable(String),
    pickup_boroct2010       Nullable(String),
    pickup_cdeligibil       Nullable(FixedString(1)),
    pickup_ntacode          Nullable(String),
    pickup_ntaname          Nullable(String),
    pickup_puma             Nullable(String),

    dropoff_nyct2010_gid    Nullable(UInt8),
    dropoff_ctlabel         Nullable(String),
    dropoff_borocode        Nullable(UInt8),
    dropoff_boroname        Nullable(String),
    dropoff_ct2010          Nullable(String),
    dropoff_boroct2010      Nullable(String),
    dropoff_cdeligibil      Nullable(String),
    dropoff_ntacode         Nullable(String),
    dropoff_ntaname         Nullable(String),
    dropoff_puma            Nullable(String)
) ENGINE = Log;

Ona ou aveese lea ma uta ia faila CSV taitasi i totonu o se laulau malaga (trips). O mea nei na mae'a i le 55 minute ma le 10 sekone. A maeʻa lenei taʻaloga, o le tele o le faʻamaumauga faʻamaumauga o le 134 GB.

$ time (for FILENAME in /ch/csv/trips_x*.csv.gz; do
            echo $FILENAME
            gunzip -c $FILENAME | 
                clickhouse-client 
                    --host=0.0.0.0 
                    --query="INSERT INTO trips FORMAT CSV"
        done)

O le saosaoa fa'aulufale mai e 155 MB o mea e le'i fa'apipi'iina CSV i le sekone. Ou te masalo o lenei mea na mafua ona o se fagu i le GZIP decompression. Atonu e sili atu le vave e tatala uma faila gzipped faʻatasi i le faʻaogaina o xargs ona uta lea o faʻamaumauga e le faʻapipiʻi. O lo'o i lalo se fa'amatalaga o mea na lipotia i le taimi o le fa'aulufaleina mai o le CSV.

$ sudo glances

ip-172-30-2-200 (Ubuntu 16.04 64bit / Linux 4.4.0-1072-aws)                                                                                                 Uptime: 0:11:42
CPU       8.2%  nice:     0.0%                           LOAD    36-core                           MEM      9.8%  active:    5.20G                           SWAP      0.0%
user:     6.0%  irq:      0.0%                           1 min:    2.24                            total:  68.7G  inactive:  61.0G                           total:       0
system:   0.9%  iowait:   1.3%                           5 min:    1.83                            used:   6.71G  buffers:   66.4M                           used:        0
idle:    91.8%  steal:    0.0%                           15 min:   1.01                            free:   62.0G  cached:    61.6G                           free:        0

NETWORK     Rx/s   Tx/s   TASKS 370 (507 thr), 2 run, 368 slp, 0 oth sorted automatically by cpu_percent, flat view
ens5        136b    2Kb
lo         343Mb  343Mb     CPU%  MEM%  VIRT   RES   PID USER        NI S    TIME+ IOR/s IOW/s Command
                           100.4   1.5 1.65G 1.06G  9909 ubuntu       0 S  1:01.33     0     0 clickhouse-client --host=0.0.0.0 --query=INSERT INTO trips FORMAT CSV
DISK I/O     R/s    W/s     85.1   0.0 4.65M  708K  9908 ubuntu       0 R  0:50.60   32M     0 gzip -d -c /ch/csv/trips_xac.csv.gz
loop0          0      0     54.9   5.1 8.14G 3.49G  8091 clickhous    0 S  1:44.23     0   45M /usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml
loop1          0      0      4.5   0.0     0     0   319 root         0 S  0:07.50    1K     0 kworker/u72:2
nvme0n1        0     3K      2.3   0.0 91.1M 28.9M  9912 root         0 R  0:01.56     0     0 /usr/bin/python3 /usr/bin/glances
nvme0n1p1      0     3K      0.3   0.0     0     0   960 root       -20 S  0:00.10     0     0 kworker/28:1H
nvme1n1    32.1M   495M      0.3   0.0     0     0  1058 root       -20 S  0:00.90     0     0 kworker/23:1H

O le a ou faʻasaʻoloto avanoa i luga o le NVMe drive e ala i le tapeina o uluai faila CSV aʻo leʻi faʻaauau.

$ sudo rm -fr /ch/csv

Su'e ile Pepa Fa'ailoga

O le Log ClickHouse engine o le a teuina faʻamaumauga i se laina-faʻatulagaina faatulagaga. Ina ia vave su'esu'eina fa'amatalaga, ou te fa'aliliuina i le fa'asologa fa'akomupiuta e fa'aaoga ai le afi MergeTree.

$ clickhouse-client --host=0.0.0.0

O mea nei na mae'a i le 34 minute ma le 50 sekone. A maeʻa lenei taʻaloga, o le tele o le faʻamaumauga faʻamaumauga o le 237 GB.

CREATE TABLE trips_mergetree
    ENGINE = MergeTree(pickup_date, pickup_datetime, 8192)
    AS SELECT
        trip_id,
        CAST(vendor_id AS Enum8('1' = 1,
                                '2' = 2,
                                'CMT' = 3,
                                'VTS' = 4,
                                'DDS' = 5,
                                'B02512' = 10,
                                'B02598' = 11,
                                'B02617' = 12,
                                'B02682' = 13,
                                'B02764' = 14)) AS vendor_id,
        toDate(pickup_datetime)                 AS pickup_date,
        ifNull(pickup_datetime, toDateTime(0))  AS pickup_datetime,
        toDate(dropoff_datetime)                AS dropoff_date,
        ifNull(dropoff_datetime, toDateTime(0)) AS dropoff_datetime,
        assumeNotNull(store_and_fwd_flag)       AS store_and_fwd_flag,
        assumeNotNull(rate_code_id)             AS rate_code_id,

        assumeNotNull(pickup_longitude)         AS pickup_longitude,
        assumeNotNull(pickup_latitude)          AS pickup_latitude,
        assumeNotNull(dropoff_longitude)        AS dropoff_longitude,
        assumeNotNull(dropoff_latitude)         AS dropoff_latitude,
        assumeNotNull(passenger_count)          AS passenger_count,
        assumeNotNull(trip_distance)            AS trip_distance,
        assumeNotNull(fare_amount)              AS fare_amount,
        assumeNotNull(extra)                    AS extra,
        assumeNotNull(mta_tax)                  AS mta_tax,
        assumeNotNull(tip_amount)               AS tip_amount,
        assumeNotNull(tolls_amount)             AS tolls_amount,
        assumeNotNull(ehail_fee)                AS ehail_fee,
        assumeNotNull(improvement_surcharge)    AS improvement_surcharge,
        assumeNotNull(total_amount)             AS total_amount,
        assumeNotNull(payment_type)             AS payment_type_,
        assumeNotNull(trip_type)                AS trip_type,

        pickup AS pickup,
        pickup AS dropoff,

        CAST(assumeNotNull(cab_type)
            AS Enum8('yellow' = 1, 'green' = 2))
                                AS cab_type,

        precipitation           AS precipitation,
        snow_depth              AS snow_depth,
        snowfall                AS snowfall,
        max_temperature         AS max_temperature,
        min_temperature         AS min_temperature,
        average_wind_speed      AS average_wind_speed,

        pickup_nyct2010_gid     AS pickup_nyct2010_gid,
        pickup_ctlabel          AS pickup_ctlabel,
        pickup_borocode         AS pickup_borocode,
        pickup_boroname         AS pickup_boroname,
        pickup_ct2010           AS pickup_ct2010,
        pickup_boroct2010       AS pickup_boroct2010,
        pickup_cdeligibil       AS pickup_cdeligibil,
        pickup_ntacode          AS pickup_ntacode,
        pickup_ntaname          AS pickup_ntaname,
        pickup_puma             AS pickup_puma,

        dropoff_nyct2010_gid    AS dropoff_nyct2010_gid,
        dropoff_ctlabel         AS dropoff_ctlabel,
        dropoff_borocode        AS dropoff_borocode,
        dropoff_boroname        AS dropoff_boroname,
        dropoff_ct2010          AS dropoff_ct2010,
        dropoff_boroct2010      AS dropoff_boroct2010,
        dropoff_cdeligibil      AS dropoff_cdeligibil,
        dropoff_ntacode         AS dropoff_ntacode,
        dropoff_ntaname         AS dropoff_ntaname,
        dropoff_puma            AS dropoff_puma
    FROM trips;

O le mea lea na foliga mai o le va'aiga na maua i le taimi o le taotoga:

ip-172-30-2-200 (Ubuntu 16.04 64bit / Linux 4.4.0-1072-aws)                                                                                                 Uptime: 1:06:09
CPU      10.3%  nice:     0.0%                           LOAD    36-core                           MEM     16.1%  active:    13.3G                           SWAP      0.0%
user:     7.9%  irq:      0.0%                           1 min:    1.87                            total:  68.7G  inactive:  52.8G                           total:       0
system:   1.6%  iowait:   0.8%                           5 min:    1.76                            used:   11.1G  buffers:   71.8M                           used:        0
idle:    89.7%  steal:    0.0%                           15 min:   1.95                            free:   57.6G  cached:    57.2G                           free:        0

NETWORK     Rx/s   Tx/s   TASKS 367 (523 thr), 1 run, 366 slp, 0 oth sorted automatically by cpu_percent, flat view
ens5         1Kb    8Kb
lo           2Kb    2Kb     CPU%  MEM%  VIRT   RES   PID USER        NI S    TIME+ IOR/s IOW/s Command
                           241.9  12.8 20.7G 8.78G  8091 clickhous    0 S 30:36.73   34M  125M /usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml
DISK I/O     R/s    W/s      2.6   0.0 90.4M 28.3M  9948 root         0 R  1:18.53     0     0 /usr/bin/python3 /usr/bin/glances
loop0          0      0      1.3   0.0     0     0   203 root         0 S  0:09.82     0     0 kswapd0
loop1          0      0      0.3   0.1  315M 61.3M 15701 ubuntu       0 S  0:00.40     0     0 clickhouse-client --host=0.0.0.0
nvme0n1        0     3K      0.3   0.0     0     0     7 root         0 S  0:00.83     0     0 rcu_sched
nvme0n1p1      0     3K      0.0   0.0     0     0   142 root         0 S  0:00.22     0     0 migration/27
nvme1n1    25.8M   330M      0.0   0.0 59.7M 1.79M  2764 ubuntu       0 S  0:00.00     0     0 (sd-pam)

I le suega mulimuli, e tele koluma na liua ma toe fa'atatau. Na ou iloa o nisi o nei galuega e le o toe galue e pei ona faʻamoemoeina i lenei faʻamaumauga. Ina ia foia lenei faafitauli, na ou aveesea galuega le talafeagai ma utaina faʻamaumauga e aunoa ma le liua i ni ituaiga sili atu.

Fa'asoaina o fa'amaumauga i luga o le fuifui

O le a ou tufatufaina atu fa'amaumauga i nodes uma e tolu. Ina ia amata, i lalo o le a ou fatuina se laulau i luga o masini uma e tolu.

$ clickhouse-client --host=0.0.0.0

CREATE TABLE trips_mergetree_third (
    trip_id                 UInt32,
    vendor_id               String,
    pickup_date             Date,
    pickup_datetime         DateTime,
    dropoff_date            Date,
    dropoff_datetime        Nullable(DateTime),
    store_and_fwd_flag      Nullable(FixedString(1)),
    rate_code_id            Nullable(UInt8),
    pickup_longitude        Nullable(Float64),
    pickup_latitude         Nullable(Float64),
    dropoff_longitude       Nullable(Float64),
    dropoff_latitude        Nullable(Float64),
    passenger_count         Nullable(UInt8),
    trip_distance           Nullable(Float64),
    fare_amount             Nullable(Float32),
    extra                   Nullable(Float32),
    mta_tax                 Nullable(Float32),
    tip_amount              Nullable(Float32),
    tolls_amount            Nullable(Float32),
    ehail_fee               Nullable(Float32),
    improvement_surcharge   Nullable(Float32),
    total_amount            Nullable(Float32),
    payment_type            Nullable(String),
    trip_type               Nullable(UInt8),
    pickup                  Nullable(String),
    dropoff                 Nullable(String),

    cab_type                Nullable(String),

    precipitation           Nullable(Int8),
    snow_depth              Nullable(Int8),
    snowfall                Nullable(Int8),
    max_temperature         Nullable(Int8),
    min_temperature         Nullable(Int8),
    average_wind_speed      Nullable(Int8),

    pickup_nyct2010_gid     Nullable(Int8),
    pickup_ctlabel          Nullable(String),
    pickup_borocode         Nullable(Int8),
    pickup_boroname         Nullable(String),
    pickup_ct2010           Nullable(String),
    pickup_boroct2010       Nullable(String),
    pickup_cdeligibil       Nullable(FixedString(1)),
    pickup_ntacode          Nullable(String),
    pickup_ntaname          Nullable(String),
    pickup_puma             Nullable(String),

    dropoff_nyct2010_gid    Nullable(UInt8),
    dropoff_ctlabel         Nullable(String),
    dropoff_borocode        Nullable(UInt8),
    dropoff_boroname        Nullable(String),
    dropoff_ct2010          Nullable(String),
    dropoff_boroct2010      Nullable(String),
    dropoff_cdeligibil      Nullable(String),
    dropoff_ntacode         Nullable(String),
    dropoff_ntaname         Nullable(String),
    dropoff_puma            Nullable(String)
) ENGINE = MergeTree(pickup_date, pickup_datetime, 8192);

Ona ou faʻamautinoa lea e mafai e le server muamua ona vaʻai i nodes uma e tolu i le fuifui.

SELECT *
FROM system.clusters
WHERE cluster = 'perftest_3shards'
FORMAT Vertical;
Row 1:
──────
cluster:          perftest_3shards
shard_num:        1
shard_weight:     1
replica_num:      1
host_name:        172.30.2.192
host_address:     172.30.2.192
port:             9000
is_local:         1
user:             default
default_database:
Row 2:
──────
cluster:          perftest_3shards
shard_num:        2
shard_weight:     1
replica_num:      1
host_name:        172.30.2.162
host_address:     172.30.2.162
port:             9000
is_local:         0
user:             default
default_database:

Row 3:
──────
cluster:          perftest_3shards
shard_num:        3
shard_weight:     1
replica_num:      1
host_name:        172.30.2.36
host_address:     172.30.2.36
port:             9000
is_local:         0
user:             default
default_database:

Ona ou faʻamalamalamaina lea o se laulau fou i luga o le server muamua e faʻavae i luga o le schema trips_mergetree_third ma fa'aaoga le afi Fa'asoa.

CREATE TABLE trips_mergetree_x3
    AS trips_mergetree_third
    ENGINE = Distributed(perftest_3shards,
                         default,
                         trips_mergetree_third,
                         rand());

Ona ou kopiina lea o faʻamatalaga mai le MergeTree faʻavae laulau i 'auʻaunaga uma e tolu. O mea nei na mae'a i le 34 minute ma le 44 sekone.

INSERT INTO trips_mergetree_x3
    SELECT * FROM trips_mergetree;

Ina ua maeʻa le taʻaloga i luga, na ou tuʻuina atu i le ClickHouse 15 minute e alu ese ai mai le maualuga o le faʻailoga maualuga. O faʻamaumauga faʻamaumauga na iu i le 264 GB, 34 GB ma le 33 GB taʻitasi i luga o 'auʻaunaga e tolu.

ClickHouse fa'avasegaga o fa'atinoga

O le mea na sosoo ai na ou vaʻaia o le taimi sili ona vave na ou vaʻaia ai le faʻatautaia o fesili taʻitasi i luga o le laulau i le tele o taimi trips_mergetree_x3.

$ clickhouse-client --host=0.0.0.0

O mea nei na mae'a ile 2.449 sekone.

SELECT cab_type, count(*)
FROM trips_mergetree_x3
GROUP BY cab_type;

O mea nei na mae'a ile 0.691 sekone.

SELECT passenger_count,
       avg(total_amount)
FROM trips_mergetree_x3
GROUP BY passenger_count;

O mea nei na mae'a ile 0 sekone.

SELECT passenger_count,
       toYear(pickup_date) AS year,
       count(*)
FROM trips_mergetree_x3
GROUP BY passenger_count,
         year;

O mea nei na mae'a ile 0.983 sekone.

SELECT passenger_count,
       toYear(pickup_date) AS year,
       round(trip_distance) AS distance,
       count(*)
FROM trips_mergetree_x3
GROUP BY passenger_count,
         year,
         distance
ORDER BY year,
         count(*) DESC;

Mo faʻatusatusaga, na ou faʻatautaia fesili tutusa i luga o le MergeTree-faʻavae laulau o loʻo nofo i luga o le server muamua.

Iloiloga faʻatinoga o le tasi ClickHouse node

O le mea na sosoo ai na ou vaʻaia o le taimi sili ona vave na ou vaʻaia ai le faʻatautaia o fesili taʻitasi i luga o le laulau i le tele o taimi trips_mergetree_x3.

O mea nei na mae'a ile 0.241 sekone.

SELECT cab_type, count(*)
FROM trips_mergetree
GROUP BY cab_type;

O mea nei na mae'a ile 0.826 sekone.

SELECT passenger_count,
       avg(total_amount)
FROM trips_mergetree
GROUP BY passenger_count;

O mea nei na mae'a ile 1.209 sekone.

SELECT passenger_count,
       toYear(pickup_date) AS year,
       count(*)
FROM trips_mergetree
GROUP BY passenger_count,
         year;

O mea nei na mae'a ile 1.781 sekone.

SELECT passenger_count,
       toYear(pickup_date) AS year,
       round(trip_distance) AS distance,
       count(*)
FROM trips_mergetree
GROUP BY passenger_count,
         year,
         distance
ORDER BY year,
         count(*) DESC;

Mafaufauga i taunuuga

O le taimi muamua lea na mafai ai e se faʻamaumauga tuʻufaʻatasia e leai se totogi o le CPU ona sili atu le faʻaogaina o le GPU i aʻu suʻega. O lena faʻamaumauga faʻavae GPU ua uma ona faia ni toe iloiloga se lua talu mai lena taimi, ae o le faʻatinoga na tuʻuina atu e le ClickHouse i luga o se node e tasi e matua manaia lava.

I le taimi lava e tasi, pe a faʻatinoina le Query 1 i luga o se afi tufatufaina, o tau i luga ole laiga o se faʻatonuga o le maualuga maualuga. Ou te faʻamoemoe na ou misia se mea i laʻu suʻesuʻega mo lenei pou aua e manaia le vaʻai i taimi o fesili e alu i lalo aʻo ou faʻaopoopoina nisi nodes i le fuifui. Ae ui i lea, e manaia pe a faʻatino isi fesili, faʻatupulaia le faʻatinoga e tusa ma le 2 taimi.

O le a manaia le vaʻai i le ClickHouse e faʻaleleia i le mafai ona tuʻu ese le teuina ma faʻatusatusa ina ia mafai ona latou fua faʻatasi. HDFS lagolago, lea na faʻaopoopoina i le tausaga talu ai, e mafai ona avea ma se laʻasaga i lenei mea. I le tulaga o le komipiuta, afai e mafai ona faatelevaveina se fesili se tasi e ala i le faʻaopoopoina o nisi nodes i le fuifui, o lona uiga o le lumanaʻi o lenei polokalama e matua susulu.

Faafetai mo le faʻaaluina o le taimi e faitau ai lenei pou. Ou te ofoina atu faufautua, tusiata, ma faʻataʻitaʻiga atinaʻe auaunaga i tagata faʻatau i Amerika i Matu ma Europa. Afai e te manaʻo e talanoa pe faʻapefea ona fesoasoani aʻu fautuaga i lau pisinisi, faʻamolemole faʻafesoʻotaʻi aʻu e ala i LinkedIn.

puna: www.habr.com

Faaopoopo i ai se faamatalaga