1.1 biljun vjaġġ bit-taxi: cluster ClickHouse 108-core

It-traduzzjoni tal-artiklu tħejjiet speċifikament għall-istudenti tal-kors Inġinier tad-Data.

1.1 biljun vjaġġ bit-taxi: cluster ClickHouse 108-core

IkklikkjaHouse hija database ta' kolonni open source. Huwa ambjent mill-aqwa fejn mijiet ta' analisti jistgħu malajr jistaqsu dejta dettaljata, anke meta jiddaħħlu għexieren ta' biljuni ta' rekords ġodda kuljum. L-ispejjeż tal-infrastruttura biex isostnu sistema bħal din jistgħu jkunu għoljin sa $100 fis-sena, u potenzjalment nofshom skont l-użu. F'ħin minnhom, l-installazzjoni ClickHouse minn Yandex Metrics kien fiha 10 triljun rekords. Minbarra Yandex, ClickHouse sabet ukoll suċċess ma 'Bloomberg u Cloudflare.

Sentejn ilu qattajt analiżi komparattiva databases li jużaw magna waħda, u sar l-aktar mgħaġġla softwer tad-database b'xejn li qatt rajt. Minn dakinhar, l-iżviluppaturi komplew iżidu karatteristiċi, inkluż appoġġ għall-kompressjoni Kafka, HDFS u ZStandard. Is-sena li għaddiet żiedu appoġġ għal metodi ta 'kompressjoni kaskata, u delta-minn-delta il-kodifikazzjoni saret possibbli. Meta tikkompressa d-dejta tas-serje tal-ħin, il-valuri tal-kejl jistgħu jiġu kkompressati sew bl-użu tal-kodifikazzjoni delta, iżda għall-counters ikun aħjar li tuża kodifikazzjoni delta b'delta. Kompressjoni tajba saret iċ-ċavetta għall-prestazzjoni ta 'ClickHouse.

ClickHouse tikkonsisti minn 170 elf linja ta 'kodiċi C++, esklużi libreriji ta' partijiet terzi, u hija waħda mill-iżgħar bażijiet ta 'kodiċi ta' database distribwiti. B'paragun, SQLite ma jappoġġjax id-distribuzzjoni u jikkonsisti f'235 elf linja ta 'kodiċi C. Minn din il-kitba, 207 inġiniera kkontribwew għal ClickHouse, u l-intensità tal-kommits qed tiżdied reċentement.

F'Marzu 2017, ClickHouse bdiet tmexxi reġistru tal-bidla bħala mod faċli biex iżżomm rekord tal-iżvilupp. Huma wkoll kissru l-fajl ta 'dokumentazzjoni monolitika f'ġerarkija ta' fajl ibbażata fuq Markdown. Kwistjonijiet u karatteristiċi huma ssorveljati permezz ta 'GitHub, u b'mod ġenerali s-softwer sar ħafna aktar aċċessibbli f'dawn l-aħħar ftit snin.

F'dan l-artikolu, ser nagħti ħarsa lejn il-prestazzjoni ta 'cluster ClickHouse fuq AWS EC2 bl-użu ta' proċessuri 36-core u ħażna NVMe.

AĠĠORNAMENT: Ġimgħa wara li oriġinarjament ippublikajt din il-kariga, erġajt it-test b'konfigurazzjoni mtejba u ksibt riżultati ferm aħjar. Din il-kariga ġiet aġġornata biex tirrifletti dawn il-bidliet.

Tnedija ta' AWS EC2 Cluster

Se nkun qed nuża tliet istanzi c5d.9xlarge EC2 għal din il-kariga. Kull wieħed minnhom fih 36 CPUs virtwali, 72 GB ta 'RAM, 900 GB ta' ħażna NVMe SSD u jappoġġja netwerk 10 Gigabit. Dawn jiswew $1,962/siegħa kull wieħed fir-reġjun tal-eu-west-1 meta jaħdmu fuq talba. Se nkun qed nuża Ubuntu Server 16.04 LTS bħala s-sistema operattiva.

Il-firewall huwa kkonfigurat sabiex kull magna tkun tista 'tikkomunika ma' xulxin mingħajr restrizzjonijiet, u l-indirizz IPv4 tiegħi biss huwa whitelisted minn SSH fil-cluster.

Drive NVMe fl-istat ta' prontezza operattiva

Biex ClickHouse taħdem, se noħloq sistema ta 'fajls fil-format EXT4 fuq drive NVMe fuq kull wieħed mis-servers.

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

Ladarba kollox ikun ikkonfigurat, tista 'tara l-punt tal-muntatura u s-783 GB ta' spazju disponibbli fuq kull sistema.

$ 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

Is-sett tad-dejta li ser nuża f'dan it-test huwa dump tad-dejta li ġġenerajt minn 1.1 biljun vjaġġ tat-taxi meħuda fi New York City fuq sitt snin. Fuq il-blog Biljun Vjaġġ bit-Taxi f'Redshift dettalji dwar kif ġbart dan is-sett tad-dejta. Huma maħżuna fl-AWS S3, għalhekk ser nikkonfigura l-AWS CLI bl-aċċess u ċ-ċwievet sigrieti tiegħi.

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

Ser nissettja l-limitu tat-talba konkorrenti tal-klijent għal 100 sabiex il-fajls jitniżżel aktar malajr mis-settings default.

$ aws configure set 
    default.s3.max_concurrent_requests 
    100

Niżżel is-sett tad-dejta tat-taxi rides minn AWS S3 u naħżen fuq drive NVMe fuq l-ewwel server. Dan is-sett tad-dejta huwa ~104GB f'format CSV kompressat b'GZIP.

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

Installazzjoni ClickHouse

Se ninstalla d-distribuzzjoni OpenJDK għal Java 8 peress li hija meħtieġa biex tħaddem Apache ZooKeeper, li hija meħtieġa għal installazzjoni mqassma ta' ClickHouse fuq it-tliet magni kollha.

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

Imbagħad issettja l-varjabbli ambjentali JAVA_HOME.

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

Imbagħad se nuża s-sistema ta 'ġestjoni tal-pakketti ta' Ubuntu biex ninstalla ClickHouse 18.16.1, ħarsiet u ZooKeeper fuq it-tliet magni kollha.

$ 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

Ser noħloq direttorju għal ClickHouse u nagħmel ukoll xi konfigurazzjoni overrides fuq it-tliet servers kollha.

$ 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

Dawn huma l-konfigurazzjoni overrides li se nkun qed nuża.

<?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>

Imbagħad se nħaddem ZooKeeper u s-server ClickHouse fuq it-tliet magni kollha.

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

Tlugħ tad-dejta fuq ClickHouse

Fuq l-ewwel server se noħloq tabella tal-vjaġġ (trips), li se jaħżen dataset ta' vjaġġi bit-taxi bl-użu tal-magna 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;

Imbagħad niġbor u tagħbija kull wieħed mill-fajls CSV f'tabella tal-vjaġġ (trips). Dan li ġej tlesta f'55 minuta u 10 sekondi. Wara din l-operazzjoni, id-daqs tad-direttorju tad-dejta kien 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)

Il-veloċità tal-importazzjoni kienet 155 MB ta' kontenut CSV mhux kompressat kull sekonda. Nissuspetta li dan kien minħabba ostakolu fid-dekompressjoni tal-GZIP. Seta' kien aktar mgħaġġel li tħoll il-fajls kollha b'żipp b'mod parallel bl-użu ta' xargs u mbagħad tgħabbi d-dejta żżippjata. Hawn taħt hawn deskrizzjoni ta 'dak li ġie rrappurtat matul il-proċess ta' importazzjoni 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

Nillibera spazju fuq id-drajv NVMe billi nħassar il-fajls CSV oriġinali qabel inkompli.

$ sudo rm -fr /ch/csv

Ikkonverti għall-Formola tal-Kolonna

Il-magna Log ClickHouse se taħżen id-dejta f'format orjentat lejn ir-ringieli. Biex nistaqsi d-dejta aktar malajr, nikkonvertiha f'format kolonni billi tuża l-magna MergeTree.

$ clickhouse-client --host=0.0.0.0

Dan li ġej tlesta f'34 minuta u 50 sekonda. Wara din l-operazzjoni, id-daqs tad-direttorju tad-dejta kien 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;

Dan huwa kif dehret l-output t'għajn waqt l-operazzjoni:

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)

Fl-aħħar test, diversi kolonni ġew konvertiti u kkalkulati mill-ġdid. Sibt li xi wħud minn dawn il-funzjonijiet m'għadhomx jaħdmu kif mistenni fuq dan is-sett tad-dejta. Biex issolvi din il-problema, neħħejt il-funzjonijiet mhux xierqa u tgħabba d-dejta mingħajr ma kkonverti għal tipi aktar granulari.

Distribuzzjoni tad-data madwar il-cluster

Se nqassam id-dejta fit-tliet nodi tal-cluster. Biex tibda, hawn taħt se noħloq tabella fuq it-tliet magni.

$ 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);

Imbagħad ser niżgura li l-ewwel server jista 'jara t-tliet nodi fil-cluster.

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:

Imbagħad ser niddefinixxi tabella ġdida fuq l-ewwel server li hija bbażata fuq l-iskema trips_mergetree_third u juża l-magna Distributed.

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

Imbagħad ser nikkopja d-dejta mit-tabella bbażata fuq MergeTree għat-tliet servers kollha. Dan li ġej tlesta f'34 minuta u 44 sekonda.

INSERT INTO trips_mergetree_x3
    SELECT * FROM trips_mergetree;

Wara l-operazzjoni ta 'hawn fuq, tajt lil ClickHouse 15-il minuta biex titbiegħed mill-marka tal-livell massimu ta' ħażna. Id-direttorji tad-dejta spiċċaw 264 GB, 34 GB u 33 GB rispettivament fuq kull wieħed mit-tliet servers.

Evalwazzjoni tal-prestazzjoni tal-cluster ClickHouse

Dak li rajt li jmiss kien l-iktar ħin mgħaġġel li rajt imexxi kull mistoqsija fuq mejda diversi drabi trips_mergetree_x3.

$ clickhouse-client --host=0.0.0.0

Dan li ġej tlesta f'2.449 sekondi.

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

Dan li ġej tlesta f'0.691 sekondi.

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

Dan li ġej tlesta f'0 sekondi.

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

Dan li ġej tlesta f'0.983 sekondi.

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;

Għal paragun, għamilt l-istess mistoqsijiet fuq tabella bbażata fuq MergeTree li tirrisjedi biss fuq l-ewwel server.

Evalwazzjoni tal-prestazzjoni ta' node ClickHouse wieħed

Dak li rajt li jmiss kien l-iktar ħin mgħaġġel li rajt imexxi kull mistoqsija fuq mejda diversi drabi trips_mergetree_x3.

Dan li ġej tlesta f'0.241 sekondi.

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

Dan li ġej tlesta f'0.826 sekondi.

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

Dan li ġej tlesta f'1.209 sekondi.

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

Dan li ġej tlesta f'1.781 sekondi.

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;

Riflessjonijiet fuq ir-riżultati

Din hija l-ewwel darba li database bbażata fuq CPU b'xejn kienet kapaċi tegħleb database bbażata fuq GPU fit-testijiet tiegħi. Dik id-database bbażata fuq GPU għaddiet minn żewġ reviżjonijiet minn dakinhar, iżda l-prestazzjoni li ClickHouse wassal fuq nodu wieħed hija madankollu impressjonanti ħafna.

Fl-istess ħin, meta tesegwixxi Mistoqsija 1 fuq magna distribwita, l-ispejjeż ġenerali huma ordni ta 'kobor ogħla. Nispera li tlift xi ħaġa fir-riċerka tiegħi għal din il-kariga għax ikun sabiħ li nara l-ħinijiet tal-mistoqsijiet jonqsu hekk kif inżid aktar nodi mal-cluster. Madankollu, huwa kbir li meta tesegwixxi mistoqsijiet oħra, il-prestazzjoni żdiedet b'madwar 2 darbiet.

Ikun sabiħ li tara ClickHouse tevolvi biex tkun tista' tissepara l-ħażna u l-komputazzjoni sabiex ikunu jistgħu jiskalaw b'mod indipendenti. L-appoġġ tal-HDFS, li ġie miżjud is-sena li għaddiet, jista 'jkun pass lejn dan. F'termini ta 'kompjuters, jekk mistoqsija waħda tista' tiġi aċċellerata billi żżid aktar nodi mal-cluster, allura l-futur ta 'dan is-software huwa qawwi ħafna.

Grazzi talli ħadt il-ħin biex taqra din il-kariga. Noffri servizzi ta' konsulenza, arkitettura u żvilupp ta' prattika lill-klijenti fl-Amerika ta' Fuq u fl-Ewropa. Jekk tixtieq tiddiskuti kif is-suġġerimenti tiegħi jistgħu jgħinu lin-negozju tiegħek, jekk jogħġbok ikkuntattjani permezz LinkedIn.

Sors: www.habr.com

Żid kumment