1.1 biliona huakaʻi kaʻa: 108-core ClickHouse cluster

Ua hoʻomākaukau ʻia ka unuhi ʻana o ka ʻatikala no nā haumāna o ka papa ʻEnekinia ʻIkepili.

1.1 biliona huakaʻi kaʻa: 108-core ClickHouse cluster

KaomiHouse he waihona kolamu kumu wehe. He wahi maikaʻi loa ia e hiki ai i nā haneli o nā mea loiloi ke nīnau wikiwiki i ka ʻikepili kikoʻī, ʻoiai ke hoʻokomo ʻia nā ʻumi piliona o nā moʻolelo hou i kēlā me kēia lā. Hiki ke kiʻekiʻe ke kumukūʻai o ka ʻoihana no ke kākoʻo ʻana i ia ʻōnaehana ma kahi o $100 i kēlā me kēia makahiki, a ʻo ka hapalua paha e pili ana i ka hoʻohana ʻana. I kekahi manawa, ua loaʻa i ka ClickHouse kahi hoʻonohonoho mai Yandex Metrics he 10 trillion moʻolelo. Ma waho aʻe o Yandex, ua loaʻa pū ʻo ClickHouse i ka kūleʻa me Bloomberg a me Cloudflare.

ʻElua makahiki i hala aku nei ua hoʻohana au kālailai hoʻohālikelike ʻikepili me ka hoʻohana ʻana i hoʻokahi mīkini, a lilo ia ka wikiwiki loa manuahi waihona lako polokalamu aʻu i ʻike ai. Mai ia manawa, ʻaʻole i ho'ōki nā mea hoʻomohala i ka hoʻohui ʻana i nā hiʻohiʻona, me ke kākoʻo no Kafka, HDFS a me ZStandard compression. I ka makahiki i hala ua hoʻohui lākou i ke kākoʻo no nā ʻano hoʻopiʻi cascading, a delta-mai-delta ua hiki ke coding. Ke kaomi ʻana i ka ʻikepili manawa, hiki ke hoʻopili maikaʻi ʻia nā waiwai ana me ka hoʻohana ʻana i ka delta encoding, akā no nā helu helu ʻoi aku ka maikaʻi o ka hoʻohana ʻana i ka delta-by-delta encoding. Ua lilo ka hoʻopili maikaʻi i ke kī i ka hana a ClickHouse.

Aia ʻo ClickHouse he 170 tausani laina o ka code C++, me ka ʻole o nā hale waihona puke ʻaoʻao ʻekolu, a ʻo ia kekahi o nā codebase waihona liʻiliʻi loa. I ka hoʻohālikelike ʻana, ʻaʻole kākoʻo ʻo SQLite i ka hāʻawi ʻana a loaʻa iā 235 tausani mau laina o ka code C. Ma kēia kākau ʻana, ua hāʻawi nā mea ʻenehana 207 iā ClickHouse, a ke piʻi nei ka ikaika o nā hana i kēia manawa.

I Malaki 2017, ua hoʻomaka ʻo ClickHouse e alakaʻi hoʻololi log ma ke ʻano he ala maʻalahi e mālama i ka hoʻomohala ʻana. Ua wāwahi pū lākou i ka faila palapala monolithic i kahi hierarchy file e pili ana iā Markdown. Hoʻopili ʻia nā pilikia a me nā hiʻohiʻona ma o GitHub, a ma ke ʻano maʻamau ua maʻalahi ka polokalamu i nā makahiki i hala.

Ma kēia ʻatikala, e nānā wau i ka hana o kahi hui ClickHouse ma AWS EC2 me ka hoʻohana ʻana i nā kaʻina hana 36-core a me ka waiho ʻana o NVMe.

UPDATE: I hoʻokahi pule ma hope o ka hoʻopuka mua ʻana i kēia pou, ua hana hou au i ka hoʻāʻo me kahi hoʻonohonoho hoʻomaikaʻi a loaʻa i nā hopena ʻoi aku ka maikaʻi. Ua hōʻano hou ʻia kēia pou e hōʻike i kēia mau loli.

Ke hoʻomaka nei i kahi AWS EC2 Cluster

E hoʻohana wau i ʻekolu mau manawa c5d.9xlarge EC2 no kēia pou. Loaʻa i kēlā me kēia o lākou nā CPU virtual 36, 72 GB o RAM, 900 GB o NVMe SSD mālama a kākoʻo i ka pūnaewele 10 Gigabit. Ua kūʻai lākou i $1,962 / hola i kēlā me kēia ma ka ʻāina eu-west-1 i ka wā e holo ana ma ke koi. E hoʻohana wau i ka Ubuntu Server 16.04 LTS e like me ka ʻōnaehana hana.

Hoʻonohonoho ʻia ka pā ahi i hiki i kēlā me kēia mīkini ke kamaʻilio me kekahi i kekahi me ka ʻole o ke kaohi ʻana, a ʻo kaʻu helu IPv4 wale nō i hoʻopaʻa inoa ʻia e SSH i loko o ka pūpū.

Hoʻokele NVMe i ke kūlana mākaukau hana

No ka hana ʻana o ClickHouse, e hana wau i kahi ʻōnaehana faila ma ka format EXT4 ma kahi kaʻa NVMe ma kēlā me kēia kikowaena.

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

Ke hoʻonohonohoʻia nā mea a pau, hiki iāʻoe keʻike i ka helu mauna a me ka 783 GB o ka hakahaka i loaʻa ma kēlā me kēia pūnaewele.

$ 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

ʻO ka ʻikepili aʻu e hoʻohana ai i kēia hoʻāʻo ʻana he waihona ʻikepili aʻu i hana ai mai 1.1 biliona holo kaʻa i lawe ʻia ma New York City i ʻeono makahiki. Ma ka blog Hoʻokahi Piliona huakaʻi kaʻaahi ma Redshift kikoʻī pehea aʻu i hōʻiliʻili ai i kēia pūʻulu ʻikepili. Mālama ʻia lākou ma AWS S3, no laila e hoʻonohonoho wau i ka AWS CLI me kaʻu komo a me nā kī huna.

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

E hoʻonoho au i ka palena noi like a ka mea kūʻai aku i 100 i ʻoi aku ka wikiwiki o ka hoʻoiho ʻana i nā faila ma mua o nā hoʻonohonoho paʻamau.

$ aws configure set 
    default.s3.max_concurrent_requests 
    100

E hoʻoiho au i ka ʻikepili kaʻa kaʻa mai AWS S3 a mālama iā ia ma kahi kaʻa NVMe ma ke kikowaena mua. ʻO kēia ʻikepili he ~104GB ma GZIP-compressed CSV format.

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

Hoʻokomo ʻia ʻo ClickHouse

E hoʻokomo wau i ka hāʻawi OpenJDK no Java 8 e like me ka mea e koi ʻia e holo Apache ZooKeeper, kahi i koi ʻia no ka hoʻokomo ʻana o ClickHouse ma nā mīkini ʻekolu.

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

A laila hoʻonoho wau i ka hoʻololi kaiapuni JAVA_HOME.

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

E hoʻohana au i ka ʻōnaehana hoʻokele pūʻolo o Ubuntu e hoʻokomo iā ClickHouse 18.16.1, glances a me ZooKeeper ma nā mīkini ʻekolu.

$ 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

E hana wau i papa kuhikuhi no ClickHouse a e hana pū kekahi i nā hoʻonohonoho hoʻonohonoho ma nā kikowaena ʻekolu.

$ 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

ʻO kēia nā hoʻonohonoho hoʻonohonoho aʻu e hoʻohana ai.

<?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>

E holo wau iā ZooKeeper a me ka server ClickHouse ma nā mīkini ʻekolu.

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

Hoʻouka i ka ʻikepili i ClickHouse

Ma ke kikowaena mua e hana au i papaʻaina huakaʻi (trips), ka mea e mālama i kahi papa helu o nā huakaʻi kaʻa me ka hoʻohana ʻana i ka mīkini 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;

A laila unuhi au a hoʻouka i kēlā me kēia o nā faila CSV i kahi papaʻaina huakaʻi (trips). Ua hoʻopau ʻia kēia mau mea i loko o 55 mau minuke a me 10 kekona. Ma hope o kēia hana, ʻo ka nui o ka papa kuhikuhi ʻikepili he 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)

ʻO 155 MB ka wikiwiki o ka lawe ʻana mai o ka maʻiʻo CSV paʻa ʻole i kēlā me kēia kekona. Manaʻo wau no ka bottleneck i ka decompression GZIP. Ua ʻoi aku ka wikiwiki o ka wehe ʻana i nā faila gzip a pau me ka hoʻohana ʻana i xargs a laila hoʻouka i ka ʻikepili unzipped. Aia ma lalo kahi wehewehe o ka mea i hōʻike ʻia i ka wā o ka hoʻokomo ʻana i ka 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

E hoʻokuʻu au i ka lewa ma ka NVMe drive ma ka holoi ʻana i nā faila CSV kumu ma mua o ka hoʻomau.

$ sudo rm -fr /ch/csv

E hoʻohuli i ke ʻano kolamu

E mālama ka ʻenekini Log ClickHouse i ka ʻikepili ma kahi ʻano lālani. No ka huli ʻana i ka ʻikepili me ka wikiwiki, hoʻololi au iā ia i ke ʻano columnar me ka hoʻohana ʻana i ka mīkini MergeTree.

$ clickhouse-client --host=0.0.0.0

Ua hoʻopau ʻia kēia mau mea i loko o 34 mau minuke a me 50 kekona. Ma hope o kēia hana, ʻo ka nui o ka papa kuhikuhi ʻikepili he 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;

ʻO kēia ke ʻano o ka puka aniani i ka wā o ka hana:

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)

Ma ka ho'āʻo hope loa, ua hoʻololi a helu hou ʻia kekahi mau kolamu. Ua ʻike au ʻaʻole hana hou kekahi o kēia mau hana e like me ka mea i manaʻo ʻia ma kēia waihona. No ka hoʻoponopono ʻana i kēia pilikia, ua wehe au i nā hana kūpono ʻole a hoʻouka i ka ʻikepili me ka ʻole o ka hoʻololi ʻana i nā ʻano granular.

Ka hāʻawi ʻana i ka ʻikepili ma ka pūʻulu

E puʻunaue au i ka ʻikepili ma waena o nā pūnana puʻupuʻu ʻekolu. No ka hoʻomaka, ma lalo wau e hana i kahi papaʻaina ma nā mīkini ʻekolu.

$ 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);

A laila, e hōʻoia wau e hiki i ke kikowaena mua ke ʻike i nā nodes ʻekolu i loko o ka pūpū.

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:

A laila, e wehewehe au i kahi papaʻaina hou ma ke kikowaena mua e pili ana i ka schema trips_mergetree_third a hoʻohana i ka ʻenekini Distributed.

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

A laila e kope wau i ka ʻikepili mai ka papa ʻaina MergeTree i nā kikowaena ʻekolu. Ua hoʻopau ʻia kēia mau mea i loko o 34 mau minuke a me 44 kekona.

INSERT INTO trips_mergetree_x3
    SELECT * FROM trips_mergetree;

Ma hope o ka hana ma luna, ua hāʻawi wau iā ClickHouse 15 mau minuke e neʻe aku mai ka māka pae kiʻekiʻe loa. Ua pau nā papa kuhikuhi ʻikepili i ka 264 GB, 34 GB a me 33 GB ma kēlā me kēia o nā kikowaena ʻekolu.

ClickHouse cluster hana loiloi

ʻO kaʻu mea i ʻike ai aʻe, ʻo ia ka manawa wikiwiki loa aʻu i ʻike ai e holo ana i kēlā me kēia nīnau ma ka papaʻaina i nā manawa he nui trips_mergetree_x3.

$ clickhouse-client --host=0.0.0.0

Ua hoʻopau ʻia ka mea ma hope i 2.449 kekona.

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

Ua hoʻopau ʻia ka mea ma hope i 0.691 kekona.

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

Ua hoʻopau ʻia kēia mau mea i 0 kekona.

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

Ua hoʻopau ʻia ka mea ma hope i 0.983 kekona.

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;

No ka hoʻohālikelike, holo wau i nā nīnau like ma kahi papa ʻaina MergeTree e noho wale ana ma ke kikowaena mua.

Ka loiloi hana o hoʻokahi node ClickHouse

ʻO kaʻu mea i ʻike ai aʻe, ʻo ia ka manawa wikiwiki loa aʻu i ʻike ai e holo ana i kēlā me kēia nīnau ma ka papaʻaina i nā manawa he nui trips_mergetree_x3.

Ua hoʻopau ʻia ka mea ma hope i 0.241 kekona.

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

Ua hoʻopau ʻia ka mea ma hope i 0.826 kekona.

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

Ua hoʻopau ʻia ka mea ma hope i 1.209 kekona.

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

Ua hoʻopau ʻia ka mea ma hope i 1.781 kekona.

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;

Nā noʻonoʻo i nā hopena

ʻO kēia ka manawa mua i hiki ai i kahi waihona waihona CPU manuahi ke hoʻokō i kahi waihona waihona GPU i kaʻu mau hoʻokolohua. Ua hele aku kēlā waihona waihona GPU i ʻelua mau hoʻoponopono mai ia manawa, akā ʻo ka hana a ClickHouse i hāʻawi ʻia ma kahi node hoʻokahi he mea kupanaha loa.

I ka manawa like, i ka wā e hoʻokō ai i ka Query 1 ma kahi ʻenekini puʻunaue, ʻoi aku ka nui o nā kumukūʻai ma luna. Manaʻo wau ua hala kekahi mea i kaʻu noiʻi no kēia pou no ka mea maikaʻi ke ʻike i ka iho ʻana o nā manawa nīnau i ka wā e hoʻohui ai au i nā nodes i ka hui. Eia nō naʻe, maikaʻi ia i ka wā e holo ana i nā nīnau ʻē aʻe, hoʻonui ʻia ka hana ma kahi o 2 mau manawa.

He mea maikaʻi ke ʻike i ka ulu ʻana o ClickHouse i ka hiki ke hoʻokaʻawale i ka waiho ʻana a me ka helu ʻana i hiki iā lākou ke hoʻonui kūʻokoʻa. ʻO ke kākoʻo HDFS, i hoʻohui ʻia i ka makahiki i hala, hiki ke hana i kēia. Ma ke ʻano o ka helu ʻana, inā hiki ke hoʻolōʻihi ʻia kahi nīnau hoʻokahi ma ka hoʻohui ʻana i nā nodes hou aʻe i ka pūʻulu, a laila ʻoi aku ka mālamalama o ka wā e hiki mai ana o kēia polokalamu.

Mahalo iā ʻoe no ka lawe ʻana i ka manawa e heluhelu ai i kēia pou. Hāʻawi wau i nā lawelawe kūkākūkā, hoʻolālā, a hoʻomaʻamaʻa hoʻomohala i nā mea kūʻai aku ma ʻAmelika ʻĀkau a me ʻEulopa. Inā makemake ʻoe e kūkākūkā pehea e hiki ai i kaʻu mau manaʻo ke kōkua i kāu ʻoihana, e ʻoluʻolu e kelepona mai iaʻu ma LinkedIn.

Source: www.habr.com

Pākuʻi i ka manaʻo hoʻopuka