This tutorial shows you essential techniques for getting fast reads and writes in CockroachDB, starting with a single-region deployment and expanding into multiple regions.
For a comprehensive list of tuning recommendations, only some of which are demonstrated here, see SQL Performance Best Practices.
Overview
Topology
You'll start with a 3-node CockroachDB cluster in a single Google Compute Engine (GCE) zone, with an extra instance for running a client application workload:
Within a single GCE zone, network latency between instances should be sub-millisecond.
You'll then scale the cluster to 9 nodes running across 3 GCE regions, with an extra instance in each region for a client application workload:
To reproduce the performance demonstrated in this tutorial:
- For each CockroachDB node, you'll use the
n1-standard-4
machine type (4 vCPUs, 15 GB memory) with the Ubuntu 16.04 OS image and a local SSD disk. - For running the client application workload, you'll use smaller instances, such as
n1-standard-1
.
Schema
Your schema and data will be based on the fictional peer-to-peer vehicle-sharing app, MovR, that was featured in the CockroachDB 2.0 demo:
A few notes about the schema:
- There are just three self-explanatory tables: In essence,
users
represents the people registered for the service,vehicles
represents the pool of vehicles for the service, andrides
represents when and where users have participated. - Each table has a composite primary key, with
city
being first in the key. Although not necessary initially in the single-region deployment, once you scale the cluster to multiple regions, these compound primary keys will enable you to geo-partition data at the row level bycity
. As such, this tutorial demonstrates a schema designed for future scaling. - The
IMPORT
feature you'll use to import the data does not support foreign keys, so you'll import the data without foreign key constraints. However, the import will create the secondary indexes required to add the foreign keys later. - The
rides
table contains bothcity
and the seemingly redundantvehicle_city
. This redundancy is necessary because, while it is not possible to apply more than one foreign key constraint to a single column, you will need to apply two foreign key constraints to therides
table, and each will require city as part of the constraint. The duplicatevehicle_city
, which is kept in sync withcity
via aCHECK
constraint, lets you overcome this limitation.
Important concepts
To understand the techniques in this tutorial, and to be able to apply them in your own scenarios, it's important to first review some important CockroachDB architectural concepts:
Term | Definition |
---|---|
Cluster | Your CockroachDB deployment, which acts as a single logical application. |
Node | An individual machine running CockroachDB. Many nodes join together to create your cluster. |
Range | CockroachDB stores all user data (tables, indexes, etc.) and almost all system data in a giant sorted map of key-value pairs. This keyspace is divided into "ranges", contiguous chunks of the keyspace, so that every key can always be found in a single range. From a SQL perspective, a table and its secondary indexes initially map to a single range, where each key-value pair in the range represents a single row in the table (also called the primary index because the table is sorted by the primary key) or a single row in a secondary index. As soon as a range reaches 64 MiB in size, it splits into two ranges. This process continues as the table and its indexes continue growing. |
Replica | CockroachDB replicates each range (3 times by default) and stores each replica on a different node. |
Leaseholder | For each range, one of the replicas holds the "range lease". This replica, referred to as the "leaseholder", is the one that receives and coordinates all read and write requests for the range. Unlike writes, read requests access the leaseholder and send the results to the client without needing to coordinate with any of the other range replicas. This reduces the network round trips involved and is possible because the leaseholder is guaranteed to be up-to-date due to the fact that all write requests also go to the leaseholder. |
Raft Leader | For each range, one of the replicas is the "leader" for write requests. Via the Raft consensus protocol, this replica ensures that a majority of replicas (the leader and enough followers) agree, based on their Raft logs, before committing the write. The Raft leader is almost always the same replica as the leaseholder. |
Raft Log | For each range, a time-ordered log of writes to the range that its replicas have agreed on. This log exists on-disk with each replica and is the range's source of truth for consistent replication. |
As mentioned above, when a query is executed, the cluster routes the request to the leaseholder for the range containing the relevant data. If the query touches multiple ranges, the request goes to multiple leaseholders. For a read request, only the leaseholder of the relevant range retrieves the data. For a write request, the Raft consensus protocol dictates that a majority of the replicas of the relevant range must agree before the write is committed.
Let's consider how these mechanics play out in some hypothetical queries.
Read scenario
First, imagine a simple read scenario where:
- There are 3 nodes in the cluster.
- There are 3 small tables, each fitting in a single range.
- Ranges are replicated 3 times (the default).
- A query is executed against node 2 to read from table 3.
In this case:
- Node 2 (the gateway node) receives the request to read from table 3.
- The leaseholder for table 3 is on node 3, so the request is routed there.
- Node 3 returns the data to node 2.
- Node 2 responds to the client.
If the query is received by the node that has the leaseholder for the relevant range, there are fewer network hops:
Write scenario
Now imagine a simple write scenario where a query is executed against node 3 to write to table 1:
In this case:
- Node 3 (the gateway node) receives the request to write to table 1.
- The leaseholder for table 1 is on node 1, so the request is routed there.
- The leaseholder is the same replica as the Raft leader (as is typical), so it simultaneously appends the write to its own Raft log and notifies its follower replicas on nodes 2 and 3.
- As soon as one follower has appended the write to its Raft log (and thus a majority of replicas agree based on identical Raft logs), it notifies the leader and the write is committed to the key-values on the agreeing replicas. In this diagram, the follower on node 2 acknowledged the write, but it could just as well have been the follower on node 3. Also note that the follower not involved in the consensus agreement usually commits the write very soon after the others.
- Node 1 returns acknowledgement of the commit to node 3.
- Node 3 responds to the client.
Just as in the read scenario, if the write request is received by the node that has the leaseholder and Raft leader for the relevant range, there are fewer network hops:
Network and I/O bottlenecks
With the above examples in mind, it's always important to consider network latency and disk I/O as potential performance bottlenecks. In summary:
- For reads, hops between the gateway node and the leaseholder add latency.
- For writes, hops between the gateway node and the leaseholder/Raft leader, and hops between the leaseholder/Raft leader and Raft followers, add latency. In addition, since Raft log entries are persisted to disk before a write is committed, disk I/O is important.
Single-region deployment
Step 1. Configure your network
CockroachDB requires TCP communication on two ports:
- 26257 (
tcp:26257
) for inter-node communication (i.e., working as a cluster) - 8080 (
tcp:8080
) for accessing the Web UI
Since GCE instances communicate on their internal IP addresses by default, you do not need to take any action to enable inter-node communication. However, if you want to access the Web UI from your local network, you must create a firewall rule for your project:
Field | Recommended Value |
---|---|
Name | cockroachweb |
Source filter | IP ranges |
Source IP ranges | Your local network's IP ranges |
Allowed protocols | tcp:8080 |
Target tags | cockroachdb |
The tag feature will let you easily apply the rule to your instances.
Step 2. Create instances
You'll start with a 3-node CockroachDB cluster in the us-east1-b
GCE zone, with an extra instance for running a client application workload.
Create 3 instances for your CockroachDB nodes. While creating each instance:
- Select the
us-east1-b
zone. - Use the
n1-standard-4
machine type (4 vCPUs, 15 GB memory). - Use the Ubuntu 16.04 OS image.
- Create and mount a local SSD.
- To apply the Web UI firewall rule you created earlier, click Management, disk, networking, SSH keys, select the Networking tab, and then enter
cockroachdb
in the Network tags field.
- Select the
Note the internal IP address of each
n1-standard-4
instance. You'll need these addresses when starting the CockroachDB nodes.Create a separate instance for running a client application workload, also in the
us-east1-b
zone. This instance can be smaller, such asn1-standard-1
.
Step 3. Start a 3-node cluster
SSH to the first
n1-standard-4
instance.Download the CockroachDB archive for Linux, extract the binary, and copy it into the
PATH
:$ curl https://binaries.cockroachdb.com/cockroach-v2.0.7.linux-amd64.tgz \ | tar -xz
$ sudo cp -i cockroach-v2.0.7.linux-amd64/cockroach /usr/local/bin/
Run the
cockroach start
command:$ cockroach start \ --insecure \ --advertise-host=<node1 internal address> \ --join=<node1 internal address>:26257,<node2 internal address>:26257,<node3 internal address>:26257 \ --locality=cloud=gce,region=us-east1,zone=us-east1-b \ --cache=.25 \ --max-sql-memory=.25 \ --background
Repeat steps 1 - 3 for the other two
n1-standard-4
instances.On any of the
n1-standard-4
instances, run thecockroach init
command:$ cockroach init --insecure --host=localhost
Each node then prints helpful details to the standard output, such as the CockroachDB version, the URL for the Web UI, and the SQL URL for clients.
Step 4. Import the Movr dataset
Now you'll import Movr data representing users, vehicles, and rides in 3 eastern US cities (New York, Boston, and Washington DC) and 3 western US cities (Los Angeles, San Francisco, and Seattle).
SSH to the fourth instance, the one not running a CockroachDB node.
Download the CockroachDB archive for Linux, and extract the binary:
$ curl https://binaries.cockroachdb.com/cockroach-v2.0.7.linux-amd64.tgz \ | tar -xz
Copy the binary into the
PATH
:$ sudo cp -i cockroach-v2.0.7.linux-amd64/cockroach /usr/local/bin/
Start the built-in SQL shell, pointing it at one of the CockroachDB nodes:
$ cockroach sql --insecure --host=<address of any node>
Create the
movr
database and set it as the default:> CREATE DATABASE movr;
> SET DATABASE = movr;
Use the
IMPORT
statement to create and populate theusers
,vehicles,
andrides
tables:> IMPORT TABLE users ( id UUID NOT NULL, city STRING NOT NULL, name STRING NULL, address STRING NULL, credit_card STRING NULL, CONSTRAINT "primary" PRIMARY KEY (city ASC, id ASC) ) CSV DATA ( 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/users/n1.0.csv' );
+--------------------+-----------+--------------------+------+---------------+----------------+-------+ | job_id | status | fraction_completed | rows | index_entries | system_records | bytes | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ | 370636591722889217 | succeeded | 1 | 0 | 0 | 0 | 0 | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ (1 row) Time: 3.409449563s
> IMPORT TABLE vehicles ( id UUID NOT NULL, city STRING NOT NULL, type STRING NULL, owner_id UUID NULL, creation_time TIMESTAMP NULL, status STRING NULL, ext JSON NULL, mycol STRING NULL, CONSTRAINT "primary" PRIMARY KEY (city ASC, id ASC), INDEX vehicles_auto_index_fk_city_ref_users (city ASC, owner_id ASC) ) CSV DATA ( 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/vehicles/n1.0.csv' );
+--------------------+-----------+--------------------+------+---------------+----------------+-------+ | job_id | status | fraction_completed | rows | index_entries | system_records | bytes | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ | 370636877487505409 | succeeded | 1 | 0 | 0 | 0 | 0 | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ (1 row) Time: 5.646142826s
> IMPORT TABLE rides ( id UUID NOT NULL, city STRING NOT NULL, vehicle_city STRING NULL, rider_id UUID NULL, vehicle_id UUID NULL, start_address STRING NULL, end_address STRING NULL, start_time TIMESTAMP NULL, end_time TIMESTAMP NULL, revenue DECIMAL(10,2) NULL, CONSTRAINT "primary" PRIMARY KEY (city ASC, id ASC), INDEX rides_auto_index_fk_city_ref_users (city ASC, rider_id ASC), INDEX rides_auto_index_fk_vehicle_city_ref_vehicles (vehicle_city ASC, vehicle_id ASC), CONSTRAINT check_vehicle_city_city CHECK (vehicle_city = city) ) CSV DATA ( 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.0.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.1.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.2.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.3.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.4.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.5.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.6.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.7.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.8.csv', 'https://s3-us-west-1.amazonaws.com/cockroachdb-movr/datasets/perf-tuning/rides/n1.9.csv' );
+--------------------+-----------+--------------------+------+---------------+----------------+-------+ | job_id | status | fraction_completed | rows | index_entries | system_records | bytes | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ | 370636986413285377 | succeeded | 1 | 0 | 0 | 0 | 0 | +--------------------+-----------+--------------------+------+---------------+----------------+-------+ (1 row) Time: 42.781522085s
Tip:You can observe the progress of imports as well as all schema change operations (e.g., adding secondary indexes) on the Jobs page of the Web UI.
Logically, there should be a number of foreign key relationships between the tables:
Referencing columns Referenced columns vehicles.city
,vehicles.owner_id
users.city
,users.id
rides.city
,rides.rider_id
users.city
,users.id
rides.vehicle_city
,rides.vehicle_id
vehicles.city
,vehicles.id
As mentioned earlier, it wasn't possible to put these relationships in place during
IMPORT
, but it was possible to create the required secondary indexes. Now, let's add the foreign key constraints:> ALTER TABLE vehicles ADD CONSTRAINT fk_city_ref_users FOREIGN KEY (city, owner_id) REFERENCES users (city, id);
> ALTER TABLE rides ADD CONSTRAINT fk_city_ref_users FOREIGN KEY (city, rider_id) REFERENCES users (city, id);
> ALTER TABLE rides ADD CONSTRAINT fk_vehicle_city_ref_vehicles FOREIGN KEY (vehicle_city, vehicle_id) REFERENCES vehicles (city, id);
Exit the built-in SQL shell:
> \q
Step 5. Install the Python client
When measuring SQL performance, it's best to run a given statement multiple times and look at the average and/or cumulative latency. For that purpose, you'll install and use a Python testing client.
Still on the fourth instance, make sure all of the system software is up-to-date:
$ sudo apt-get update && sudo apt-get -y upgrade
Install the
psycopg2
driver:$ sudo apt-get install python-psycopg2
Download the Python client:
$ wget https://raw.githubusercontent.com/cockroachdb/docs/master/_includes/v2.0/performance/tuning.py \ && chmod +x tuning.py
As you'll see below, this client lets you pass command-line flags:
Flag Description --host
The IP address of the target node. This is used in the client's connection string. --statement
The SQL statement to execute. --repeat
The number of times to repeat the statement. This defaults to 20. When run, the client prints the average time in seconds across all repetitions of the statement. Optionally, you can pass two other flags,
--time
to print the execution time in seconds for each repetition of the statement, and--cumulative
to print the cumulative time in seconds for all repetitions.--cumulative
is particularly useful when testing writes.Tip:To get similar help directly in your shell, use
./tuning.py --help
.
Step 6. Test/tune read performance
- Filtering by the primary key
- Filtering by a non-indexed column (full table scan)
- Filtering by a secondary index
- Filtering by a secondary index storing additional columns
- Joining data from different tables
- Using
IN (list)
with a subquery - Using
IN (list)
with explicit values
Filtering by the primary key
Retrieving a single row based on the primary key will usually return in 2ms or less:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT * FROM rides WHERE city = 'boston' AND id = '000007ef-fa0f-4a6e-a089-ce74aa8d2276'" \
--repeat=50 \
--times
Result:
['id', 'city', 'vehicle_city', 'rider_id', 'vehicle_id', 'start_address', 'end_address', 'start_time', 'end_time', 'revenue']
['000007ef-fa0f-4a6e-a089-ce74aa8d2276', 'boston', 'boston', 'd66c386d-4b7b-48a7-93e6-f92b5e7916ab', '6628bbbc-00be-4891-bc00-c49f2f16a30b', '4081 Conner Courts\nSouth Taylor, VA 86921', '2808 Willis Wells Apt. 931\nMccoyberg, OH 10303-4879', '2018-07-20 01:46:46.003070', '2018-07-20 02:27:46.003070', '44.25']
Times (milliseconds):
[24.547100067138672, 0.7688999176025391, 0.6949901580810547, 0.8230209350585938, 0.698089599609375, 0.7278919219970703, 0.6978511810302734, 0.5998611450195312, 0.7150173187255859, 0.7338523864746094, 0.6768703460693359, 0.7460117340087891, 0.7028579711914062, 0.7121562957763672, 0.7579326629638672, 0.8080005645751953, 1.0089874267578125, 0.7259845733642578, 0.6411075592041016, 0.7269382476806641, 0.6339550018310547, 0.7460117340087891, 0.9441375732421875, 0.8139610290527344, 0.6990432739257812, 0.6339550018310547, 0.7319450378417969, 0.637054443359375, 0.6501674652099609, 0.7278919219970703, 0.7069110870361328, 0.5779266357421875, 0.6208419799804688, 0.9050369262695312, 0.7741451263427734, 0.5650520324707031, 0.6079673767089844, 0.6191730499267578, 0.7388591766357422, 0.5598068237304688, 0.6401538848876953, 0.6659030914306641, 0.6489753723144531, 0.621795654296875, 0.7548332214355469, 0.6010532379150391, 0.6990432739257812, 0.6699562072753906, 0.6210803985595703, 0.7240772247314453]
Average time (milliseconds):
1.18108272552
When reading from a table or index for the first time in a session, the query will be slower than usual because the node issuing the query loads the schema of the table or index into memory first. For this reason, the first query took 24ms, whereas all others were sub-millisecond.
Retrieving a subset of columns will usually be even faster:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT rider_id, vehicle_id \
FROM rides \
WHERE city = 'boston' AND id = '000007ef-fa0f-4a6e-a089-ce74aa8d2276'" \
--repeat=50 \
--times
Result:
['rider_id', 'vehicle_id']
['d66c386d-4b7b-48a7-93e6-f92b5e7916ab', '6628bbbc-00be-4891-bc00-c49f2f16a30b']
Times (milliseconds):
[1.2311935424804688, 0.7009506225585938, 0.5898475646972656, 0.6151199340820312, 0.5660057067871094, 0.6620883941650391, 0.5691051483154297, 0.5369186401367188, 0.5609989166259766, 0.5290508270263672, 0.5939006805419922, 0.5769729614257812, 0.5638599395751953, 0.5381107330322266, 0.61798095703125, 0.5879402160644531, 0.6008148193359375, 0.5900859832763672, 0.5190372467041016, 0.5409717559814453, 0.51116943359375, 0.5400180816650391, 0.5490779876708984, 0.4870891571044922, 0.5340576171875, 0.49591064453125, 0.5669593811035156, 0.4971027374267578, 0.5729198455810547, 0.514984130859375, 0.5309581756591797, 0.5099773406982422, 0.5550384521484375, 0.5328655242919922, 0.5559921264648438, 0.5319118499755859, 0.5059242248535156, 0.5719661712646484, 0.49614906311035156, 0.6041526794433594, 0.5080699920654297, 0.5240440368652344, 0.49591064453125, 0.5681514739990234, 0.5118846893310547, 0.5359649658203125, 0.5450248718261719, 0.5650520324707031, 0.5249977111816406, 0.5669593811035156]
Average time (milliseconds):
0.566024780273
Filtering by a non-indexed column (full table scan)
You'll get generally poor performance when retrieving a single row based on a column that is not in the primary key or any secondary index:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT * FROM users WHERE name = 'Natalie Cunningham'" \
--repeat=50 \
--times
Result:
['id', 'city', 'name', 'address', 'credit_card']
['02cc9e5b-1e91-4cdb-87c4-726b4ea7219a', 'boston', 'Natalie Cunningham', '97477 Lee Path\nKimberlyport, CA 65960', '4532613656695680']
Times (milliseconds):
[31.939983367919922, 4.055023193359375, 3.988981246948242, 4.395008087158203, 4.045009613037109, 3.838062286376953, 6.09898567199707, 4.03904914855957, 3.9091110229492188, 5.933046340942383, 6.157875061035156, 6.323814392089844, 4.379987716674805, 3.982067108154297, 4.28009033203125, 4.118919372558594, 4.222869873046875, 4.041910171508789, 3.9288997650146484, 4.031896591186523, 4.085063934326172, 3.996133804321289, 4.001140594482422, 6.031990051269531, 5.98597526550293, 4.163026809692383, 5.931854248046875, 5.897998809814453, 3.9229393005371094, 3.8909912109375, 3.7729740142822266, 3.9768218994140625, 3.9958953857421875, 4.265069961547852, 4.204988479614258, 4.142999649047852, 4.3659210205078125, 6.074190139770508, 4.015922546386719, 4.418849945068359, 3.9381980895996094, 4.222869873046875, 4.694938659667969, 3.9060115814208984, 3.857851028442383, 3.8509368896484375, 3.969907760620117, 4.241943359375, 4.032135009765625, 3.9670467376708984]
Average time (milliseconds):
4.99066352844
To understand why this query performs poorly, use the SQL client built into the cockroach
binary to EXPLAIN
the query plan:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT * FROM users WHERE name = 'Natalie Cunningham';"
+------+-------+---------------+
| Tree | Field | Description |
+------+-------+---------------+
| scan | | |
| | table | users@primary |
| | spans | ALL |
+------+-------+---------------+
(3 rows)
The row with spans | ALL
shows you that, without a secondary index on the name
column, CockroachDB scans every row of the users
table, ordered by the primary key (city
/id
), until it finds the row with the correct name
value.
Filtering by a secondary index
To speed up this query, add a secondary index on name
:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="CREATE INDEX on users (name);"
The query will now return much faster:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT * FROM users WHERE name = 'Natalie Cunningham'" \
--repeat=50 \
--times
Result:
['id', 'city', 'name', 'address', 'credit_card']
['02cc9e5b-1e91-4cdb-87c4-726b4ea7219a', 'boston', 'Natalie Cunningham', '97477 Lee Path\nKimberlyport, CA 65960', '4532613656695680']
Times (milliseconds):
[3.4589767456054688, 1.6651153564453125, 1.547098159790039, 1.9190311431884766, 1.7499923706054688, 1.6219615936279297, 1.5749931335449219, 1.7859935760498047, 1.5561580657958984, 1.6391277313232422, 1.5120506286621094, 1.5139579772949219, 1.6808509826660156, 1.708984375, 1.4798641204833984, 1.544952392578125, 1.653909683227539, 1.6129016876220703, 1.7309188842773438, 1.5811920166015625, 1.7628669738769531, 1.5459060668945312, 1.6429424285888672, 1.6558170318603516, 1.7898082733154297, 1.6138553619384766, 1.6868114471435547, 1.5490055084228516, 1.7120838165283203, 1.6911029815673828, 1.5289783477783203, 1.5990734100341797, 1.6109943389892578, 1.5058517456054688, 1.5058517456054688, 1.6798973083496094, 1.7499923706054688, 1.5850067138671875, 1.4929771423339844, 1.6651153564453125, 1.5921592712402344, 1.6739368438720703, 1.6529560089111328, 1.6019344329833984, 1.6429424285888672, 1.5649795532226562, 1.605987548828125, 1.550912857055664, 1.6069412231445312, 1.6779899597167969]
Average time (milliseconds):
1.66565418243
To understand why performance improved from 4.99ms (without index) to 1.66ms (with index), use EXPLAIN
to see the new query plan:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT * FROM users WHERE name = 'Natalie Cunningham';"
+------------+-------+-------------------------------------------------------+
| Tree | Field | Description |
+------------+-------+-------------------------------------------------------+
| index-join | | |
| ├── scan | | |
| │ | table | users@users_name_idx |
| │ | spans | /"Natalie Cunningham"-/"Natalie Cunningham"/PrefixEnd |
| └── scan | | |
| | table | users@primary |
+------------+-------+-------------------------------------------------------+
(6 rows)
This shows you that CockroachDB starts with the secondary index (table | users@users_name_idx
). Because it is sorted by name
, the query can jump directly to the relevant value (spans | /"Natalie Cunningham"-/"Natalie Cunningham"/PrefixEnd
). However, the query needs to return values not in the secondary index, so CockroachDB grabs the primary key (city
/id
) stored with the name
value (the primary key is always stored with entries in a secondary index), jumps to that value in the primary index, and then returns the full row.
Thinking back to the earlier discussion of ranges and leaseholders, because the users
table is small (under 64 MiB), the primary index and all secondary indexes are contained in a single range with a single leaseholder. If the table were bigger, however, the primary index and secondary index could reside in separate ranges, each with its own leaseholder. In this case, if the leaseholders were on different nodes, the query would require more network hops, further increasing latency.
Filtering by a secondary index storing additional columns
When you have a query that filters by a specific column but retrieves a subset of the table's total columns, you can improve performance by storing those additional columns in the secondary index to prevent the query from needing to scan the primary index as well.
For example, let's say you frequently retrieve a user's name and credit card number:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT name, credit_card FROM users WHERE name = 'Natalie Cunningham'" \
--repeat=50 \
--times
Result:
['name', 'credit_card']
['Natalie Cunningham', '4532613656695680']
Times (milliseconds):
[2.338886260986328, 1.7859935760498047, 1.9490718841552734, 1.550912857055664, 1.4331340789794922, 1.4619827270507812, 1.425027847290039, 1.8270015716552734, 1.6829967498779297, 1.6028881072998047, 1.628875732421875, 1.4889240264892578, 1.497030258178711, 1.5380382537841797, 1.486063003540039, 1.5859603881835938, 1.7290115356445312, 1.7409324645996094, 1.5869140625, 1.6489028930664062, 1.7418861389160156, 1.5971660614013672, 1.619100570678711, 1.6379356384277344, 1.6028881072998047, 1.6531944274902344, 1.667022705078125, 1.6241073608398438, 1.5468597412109375, 1.5778541564941406, 1.6779899597167969, 1.5718936920166016, 1.5950202941894531, 1.6407966613769531, 1.538991928100586, 1.8379688262939453, 1.7008781433105469, 1.837015151977539, 1.5687942504882812, 1.7828941345214844, 1.7290115356445312, 1.6810894012451172, 1.7969608306884766, 1.5821456909179688, 1.569986343383789, 1.5740394592285156, 1.8229484558105469, 1.7371177673339844, 1.7681121826171875, 1.6360282897949219]
Average time (milliseconds):
1.65812492371
With the current secondary index on name
, CockroachDB still needs to scan the primary index to get the credit card number:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT name, credit_card FROM users WHERE name = 'Natalie Cunningham';"
+-----------------+-------+-------------------------------------------------------+
| Tree | Field | Description |
+-----------------+-------+-------------------------------------------------------+
| render | | |
| └── index-join | | |
| ├── scan | | |
| │ | table | users@users_name_idx |
| │ | spans | /"Natalie Cunningham"-/"Natalie Cunningham"/PrefixEnd |
| └── scan | | |
| | table | users@primary |
+-----------------+-------+-------------------------------------------------------+
(7 rows)
Let's drop and recreate the index on name
, this time storing the credit_card
value in the index:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="DROP INDEX users_name_idx;"
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="CREATE INDEX ON users (name) STORING (credit_card);"
Now that credit_card
values are stored in the index on name
, CockroachDB only needs to scan that index:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT name, credit_card FROM users WHERE name = 'Natalie Cunningham';"
+-----------+-------+-------------------------------------------------------+
| Tree | Field | Description |
+-----------+-------+-------------------------------------------------------+
| render | | |
| └── scan | | |
| | table | users@users_name_idx |
| | spans | /"Natalie Cunningham"-/"Natalie Cunningham"/PrefixEnd |
+-----------+-------+-------------------------------------------------------+
(4 rows)
This results in even faster performance, reducing latency from 1.65ms (index without storing) to 1.04ms (index with storing):
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT name, credit_card FROM users WHERE name = 'Natalie Cunningham'" \
--repeat=50 \
--times
Result:
['name', 'credit_card']
['Natalie Cunningham', '4532613656695680']
Times (milliseconds):
[1.8949508666992188, 1.2660026550292969, 1.2140274047851562, 1.110076904296875, 1.4989376068115234, 1.1739730834960938, 1.2331008911132812, 0.9701251983642578, 0.9019374847412109, 0.9038448333740234, 1.016855239868164, 0.9331703186035156, 0.9179115295410156, 0.9288787841796875, 0.888824462890625, 0.9429454803466797, 0.9410381317138672, 1.001119613647461, 0.9438991546630859, 0.9849071502685547, 1.0221004486083984, 1.013040542602539, 1.0149478912353516, 0.9579658508300781, 1.0061264038085938, 1.0559558868408203, 1.0788440704345703, 1.0411739349365234, 0.9610652923583984, 0.9639263153076172, 1.1239051818847656, 0.9639263153076172, 1.058816909790039, 0.949859619140625, 0.9739398956298828, 1.046895980834961, 0.9260177612304688, 1.0569095611572266, 1.033782958984375, 1.1029243469238281, 0.9710788726806641, 1.0311603546142578, 0.9870529174804688, 1.1179447174072266, 1.0349750518798828, 1.088857650756836, 1.1060237884521484, 1.0170936584472656, 1.0180473327636719, 1.0519027709960938]
Average time (milliseconds):
1.04885578156
Joining data from different tables
Secondary indexes are crucial when joining data from different tables as well.
For example, let's say you want to count the number of users who started rides on a given day. To do this, you need to use a join to get the relevant rides from the rides
table and then map the rider_id
for each of those rides to the corresponding id
in the users
table, counting each mapping only once:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT count(DISTINCT users.id) \
FROM users \
INNER JOIN rides ON rides.rider_id = users.id \
WHERE start_time BETWEEN '2018-07-20 00:00:00' AND '2018-07-21 00:00:00'" \
--repeat=50 \
--times
Result:
['count']
['1998']
Times (milliseconds):
[1663.2239818572998, 841.871976852417, 844.9788093566895, 1043.7190532684326, 1047.544002532959, 1049.0870475769043, 1079.737901687622, 1049.543857574463, 1069.1118240356445, 1104.2020320892334, 1071.1669921875, 1080.1141262054443, 1066.741943359375, 1071.8858242034912, 1073.8670825958252, 1054.008960723877, 1089.4761085510254, 1048.2399463653564, 1033.8318347930908, 1078.5980224609375, 1054.8391342163086, 1095.6230163574219, 1056.9767951965332, 1082.8359127044678, 1048.3272075653076, 1050.3859519958496, 1084.2180252075195, 1082.1950435638428, 1101.97114944458, 1079.9469947814941, 1065.234899520874, 1051.058053970337, 1105.48996925354, 1119.469165802002, 1089.8759365081787, 1082.5989246368408, 1074.9430656433105, 1067.4428939819336, 1066.5888786315918, 1069.6449279785156, 1067.9738521575928, 1082.4880599975586, 1037.9269123077393, 1042.2871112823486, 1130.7330131530762, 1150.7518291473389, 1165.3728485107422, 1136.9531154632568, 1120.3861236572266, 1126.8589496612549]
Average time (milliseconds):
1081.04698181
To understand what's happening, use EXPLAIN
to see the query plan:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT count(DISTINCT users.id) \
FROM users \
INNER JOIN rides ON rides.rider_id = users.id \
WHERE start_time BETWEEN '2018-07-20 00:00:00' AND '2018-07-21 00:00:00';"
+---------------------+----------+-------------------+
| Tree | Field | Description |
+---------------------+----------+-------------------+
| group | | |
| └── render | | |
| └── join | | |
| │ | type | inner |
| │ | equality | (id) = (rider_id) |
| ├── scan | | |
| │ | table | users@primary |
| │ | spans | ALL |
| └── scan | | |
| | table | rides@primary |
| | spans | ALL |
+---------------------+----------+-------------------+
(11 rows)
Reading from bottom up, you can see that CockroachDB does a full table scan (spans | ALL
) first on rides
to get all rows with a start_time
in the specified range and then does another full table scan on users
to find matching rows and calculate the count.
Given that the rides
table is large, its data is split across several ranges. Each range is replicated and has a leaseholder. At least some of these leaseholders are likely located on different nodes. This means that the full table scan of rides
involves several network hops to various leaseholders before finally going to the leaseholder for users
to do a full table scan there.
To track this specifically, let's use the SHOW EXPERIMENTAL_RANGES
statement to find out where the relevant leaseholders reside for rides
and users
:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW EXPERIMENTAL_RANGES FROM TABLE rides;"
+------------------------------------------------------------------------+------------------------------------------------------------------------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+------------------------------------------------------------------------+------------------------------------------------------------------------+----------+----------+--------------+
| NULL | /"boston"/"\xfe\xdd?\xbb4\xabOV\x84\x00M\x89#-a6"/PrefixEnd | 23 | {1,2,3} | 1 |
| /"boston"/"\xfe\xdd?\xbb4\xabOV\x84\x00M\x89#-a6"/PrefixEnd | /"los angeles"/"\xf1\xe8\x99eǵI\x16\xb9w\a\xd01\xcc\b\xa4"/PrefixEnd | 25 | {1,2,3} | 2 |
| /"los angeles"/"\xf1\xe8\x99eǵI\x16\xb9w\a\xd01\xcc\b\xa4"/PrefixEnd | /"new york"/"\xebV\xf5\xe6P%L$\x92\xd2\xdf&\a\x81\xeeO"/PrefixEnd | 26 | {1,2,3} | 1 |
| /"new york"/"\xebV\xf5\xe6P%L$\x92\xd2\xdf&\a\x81\xeeO"/PrefixEnd | /"san francisco"/"\xda\xc5B\xe0\x0e\fK)\x98:\xe6[@\x05\x91*"/PrefixEnd | 27 | {1,2,3} | 2 |
| /"san francisco"/"\xda\xc5B\xe0\x0e\fK)\x98:\xe6[@\x05\x91*"/PrefixEnd | /"seattle"/"\xd4ˆ?\x98\x98FA\xa7m\x84\xba\xac\xf5\xbfI"/PrefixEnd | 28 | {1,2,3} | 3 |
| /"seattle"/"\xd4ˆ?\x98\x98FA\xa7m\x84\xba\xac\xf5\xbfI"/PrefixEnd | /"washington dc"/"Ņ\x06\x9d\xc2LEq\xb8<KG\a(\x18\xf6"/PrefixEnd | 29 | {1,2,3} | 1 |
| /"washington dc"/"Å…\x06\x9d\xc2LEq\xb8<KG\a(\x18\xf6"/PrefixEnd | NULL | 30 | {1,2,3} | 1 |
+------------------------------------------------------------------------+------------------------------------------------------------------------+----------+----------+--------------+
(7 rows)
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW EXPERIMENTAL_RANGES FROM TABLE users;"
+-----------+---------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+-----------+---------+----------+----------+--------------+
| NULL | NULL | 51 | {1,2,3} | 2 |
+-----------+---------+----------+----------+--------------+
(1 row)
The results above tell us:
- The
rides
table is split across 7 ranges, with four leaseholders on node 1, two leaseholders on node 2, and one leaseholder on node 3. - The
users
table is just a single range with its leaseholder on node 2.
Now, given the WHERE
condition of the join, the full table scan of rides
, across all of its 7 ranges, is particularly wasteful. To speed up the query, you can create a secondary index on the WHERE
condition (rides.start_time
) storing the join key (rides.rider_id
):
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="CREATE INDEX ON rides (start_time) STORING (rider_id);"
The rides
table contains 1 million rows, so adding this index will take a few minutes.
Adding the secondary index reduced the query time from 1081.04ms to 71.89ms:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT count(DISTINCT users.id) \
FROM users \
INNER JOIN rides ON rides.rider_id = users.id \
WHERE start_time BETWEEN '2018-07-20 00:00:00' AND '2018-07-21 00:00:00'" \
--repeat=50 \
--times
Result:
['count']
['1998']
Times (milliseconds):
[124.19795989990234, 83.74285697937012, 84.76495742797852, 76.9808292388916, 65.74702262878418, 62.478065490722656, 60.26411056518555, 59.99302864074707, 67.10195541381836, 73.45199584960938, 67.09504127502441, 60.45889854431152, 68.6960220336914, 61.94710731506348, 61.53106689453125, 60.44197082519531, 62.22796440124512, 89.34903144836426, 77.64196395874023, 71.43712043762207, 66.09010696411133, 63.668012619018555, 65.31286239624023, 77.1780014038086, 73.52113723754883, 68.84908676147461, 65.11712074279785, 65.34600257873535, 65.8869743347168, 76.90095901489258, 76.9491195678711, 69.39697265625, 64.23306465148926, 75.0880241394043, 69.34094429016113, 57.55496025085449, 65.79995155334473, 83.74285697937012, 75.32310485839844, 74.08809661865234, 77.33798027038574, 73.95505905151367, 71.85482978820801, 77.95405387878418, 74.30601119995117, 72.24106788635254, 75.28901100158691, 78.2630443572998, 74.97286796569824, 79.50282096862793]
Average time (milliseconds):
71.8922615051
To understand why performance improved, again use EXPLAIN
to see the new query plan:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT count(DISTINCT users.id) \
FROM users \
INNER JOIN rides ON rides.rider_id = users.id \
WHERE start_time BETWEEN '2018-07-20 00:00:00' AND '2018-07-21 00:00:00';"
+---------------------+----------+-------------------------------------------------------+
| Tree | Field | Description |
+---------------------+----------+-------------------------------------------------------+
| group | | |
| └── render | | |
| └── join | | |
| │ | type | inner |
| │ | equality | (id) = (rider_id) |
| ├── scan | | |
| │ | table | users@primary |
| │ | spans | ALL |
| └── scan | | |
| | table | rides@rides_start_time_idx |
| | spans | /2018-07-20T00:00:00Z-/2018-07-21T00:00:00.000000001Z |
+---------------------+----------+-------------------------------------------------------+
(11 rows)
Notice that CockroachDB now starts by using rides@rides_start_time_idx
secondary index to retrieve the relevant rides without needing to scan the full rides
table.
Let's check the ranges for the new index:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW EXPERIMENTAL_RANGES FROM INDEX rides@rides_start_time_idx;"
+-----------------------------------------------------------------------------------------+-----------------------------------------------------------------------------------------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+-----------------------------------------------------------------------------------------+-----------------------------------------------------------------------------------------+----------+----------+--------------+
| NULL | /2018-07-15T02:32:47.564891Z/"seattle"/"r\x8f\xbc\xd4\f\x18E\x9f\x85\xc2\"H\\\xe7k\xf1" | 34 | {1,2,3} | 1 |
| /2018-07-15T02:32:47.564891Z/"seattle"/"r\x8f\xbc\xd4\f\x18E\x9f\x85\xc2\"H\\\xe7k\xf1" | NULL | 35 | {1,2,3} | 1 |
+-----------------------------------------------------------------------------------------+-----------------------------------------------------------------------------------------+----------+----------+--------------+
(2 rows)
This tells us that the index is stored in 2 ranges, with the leaseholders for both of them on node 1. Based on the output of SHOW EXPERIMENTAL_RANGES FROM TABLE users
that we saw earlier, we already know that the leaseholder for the users
table is on node 2.
Using IN (list)
with a subquery
Now let's say you want to get the latest ride of each of the 5 most used vehicles. To do this, you might think to use a subquery to get the IDs of the 5 most frequent vehicles from the rides
table, passing the results into the IN
list of another query to get the most recent ride of each of the 5 vehicles:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT vehicle_id, max(end_time) \
FROM rides \
WHERE vehicle_id IN ( \
SELECT vehicle_id \
FROM rides \
GROUP BY vehicle_id \
ORDER BY count(*) DESC \
LIMIT 5 \
) \
GROUP BY vehicle_id" \
--repeat=20 \
--times
Result:
['vehicle_id', 'max']
['c6541da5-9858-4e3f-9b49-992e206d2c50', '2018-08-02 02:14:50.543760']
['78fdd6f8-c6a1-42df-a89f-cd65b7bb8be9', '2018-08-02 02:47:43.755989']
['3c950d36-c2b8-48d0-87d3-e0d6f570af62', '2018-08-02 03:06:31.293184']
['35752c4c-b878-4436-8330-8d7246406a55', '2018-08-02 03:08:49.823209']
['0962cdca-9d85-457c-9616-cc2ae2d32008', '2018-08-02 03:01:25.414512']
Times (milliseconds):
[4368.9610958099365, 4373.898029327393, 4396.070957183838, 4382.591962814331, 4274.624824523926, 4369.847059249878, 4373.079061508179, 4287.877082824707, 4307.362079620361, 4368.865966796875, 4363.792896270752, 4310.600996017456, 4378.695011138916, 4340.383052825928, 4338.238000869751, 4373.046875, 4327.131986618042, 4386.303901672363, 4429.6300411224365, 4383.068084716797]
Average time (milliseconds):
4356.7034483
However, as you can see, this query is slow because, currently, when the WHERE
condition of a query comes from the result of a subquery, CockroachDB scans the entire table, even if there is an available index. Use EXPLAIN
to see this in more detail:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="EXPLAIN SELECT vehicle_id, max(end_time) \
FROM rides \
WHERE vehicle_id IN ( \
SELECT vehicle_id \
FROM rides \
GROUP BY vehicle_id \
ORDER BY count(*) DESC \
LIMIT 5 \
) \
GROUP BY vehicle_id;"
+------------------------------------+-----------+--------------------------------------------------------------------------+
| Tree | Field | Description |
+------------------------------------+-----------+--------------------------------------------------------------------------+
| root | | |
| ├── group | | |
| │ │ | group by | @1-@1 |
| │ └── render | | |
| │ └── scan | | |
| │ | table | rides@primary |
| │ | spans | ALL |
| └── subquery | | |
| │ | id | @S1 |
| │ | sql | (SELECT vehicle_id FROM rides GROUP BY vehicle_id ORDER BY count(*) DESC |
| | | LIMIT 5) |
| │ | exec mode | all rows normalized |
| └── limit | | |
| └── sort | | |
| │ | order | -count |
| │ | strategy | top 5 |
| └── group | | |
| │ | group by | @1-@1 |
| └── render | | |
| └── scan | | |
| | table | rides@primary |
| | spans | ALL |
+------------------------------------+-----------+--------------------------------------------------------------------------+
(21 rows)
This is a complex query plan, but the important thing to note is the full table scan of rides@primary
above the subquery
. This shows you that, after the subquery returns the IDs of the top 5 vehicles, CockroachDB scans the entire primary index to find the rows with max(end_time)
for each vehicle_id
, although you might expect CockroachDB to more efficiently use the secondary index on vehicle_id
(CockroachDB is working to remove this limitation in a future version).
Using IN (list)
with explicit values
Because CockroachDB will not use an available secondary index when using IN (list)
with a subquery, it's much more performant to have your application first select the top 5 vehicles:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT vehicle_id \
FROM rides \
GROUP BY vehicle_id \
ORDER BY count(*) DESC \
LIMIT 5" \
--repeat=20 \
--times
Result:
['vehicle_id']
['35752c4c-b878-4436-8330-8d7246406a55']
['0962cdca-9d85-457c-9616-cc2ae2d32008']
['c6541da5-9858-4e3f-9b49-992e206d2c50']
['78fdd6f8-c6a1-42df-a89f-cd65b7bb8be9']
['3c950d36-c2b8-48d0-87d3-e0d6f570af62']
Times (milliseconds):
[787.0969772338867, 782.2480201721191, 741.5878772735596, 790.3921604156494, 767.4920558929443, 733.0870628356934, 768.8038349151611, 754.1589736938477, 716.4630889892578, 726.3698577880859, 721.092939376831, 737.1737957000732, 747.978925704956, 736.1149787902832, 727.1649837493896, 725.5918979644775, 746.1550235748291, 752.6230812072754, 728.59787940979, 733.4978580474854]
Average time (milliseconds):
746.184563637
And then put the results into the IN
list to get the most recent rides of the vehicles:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT vehicle_id, max(end_time) \
FROM rides \
WHERE vehicle_id IN ( \
'35752c4c-b878-4436-8330-8d7246406a55', \
'0962cdca-9d85-457c-9616-cc2ae2d32008', \
'c6541da5-9858-4e3f-9b49-992e206d2c50', \
'78fdd6f8-c6a1-42df-a89f-cd65b7bb8be9', \
'3c950d36-c2b8-48d0-87d3-e0d6f570af62' \
) \
GROUP BY vehicle_id;" \
--repeat=20 \
--times
Result:
['vehicle_id', 'max']
['3c950d36-c2b8-48d0-87d3-e0d6f570af62', '2018-08-02 03:06:31.293184']
['78fdd6f8-c6a1-42df-a89f-cd65b7bb8be9', '2018-08-02 02:47:43.755989']
['35752c4c-b878-4436-8330-8d7246406a55', '2018-08-02 03:08:49.823209']
['0962cdca-9d85-457c-9616-cc2ae2d32008', '2018-08-02 03:01:25.414512']
['c6541da5-9858-4e3f-9b49-992e206d2c50', '2018-08-02 02:14:50.543760']
Times (milliseconds):
[828.5520076751709, 826.6720771789551, 837.0990753173828, 865.441083908081, 870.556116104126, 842.6721096038818, 859.3161106109619, 861.4299297332764, 866.6350841522217, 833.0469131469727, 838.021993637085, 841.0389423370361, 878.7519931793213, 879.6770572662354, 861.1328601837158, 855.1840782165527, 856.5502166748047, 882.9760551452637, 873.0340003967285, 858.4709167480469]
Average time (milliseconds):
855.812931061
This approach reduced the query time from 4356.70ms (query with subquery) to 1601.99ms (2 distinct queries).
Step 7. Test/tune write performance
- Bulk inserting into an existing table
- Minimizing unused indexes
- Retrieving the ID of a newly inserted row
Bulk inserting into an existing table
Moving on to writes, let's imagine that you have a batch of 100 new users to insert into the users
table. The most obvious approach is to insert each row using 100 separate INSERT
statements:
For the purpose of demonstration, the command below inserts the same user 100 times, each time with a different unique ID. Note also that you're now adding the --cumulative
flag to print the total time across all 100 inserts.
$ ./tuning.py \
--host=<address of any node> \
--statement="INSERT INTO users VALUES (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347')" \
--repeat=100 \
--times \
--cumulative
Times (milliseconds):
[33.28299522399902, 13.558149337768555, 14.67585563659668, 8.835077285766602, 9.104013442993164, 8.157968521118164, 10.174989700317383, 8.877992630004883, 9.196996688842773, 8.93402099609375, 9.894132614135742, 9.97304916381836, 8.221149444580078, 9.334087371826172, 9.270191192626953, 8.980035781860352, 7.210969924926758, 8.212089538574219, 8.048057556152344, 7.8639984130859375, 7.489204406738281, 9.547948837280273, 9.073972702026367, 9.660005569458008, 9.325981140136719, 9.338140487670898, 9.240865707397461, 7.958889007568359, 8.417844772338867, 8.075952529907227, 7.896184921264648, 9.118080139160156, 8.161067962646484, 9.071111679077148, 8.996963500976562, 7.790803909301758, 7.8220367431640625, 9.695053100585938, 9.470939636230469, 8.415937423706055, 9.287118911743164, 9.29117202758789, 9.618043899536133, 9.107828140258789, 8.491039276123047, 7.998943328857422, 9.282827377319336, 7.735013961791992, 9.161949157714844, 9.70005989074707, 8.910894393920898, 9.124994277954102, 9.028911590576172, 9.568929672241211, 10.931968688964844, 8.813858032226562, 14.040946960449219, 7.773876190185547, 9.801864624023438, 7.989168167114258, 8.188962936401367, 9.398937225341797, 9.705066680908203, 9.213924407958984, 9.569168090820312, 9.19198989868164, 9.664058685302734, 9.52601432800293, 8.01396369934082, 8.30698013305664, 8.03995132446289, 8.166074752807617, 9.335994720458984, 7.915019989013672, 9.584903717041016, 8.049964904785156, 7.803916931152344, 8.125066757202148, 9.367942810058594, 9.21487808227539, 9.630918502807617, 9.505033493041992, 9.830951690673828, 8.285045623779297, 8.095979690551758, 9.876012802124023, 8.067131042480469, 9.438037872314453, 8.147001266479492, 8.9111328125, 9.560108184814453, 8.78596305847168, 9.341955184936523, 10.293006896972656, 9.062051773071289, 14.008045196533203, 9.293079376220703, 9.57798957824707, 14.974832534790039, 8.59689712524414]
Average time (milliseconds):
9.41696166992
Cumulative time (milliseconds):
941.696166992
The 100 inserts took 941.69ms to complete, which isn't bad. However, it's significantly faster to use a single INSERT
statement with 100 comma-separated VALUES
clauses:
$ ./tuning.py \
--host=<address of any node> \
--statement="INSERT INTO users VALUES \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), \
(gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347'), (gen_random_uuid(), 'new york', 'Max Roach', '411 Drum Street', '173635282937347')" \
--repeat=1 \
--cumulative
Average time (milliseconds):
18.965959549
Cumulative time (milliseconds):
18.965959549
As you can see, this multi-row INSERT
technique reduced the total time for 100 inserts from 941.69ms to 18.96ms. It's useful to note that this technique is equally effective for UPSERT
and DELETE
statements as well.
Minimizing unused indexes
Earlier, we saw how important secondary indexes are for read performance. For writes, however, it's important to recognized the overhead that they create.
Let's consider the users
table:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW INDEXES FROM users;"
+-------+----------------+--------+-----+-------------+-----------+---------+----------+
| Table | Name | Unique | Seq | Column | Direction | Storing | Implicit |
+-------+----------------+--------+-----+-------------+-----------+---------+----------+
| users | primary | true | 1 | city | ASC | false | false |
| users | primary | true | 2 | id | ASC | false | false |
| users | users_name_idx | false | 1 | name | ASC | false | false |
| users | users_name_idx | false | 2 | credit_card | N/A | true | false |
| users | users_name_idx | false | 3 | city | ASC | false | true |
| users | users_name_idx | false | 4 | id | ASC | false | true |
+-------+----------------+--------+-----+-------------+-----------+---------+----------+
(6 rows)
This table has the primary index (the full table) and a secondary index on name
that is also storing credit_card
. This means that whenever a row is inserted, or whenever name
, credit_card
, city
, or id
are modified in existing rows, both indexes are updated.
To make this more concrete, let's count how many rows have a name that starts with C
and then update those rows to all have the same name:
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT count(*) \
FROM users \
WHERE name LIKE 'C%'" \
--repeat=1
Result:
['count']
['179']
Average time (milliseconds):
2.52413749695
$ ./tuning.py \
--host=<address of any node> \
--statement="UPDATE users \
SET name = 'Carl Kimball' \
WHERE name LIKE 'C%'" \
--repeat=1
Average time (milliseconds):
110.701799393
Because name
is in both the primary
and users_name_idx
indexes, for each of the 168 rows, 2 keys were updated.
Now, assuming that the users_name_idx
index is no longer needed, lets drop the index and execute an equivalent query:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="DROP INDEX users_name_idx;"
$ ./tuning.py \
--host=<address of any node> \
--statement="UPDATE users \
SET name = 'Peedie Hirata' \
WHERE name = 'Carl Kimball'" \
--repeat=1
Average time (milliseconds):
21.7709541321
Before, when both the primary and secondary indexes needed to be updated, the updates took 110.70ms. Now, after dropping the secondary index, an equivalent update took only 21.77ms.
Retrieving the ID of a newly inserted row
Now let's focus on the common case of inserting a row into a table and then retrieving the ID of the new row to do some follow-up work. One approach is to execute two statements, an INSERT
to insert the row and then a SELECT
to get the new ID:
$ ./tuning.py \
--host=<address of any node> \
--statement="INSERT INTO users VALUES (gen_random_uuid(), 'new york', 'Toni Brooks', '800 Camden Lane, Brooklyn, NY 11218', '98244843845134960')" \
--repeat=1
Average time (milliseconds):
9.97304916382
$ ./tuning.py \
--host=<address of any node> \
--statement="SELECT id FROM users WHERE name = 'Toni Brooks'" \
--repeat=1
Result:
['id']
['cc83e0bd-2ea0-4507-a683-a707cfbe0aba']
Average time (milliseconds):
7.32207298279
Combined, these statements are relatively fast, at 17.29ms, but an even more performant approach is to append RETURNING id
to the end of the INSERT
:
$ ./tuning.py \
--host=<address of any node> \
--statement="INSERT INTO users VALUES (gen_random_uuid(), 'new york', 'Brian Brooks', '800 Camden Lane, Brooklyn, NY 11218', '98244843845134960') \
RETURNING id" \
--repeat=1
Result:
['id']
['3d16500e-cb2e-462e-9c83-db0965d6deaf']
Average time (milliseconds):
9.48596000671
At just 9.48ms, this approach is faster due to the write and read executing in one instead of two client-server roundtrips. Note also that, as discussed earlier, if the leaseholder for the table happens to be on a different node than the query is running against, that introduces additional network hops and latency.
Multi-region deployment
Given that Movr is active on both US coasts, you'll now scale the cluster into two new regions, us-west1-a
and us-west2-a
, each with 3 nodes and an extra instance for simulating regional client traffic.
Step 8. Create more instances
Create 6 more instances, 3 in the
us-west1-a
zone (Oregon), and 3 in theus-west2-a
zone (Los Angeles). While creating each instance:- Use the
n1-standard-4
machine type (4 vCPUs, 15 GB memory). - Use the Ubuntu 16.04 OS image.
- Create and mount a local SSD.
- To apply the Web UI firewall rule you created earlier, click Management, disk, networking, SSH keys, select the Networking tab, and then enter
cockroachdb
in the Network tags field.
- Use the
Note the internal IP address of each
n1-standard-4
instance. You'll need these addresses when starting the CockroachDB nodes.Create an additional instance in the
us-west1-a
andus-west2-a
zones. These can be smaller, such asn1-standard-1
.
Step 9. Scale the cluster
SSH to one of the
n1-standard-4
instances in theus-west1-a
zone.Download the CockroachDB archive for Linux, extract the binary, and copy it into the
PATH
:$ curl https://binaries.cockroachdb.com/cockroach-v2.0.7.linux-amd64.tgz \ | tar -xz
$ sudo cp -i cockroach-v2.0.7.linux-amd64/cockroach /usr/local/bin/
Run the
cockroach start
command:$ cockroach start \ --insecure \ --advertise-host=<node internal address> \ --join=<same as earlier> \ --locality=cloud=gce,region=us-west1,zone=us-west1-a \ --cache=.25 \ --max-sql-memory=.25 \ --background
Repeat steps 1 - 3 for the other two
n1-standard-4
instances in theus-west1-a
zone.SSH to one of the
n1-standard-4
instances in theus-west2-a
zone.Download the CockroachDB archive for Linux, extract the binary, and copy it into the
PATH
:$ curl https://binaries.cockroachdb.com/cockroach-v2.0.7.linux-amd64.tgz \ | tar -xz
$ sudo cp -i cockroach-v2.0.7.linux-amd64/cockroach /usr/local/bin/
Run the
cockroach start
command:$ cockroach start \ --insecure \ --advertise-host=<node1 internal address> \ --join=<same as earlier> \ --locality=cloud=gce,region=us-west2,zone=us-west2-a \ --cache=.25 \ --max-sql-memory=.25 \ --background
Repeat steps 5 - 7 for the other two
n1-standard-4
instances in theus-west2-a
zone.
Step 10. Install the Python client
In each of the new zones, SSH to the instance not running a CockroachDB node, and install the Python client as described in step 5 above.
Step 11. Check rebalancing
Since you started each node with the --locality
flag set to its GCE zone, over the next minutes, CockroachDB will rebalance data evenly across the zones.
To check this, access the Web UI on any node at <node address>:8080
and look at the Node List. You'll see that the range count is more or less even across all nodes:
For reference, here's how the nodes map to zones:
Node IDs | Zone |
---|---|
1-3 | us-east1-b (South Carolina) |
4-6 | us-west1-a (Oregon) |
7-9 | us-west2-a (Los Angeles) |
To verify even balancing at range level, SSH to one of the instances not running CockroachDB and run the SHOW EXPERIMENTAL_RANGES
statement:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW EXPERIMENTAL_RANGES FROM TABLE vehicles;"
+-----------+---------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+-----------+---------+----------+----------+--------------+
| NULL | NULL | 22 | {1,6,9} | 6 |
+-----------+---------+----------+----------+--------------+
(1 row)
In this case, we can see that, for the single range containing vehicles
data, one replica is in each zone, and the leaseholder is in the us-west1-a
zone.
Step 12. Test performance
In general, all of the tuning techniques featured in the single-region scenario above still apply in a multi-region deployment. However, the fact that data and leaseholders are spread across the US means greater latencies in many cases.
Reads
For example, imagine we are a Movr administrator in New York, and we want to get the IDs and descriptions of all New York-based bikes that are currently in use:
SSH to the instance in
us-east1-b
with the Python client.Query for the data:
$ ./tuning.py \ --host=<address of a node in us-east1-b> \ --statement="SELECT id, ext FROM vehicles \ WHERE city = 'new york' \ AND type = 'bike' \ AND status = 'in_use'" \ --repeat=50 \ --times
Result: ['id', 'ext'] ['0068ee24-2dfb-437d-9a5d-22bb742d519e', "{u'color': u'green', u'brand': u'Kona'}"] ['01b80764-283b-4232-8961-a8d6a4121a08', "{u'color': u'green', u'brand': u'Pinarello'}"] ['02a39628-a911-4450-b8c0-237865546f7f', "{u'color': u'black', u'brand': u'Schwinn'}"] ['02eb2a12-f465-4575-85f8-a4b77be14c54', "{u'color': u'black', u'brand': u'Pinarello'}"] ['02f2fcc3-fea6-4849-a3a0-dc60480fa6c2', "{u'color': u'red', u'brand': u'FujiCervelo'}"] ['034d42cf-741f-428c-bbbb-e31820c68588', "{u'color': u'yellow', u'brand': u'Santa Cruz'}"] ... Times (milliseconds): [1123.0790615081787, 190.16599655151367, 127.28595733642578, 72.94511795043945, 72.0360279083252, 70.50704956054688, 70.83487510681152, 73.11201095581055, 72.81899452209473, 71.35510444641113, 71.6249942779541, 70.8611011505127, 72.17597961425781, 71.78997993469238, 70.75691223144531, 76.08985900878906, 72.6480484008789, 71.91896438598633, 70.59216499328613, 71.07686996459961, 71.86722755432129, 71.01583480834961, 71.29812240600586, 71.74086570739746, 72.67093658447266, 71.03395462036133, 71.78306579589844, 71.5029239654541, 70.33801078796387, 72.91483879089355, 71.23708724975586, 72.81684875488281, 71.70701026916504, 71.32506370544434, 71.68197631835938, 70.78695297241211, 72.80707359313965, 73.0600357055664, 71.69818878173828, 71.40707969665527, 70.53804397583008, 71.83694839477539, 70.08099555969238, 71.96617126464844, 71.03586196899414, 72.6020336151123, 71.23398780822754, 71.03800773620605, 72.12519645690918, 71.77996635437012] Average time (milliseconds): 96.2521076202
As we saw earlier, the leaseholder for the vehicles
table is in us-west1-a
(Oregon), so our query had to go from the gateway node in us-east1-b
all the way to the west coast and then back again before returning data to the client.
For contrast, imagine we are now a Movr administrator in Seattle, and we want to get the IDs and descriptions of all Seattle-based bikes that are currently in use:
SSH to the instance in
us-west1-a
with the Python client.Query for the data:
$ ./tuning.py \ --host=<address of a node in us-west1-a> \ --statement="SELECT id, ext FROM vehicles \ WHERE city = 'seattle' \ AND type = 'bike' \ AND status = 'in_use'" \ --repeat=50 \ --times
Result: ['id', 'ext'] ['00078349-94d4-43e6-92be-8b0d1ac7ee9f', "{u'color': u'blue', u'brand': u'Merida'}"] ['003f84c4-fa14-47b2-92d4-35a3dddd2d75', "{u'color': u'red', u'brand': u'Kona'}"] ['0107a133-7762-4392-b1d9-496eb30ee5f9', "{u'color': u'yellow', u'brand': u'Kona'}"] ['0144498b-4c4f-4036-8465-93a6bea502a3', "{u'color': u'blue', u'brand': u'Pinarello'}"] ['01476004-fb10-4201-9e56-aadeb427f98a', "{u'color': u'black', u'brand': u'Merida'}"] Times (milliseconds): [83.34112167358398, 35.54201126098633, 36.23318672180176, 35.546064376831055, 39.82996940612793, 35.067081451416016, 35.12001037597656, 34.34896469116211, 35.05301475524902, 35.52699089050293, 34.442901611328125, 33.95986557006836, 35.25996208190918, 35.26592254638672, 35.75301170349121, 35.50601005554199, 35.93301773071289, 32.97090530395508, 35.09712219238281, 35.33005714416504, 34.66916084289551, 34.97791290283203, 34.68203544616699, 34.09695625305176, 35.676002502441406, 33.01596641540527, 35.39609909057617, 33.804893493652344, 33.6918830871582, 34.37995910644531, 33.71405601501465, 35.18819808959961, 34.35802459716797, 34.191131591796875, 33.44106674194336, 34.84678268432617, 35.51292419433594, 33.80894660949707, 33.6911678314209, 36.14497184753418, 34.671783447265625, 35.28904914855957, 33.84900093078613, 36.21387481689453, 36.26894950866699, 34.7599983215332, 34.73687171936035, 34.715890884399414, 35.101890563964844, 35.4609489440918] Average time (milliseconds): 35.9096717834
Because the leaseholder for vehicles
is in the same zone as the client request, this query took just 35.90ms compared to the similar query in New York that took 96.25ms.
Writes
The geographic distribution of data impacts write performance as well. For example, imagine 100 people in New York and 100 people in Los Angeles want to create new Movr accounts:
SSH to the instance in
us-east1-b
with the Python client.Create 100 NY-based users:
./tuning.py \ --host=<address of a node in us-east1-b> \ --statement="INSERT INTO users VALUES (gen_random_uuid(), 'new york', 'New Yorker', '111 East Street', '1736352379937347')" \ --repeat=100 \ --times
Times (milliseconds): [710.5610370635986, 75.03294944763184, 76.18403434753418, 76.6599178314209, 75.54292678833008, 77.10099220275879, 76.49803161621094, 76.12395286560059, 75.13093948364258, 76.4460563659668, 74.74899291992188, 76.11799240112305, 74.95307922363281, 75.22797584533691, 75.01792907714844, 76.11393928527832, 75.35195350646973, 76.23100280761719, 75.17099380493164, 76.05600357055664, 76.4470100402832, 76.4310359954834, 75.02388954162598, 76.38192176818848, 78.89008522033691, 76.27677917480469, 75.12402534484863, 74.9521255493164, 75.08397102355957, 76.21502876281738, 75.15192031860352, 77.74996757507324, 73.84800910949707, 85.68978309631348, 75.08993148803711, 77.28886604309082, 76.8439769744873, 76.6448974609375, 75.1500129699707, 76.38287544250488, 75.12092590332031, 76.92408561706543, 76.86591148376465, 76.45702362060547, 76.61795616149902, 75.77109336853027, 81.47501945495605, 83.72306823730469, 76.41983032226562, 75.19102096557617, 74.01609420776367, 77.21996307373047, 76.61914825439453, 75.56986808776855, 76.94005966186523, 75.74892044067383, 76.63488388061523, 76.73311233520508, 75.73890686035156, 75.3028392791748, 76.58910751342773, 76.70807838439941, 76.36213302612305, 75.05607604980469, 76.99084281921387, 79.19192314147949, 75.69003105163574, 76.53594017028809, 75.3641128540039, 76.4620304107666, 75.81305503845215, 76.84993743896484, 75.74915885925293, 77.1799087524414, 76.67183876037598, 75.85597038269043, 77.18396186828613, 78.25303077697754, 76.66516304016113, 75.4399299621582, 76.98297500610352, 75.69122314453125, 77.4688720703125, 81.50601387023926, 76.74908638000488, 76.9951343536377, 75.34193992614746, 76.82991027832031, 76.4460563659668, 75.76298713684082, 76.63083076477051, 75.43802261352539, 76.47705078125, 78.95708084106445, 75.60205459594727, 75.70815086364746, 76.48301124572754, 76.65586471557617, 75.71196556091309, 74.09906387329102] Average time (milliseconds): 82.7817606926
SSH to the instance in
us-west2-a
with the Python client.Create 100 new Los Angeles-based users:
./tuning.py \ --host=<address of a node in us-west2-a> \ --statement="INSERT INTO users VALUES (gen_random_uuid(), 'los angeles', 'Los Angel', '111 West Street', '9822222379937347')" \ --repeat=100 \ --times
Times (milliseconds): [213.47904205322266, 140.0778293609619, 138.11588287353516, 138.22197914123535, 143.43595504760742, 139.0368938446045, 138.3199691772461, 138.7031078338623, 139.38307762145996, 139.53304290771484, 138.78607749938965, 140.59996604919434, 138.1399631500244, 138.94009590148926, 138.17405700683594, 137.9709243774414, 138.02003860473633, 137.82405853271484, 140.13099670410156, 139.08815383911133, 138.0600929260254, 139.01615142822266, 138.05103302001953, 137.76111602783203, 139.38617706298828, 137.42399215698242, 137.89701461791992, 138.40818405151367, 138.6868953704834, 139.13893699645996, 139.24717903137207, 138.7009620666504, 137.4349594116211, 137.24017143249512, 138.99493217468262, 138.77201080322266, 138.624906539917, 139.19997215270996, 139.4331455230713, 143.18394660949707, 138.0319595336914, 137.6488208770752, 137.27498054504395, 136.3968849182129, 139.0249729156494, 137.9079818725586, 139.37997817993164, 139.32204246520996, 140.045166015625, 137.9718780517578, 139.36805725097656, 139.6927833557129, 139.63794708251953, 138.016939163208, 145.32899856567383, 138.261079788208, 139.56904411315918, 139.6658420562744, 138.02599906921387, 139.7988796234131, 138.24796676635742, 139.9519443511963, 136.5041732788086, 139.43004608154297, 138.16499710083008, 138.2119655609131, 139.69111442565918, 140.30194282531738, 138.14496994018555, 140.00296592712402, 139.44697380065918, 139.35494422912598, 137.9709243774414, 140.78497886657715, 136.4901065826416, 138.44680786132812, 138.69094848632812, 139.2819881439209, 140.45214653015137, 138.3049488067627, 139.4188404083252, 139.9250030517578, 140.40303230285645, 138.7009620666504, 136.9321346282959, 139.20903205871582, 138.14496994018555, 140.14315605163574, 139.30511474609375, 139.58096504211426, 141.16501808166504, 138.66591453552246, 138.3810043334961, 137.39800453186035, 139.9540901184082, 138.4589672088623, 138.72814178466797, 138.3681297302246, 139.1599178314209, 139.29295539855957] Average time (milliseconds): 139.702253342
On average, it took 82.78ms to create a user in New York and 139.70ms to create a user in Los Angeles. To better understand this discrepancy, let's look at the distribution of data for the users
table:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SHOW EXPERIMENTAL_RANGES FROM TABLE users;"
+-----------+---------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+-----------+---------+----------+----------+--------------+
| NULL | NULL | 51 | {2,6,7} | 2 |
+-----------+---------+----------+----------+--------------+
(1 row)
For the single range containing users
data, one replica is in each zone, with the leaseholder in the us-east1-b
zone. This means that:
- When creating a user in New York, the request doesn't have to leave the zone to reach the leaseholder. However, since a write requires consensus from its replica group, the write has to wait for confirmation from either the replica in
us-west1-a
(Oregon) orus-west2-a
(Los Angeles) before committing and then returning confirmation to the client. - When creating a user in Los Angeles, there are more network hops and, thus, increased latency. The request first needs to travel across the continent to the leaseholder in
us-east1-b
. It then has to wait for confirmation from either the replica inus-west1-a
(Oregon) orus-west2-a
(Los Angeles) before committing and then returning confirmation to the client back in the west.
Step 13. Partition data by city
For this service, the most effective technique for improving read and write latency is to geo-partition the data by city. In essence, this means changing the way data is mapped to ranges. Instead of an entire table and its indexes mapping to a specific range or set of ranges, all rows in the table and its indexes with a given city will map to a range or set of ranges. Once ranges are defined in this way, we can then use the replication zone feature to pin partitions to specific locations, ensuring that read and write requests from users in a specific city do not have to leave that region.
Partitioning is an enterprise feature, so start off by registering for a 30-day trial license.
Once you've received the trial license, SSH to any node in your cluster and apply the license:
$ cockroach sql \ --insecure \ --host=<address of any node> \ --execute="SET CLUSTER SETTING cluster.organization = '<your org name>';"
$ cockroach sql \ --insecure \ --host=<address of any node> \ --execute="SET CLUSTER SETTING enterprise.license = '<your license>';"
Define partitions for all tables and their secondary indexes.
Start with the
users
table:$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER TABLE users \ PARTITION BY LIST (city) ( \ PARTITION new_york VALUES IN ('new york'), \ PARTITION boston VALUES IN ('boston'), \ PARTITION washington_dc VALUES IN ('washington dc'), \ PARTITION seattle VALUES IN ('seattle'), \ PARTITION san_francisco VALUES IN ('san francisco'), \ PARTITION los_angeles VALUES IN ('los angeles') \ );"
Now define partitions for the
vehicles
table and its secondary indexes:$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER TABLE vehicles \ PARTITION BY LIST (city) ( \ PARTITION new_york VALUES IN ('new york'), \ PARTITION boston VALUES IN ('boston'), \ PARTITION washington_dc VALUES IN ('washington dc'), \ PARTITION seattle VALUES IN ('seattle'), \ PARTITION san_francisco VALUES IN ('san francisco'), \ PARTITION los_angeles VALUES IN ('los angeles') \ );"
$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER INDEX vehicles_auto_index_fk_city_ref_users \ PARTITION BY LIST (city) ( \ PARTITION new_york_idx VALUES IN ('new york'), \ PARTITION boston_idx VALUES IN ('boston'), \ PARTITION washington_dc_idx VALUES IN ('washington dc'), \ PARTITION seattle_idx VALUES IN ('seattle'), \ PARTITION san_francisco_idx VALUES IN ('san francisco'), \ PARTITION los_angeles_idx VALUES IN ('los angeles') \ );"
Next, define partitions for the
rides
table and its secondary indexes:$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER TABLE rides \ PARTITION BY LIST (city) ( \ PARTITION new_york VALUES IN ('new york'), \ PARTITION boston VALUES IN ('boston'), \ PARTITION washington_dc VALUES IN ('washington dc'), \ PARTITION seattle VALUES IN ('seattle'), \ PARTITION san_francisco VALUES IN ('san francisco'), \ PARTITION los_angeles VALUES IN ('los angeles') \ );"
$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER INDEX rides_auto_index_fk_city_ref_users \ PARTITION BY LIST (city) ( \ PARTITION new_york_idx1 VALUES IN ('new york'), \ PARTITION boston_idx1 VALUES IN ('boston'), \ PARTITION washington_dc_idx1 VALUES IN ('washington dc'), \ PARTITION seattle_idx1 VALUES IN ('seattle'), \ PARTITION san_francisco_idx1 VALUES IN ('san francisco'), \ PARTITION los_angeles_idx1 VALUES IN ('los angeles') \ );"
$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="ALTER INDEX rides_auto_index_fk_vehicle_city_ref_vehicles \ PARTITION BY LIST (vehicle_city) ( \ PARTITION new_york_idx2 VALUES IN ('new york'), \ PARTITION boston_idx2 VALUES IN ('boston'), \ PARTITION washington_dc_idx2 VALUES IN ('washington dc'), \ PARTITION seattle_idx2 VALUES IN ('seattle'), \ PARTITION san_francisco_idx2 VALUES IN ('san francisco'), \ PARTITION los_angeles_idx2 VALUES IN ('los angeles') \ );"
Finally, drop an unused index on
rides
rather than partition it:$ cockroach sql \ --insecure \ --database=movr \ --host=<address of any node> \ --execute="DROP INDEX rides_start_time_idx;"
Note:The
rides
table contains 1 million rows, so dropping this index will take a few minutes.Now create replication zones to require city data to be stored on specific nodes based on node locality.
City Locality New York zone=us-east1-b
Boston zone=us-east1-b
Washington DC zone=us-east1-b
Seattle zone=us-west1-a
San Francisco zone=us-west2-a
Los Angeles zone=us-west2-a
Note:Since our nodes are located in 3 specific GCE zones, we're only going to use the
zone=
portion of node locality. If we were using multiple zones per regions, we would likely use theregion=
portion of the node locality instead.Start with the
users
table partitions:$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.users.new_york \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.users.boston \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.users.washington_dc \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.users.seattle \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.users.san_francisco \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.users.los_angeles \ --insecure \ --host=<address of any node> \ -f -
Move on to the
vehicles
table and secondary index partitions:$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.new_york \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.new_york_idx \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.boston \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.boston_idx \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.washington_dc \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.vehicles.washington_dc_idx \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.vehicles.seattle \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.vehicles.seattle_idx \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.vehicles.san_francisco \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.vehicles.san_francisco_idx \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.vehicles.los_angeles \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.vehicles.los_angeles_idx \ --insecure \ --host=<address of any node> \ -f -
Finish with the
rides
table and secondary index partitions:$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.new_york \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.new_york_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.new_york_idx2 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.boston \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.boston_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.boston_idx2 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.washington_dc \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.washington_dc_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-east1-b]' | \ cockroach zone set movr.rides.washington_dc_idx2 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.rides.seattle \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.rides.seattle_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west1-a]' | \ cockroach zone set movr.rides.seattle_idx2 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.san_francisco \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.san_francisco_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.san_francisco_idx2 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.los_angeles \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.los_angeles_idx1 \ --insecure \ --host=<address of any node> \ -f -
$ echo 'constraints: [+zone=us-west2-a]' | \ cockroach zone set movr.rides.los_angeles_idx2 \ --insecure \ --host=<address of any node> \ -f -
Step 14. Check rebalancing after partitioning
Over the next minutes, CockroachDB will rebalance all partitions based on the constraints you defined.
To check this at a high level, access the Web UI on any node at <node address>:8080
and look at the Node List. You'll see that the range count is still close to even across all nodes but much higher than before partitioning:
To check at a more granular level, SSH to one of the instances not running CockroachDB and run the SHOW EXPERIMENTAL_RANGES
statement on the vehicles
table:
$ cockroach sql \
--insecure \
--host=<address of any node> \
--database=movr \
--execute="SELECT * FROM \
[SHOW EXPERIMENTAL_RANGES FROM TABLE vehicles] \
WHERE \"Start Key\" IS NOT NULL \
AND \"Start Key\" NOT LIKE '%Prefix%';"
+------------------+----------------------------+----------+----------+--------------+
| Start Key | End Key | Range ID | Replicas | Lease Holder |
+------------------+----------------------------+----------+----------+--------------+
| /"boston" | /"boston"/PrefixEnd | 95 | {1,2,3} | 2 |
| /"los angeles" | /"los angeles"/PrefixEnd | 111 | {7,8,9} | 9 |
| /"new york" | /"new york"/PrefixEnd | 91 | {1,2,3} | 1 |
| /"san francisco" | /"san francisco"/PrefixEnd | 107 | {7,8,9} | 7 |
| /"seattle" | /"seattle"/PrefixEnd | 103 | {4,5,6} | 4 |
| /"washington dc" | /"washington dc"/PrefixEnd | 99 | {1,2,3} | 1 |
+------------------+----------------------------+----------+----------+--------------+
(6 rows)
For reference, here's how the nodes map to zones:
Node IDs | Zone |
---|---|
1-3 | us-east1-b (South Carolina) |
4-6 | us-west1-a (Oregon) |
7-9 | us-west2-a (Los Angeles) |
We can see that, after partitioning, the replicas for New York, Boston, and Washington DC are located on nodes 1-3 in us-east1-b
, replicas for Seattle are located on nodes 4-6 in us-west1-a
, and replicas for San Francisco and Los Angeles are located on nodes 7-9 in us-west2-a
.
Step 15. Test performance after partitioning
After partitioning, reads and writers for a specific city will be much faster because all replicas for that city are now located on the nodes closest to the city.
To check this, let's repeat a few of the read and write queries that we executed before partitioning in step 12.
Reads
Again imagine we are a Movr administrator in New York, and we want to get the IDs and descriptions of all New York-based bikes that are currently in use:
SSH to the instance in
us-east1-b
with the Python client.Query for the data:
$ ./tuning.py \ --host=<address of a node in us-east1-b> \ --statement="SELECT id, ext FROM vehicles \ WHERE city = 'new york' \ AND type = 'bike' \ AND status = 'in_use'" \ --repeat=50 \ --times
Result: ['id', 'ext'] ['0068ee24-2dfb-437d-9a5d-22bb742d519e', "{u'color': u'green', u'brand': u'Kona'}"] ['01b80764-283b-4232-8961-a8d6a4121a08', "{u'color': u'green', u'brand': u'Pinarello'}"] ['02a39628-a911-4450-b8c0-237865546f7f', "{u'color': u'black', u'brand': u'Schwinn'}"] ['02eb2a12-f465-4575-85f8-a4b77be14c54', "{u'color': u'black', u'brand': u'Pinarello'}"] ['02f2fcc3-fea6-4849-a3a0-dc60480fa6c2', "{u'color': u'red', u'brand': u'FujiCervelo'}"] ['034d42cf-741f-428c-bbbb-e31820c68588', "{u'color': u'yellow', u'brand': u'Santa Cruz'}"] ... Times (milliseconds): [17.27890968322754, 9.554147720336914, 7.483959197998047, 7.407903671264648, 7.538795471191406, 7.39288330078125, 7.623910903930664, 7.172822952270508, 7.15184211730957, 7.201910018920898, 7.063865661621094, 7.602930068969727, 7.246971130371094, 6.966829299926758, 7.369041442871094, 7.277965545654297, 7.650852203369141, 7.177829742431641, 7.266998291015625, 7.150173187255859, 7.303953170776367, 7.1048736572265625, 7.218122482299805, 7.168054580688477, 7.258176803588867, 7.375955581665039, 7.013797760009766, 7.2078704833984375, 7.277965545654297, 7.352113723754883, 7.0400238037109375, 7.379055023193359, 7.227897644042969, 7.266044616699219, 6.883859634399414, 7.344961166381836, 7.222175598144531, 7.149934768676758, 7.241010665893555, 6.999969482421875, 7.40504264831543, 7.191896438598633, 7.192134857177734, 7.2231292724609375, 7.10296630859375, 7.291078567504883, 6.976127624511719, 7.338047027587891, 6.918191909790039, 7.070064544677734] Average time (milliseconds): 7.48650074005
Before partitioning, this query took 96.25ms on average. After partitioning, the query took only 7.48ms on average.
Writes
Now let's again imagine 100 people in New York and 100 people in Los Angeles want to create new Movr accounts:
SSH to the instance in
us-east1-b
with the Python client.Create 100 NY-based users:
./tuning.py \ --host=<address of a node in us-east1-b> \ --statement="INSERT INTO users VALUES (gen_random_uuid(), 'new york', 'New Yorker', '111 East Street', '1736352379937347')" \ --repeat=100 \ --times
Times (milliseconds): [9.378910064697266, 7.173061370849609, 9.769916534423828, 8.235931396484375, 9.124040603637695, 9.358882904052734, 8.581161499023438, 7.482051849365234, 8.441925048828125, 8.306026458740234, 8.775949478149414, 8.685827255249023, 6.851911544799805, 9.104013442993164, 9.664058685302734, 7.126092910766602, 8.738994598388672, 8.75997543334961, 9.040117263793945, 8.374929428100586, 8.384943008422852, 10.58506965637207, 8.538961410522461, 7.405996322631836, 9.508132934570312, 8.268117904663086, 11.46697998046875, 9.343147277832031, 8.31294059753418, 7.085084915161133, 8.779048919677734, 7.356166839599609, 8.732080459594727, 9.31406021118164, 8.460044860839844, 8.933067321777344, 8.610963821411133, 7.01904296875, 9.474039077758789, 8.276939392089844, 9.40704345703125, 9.205818176269531, 8.270025253295898, 7.443904876708984, 8.999824523925781, 8.215904235839844, 8.124828338623047, 8.324861526489258, 8.156061172485352, 8.740901947021484, 8.39996337890625, 7.437944412231445, 8.78000259399414, 8.615970611572266, 8.795022964477539, 8.683919906616211, 7.111072540283203, 7.770061492919922, 8.922100067138672, 9.526968002319336, 7.8411102294921875, 8.287191390991211, 10.084152221679688, 8.744001388549805, 8.032083511352539, 7.095098495483398, 8.343935012817383, 8.038997650146484, 8.939027786254883, 8.714914321899414, 6.999969482421875, 7.087945938110352, 9.23299789428711, 8.90803337097168, 7.808923721313477, 8.558034896850586, 7.122993469238281, 8.755922317504883, 8.379936218261719, 8.464813232421875, 8.405923843383789, 7.163047790527344, 9.139060974121094, 8.706092834472656, 7.130146026611328, 12.811899185180664, 9.733915328979492, 7.981061935424805, 9.001016616821289, 8.28409194946289, 7.188081741333008, 9.055137634277344, 9.569883346557617, 7.223844528198242, 8.78596305847168, 6.941080093383789, 8.934974670410156, 8.980989456176758, 7.564067840576172, 9.202003479003906] Average time (milliseconds): 8.51003170013
Before partitioning, this query took 82.78ms on average. After partitioning, the query took only 8.51ms on average.
SSH to the instance in
us-west2-a
with the Python client.Create 100 new Los Angeles-based users:
./tuning.py \ --host=<address of a node in us-west2-a> \ --statement="INSERT INTO users VALUES (gen_random_uuid(), 'los angeles', 'Los Angel', '111 West Street', '9822222379937347')" \ --repeat=100 \ --times
Times (milliseconds): [20.322084426879883, 14.09602165222168, 14.353036880493164, 25.568008422851562, 15.157938003540039, 27.19593048095703, 29.092073440551758, 14.515876770019531, 14.114141464233398, 19.414901733398438, 15.073060989379883, 13.965845108032227, 13.913869857788086, 15.218019485473633, 13.844013214111328, 14.110088348388672, 13.943910598754883, 13.73600959777832, 13.262033462524414, 14.648914337158203, 14.066219329833984, 13.91911506652832, 14.122962951660156, 14.724016189575195, 17.747879028320312, 16.537904739379883, 13.921022415161133, 14.027118682861328, 15.810012817382812, 14.811992645263672, 14.551877975463867, 14.912128448486328, 14.078140258789062, 14.576196670532227, 19.381046295166016, 14.536857604980469, 14.664888381958008, 14.539957046508789, 15.054941177368164, 17.20881462097168, 14.64700698852539, 14.211177825927734, 15.089988708496094, 14.193058013916016, 14.544010162353516, 14.680862426757812, 14.32490348815918, 15.841007232666016, 14.069080352783203, 14.59503173828125, 14.837026596069336, 14.315128326416016, 14.558792114257812, 14.645099639892578, 14.82701301574707, 14.699935913085938, 15.035152435302734, 14.724016189575195, 16.10708236694336, 14.612913131713867, 14.641046524047852, 14.706850051879883, 14.29295539855957, 14.779090881347656, 15.485048294067383, 17.444133758544922, 15.172004699707031, 20.865917205810547, 14.388084411621094, 14.241218566894531, 14.343976974487305, 14.602899551391602, 14.64390754699707, 13.908147811889648, 20.69687843322754, 15.130043029785156, 14.754056930541992, 14.123916625976562, 14.760017395019531, 14.25480842590332, 14.446020126342773, 14.229059219360352, 15.10000228881836, 14.275789260864258, 14.42098617553711, 14.935970306396484, 15.175819396972656, 27.69613265991211, 14.856815338134766, 14.902830123901367, 15.029191970825195, 15.143871307373047, 15.524148941040039, 14.510869979858398, 18.740177154541016, 14.97197151184082, 15.30003547668457, 15.158891677856445, 14.423847198486328, 35.25400161743164] Average time (milliseconds): 15.7462859154
Before partitioning, this query took 139.70ms on average. After partitioning, the query took only 15.74ms on average.