Methods to Do Load Testing with Rockset

What’s load testing and why does it matter?

Load testing is a essential course of for any database or information service, together with Rockset. By doing load testing, we purpose to evaluate the system’s habits underneath each regular and peak circumstances. This course of helps in evaluating essential metrics like Queries Per Second (QPS), concurrency, and question latency. Understanding these metrics is crucial for sizing your compute sources accurately, and making certain that they will deal with the anticipated load. This, in flip, helps in reaching Service Degree Agreements (SLAs) and ensures a easy, uninterrupted person expertise. That is particularly essential for customer-facing use instances, the place finish customers anticipate a handy guide a rough person expertise. Load testing is typically additionally referred to as efficiency or stress testing.

“53% of visits are more likely to be deserted if pages take longer than 3 seconds to load” — Google

Rockset compute sources (referred to as digital situations or VIs) come in numerous sizes, starting from Small to 16XL, and every dimension has a predefined variety of vCPUs and reminiscence out there. Selecting an applicable dimension is dependent upon your question complexity, dataset dimension and selectivity of your queries, variety of queries which are anticipated to run concurrently and goal question efficiency latency. Moreover, in case your VI can also be used for ingestion, it’s best to think about sources wanted to deal with ingestion and indexing in parallel to question execution. Fortunately, we provide two options that may assist with this:

• Auto-scaling – with this characteristic, Rockset will mechanically scale the VI up and down relying on the present load. That is essential in case you have some variability in your load and/or use your VI to do each ingestion and querying.
• Compute-compute separation – that is helpful as a result of you’ll be able to create VIs which are devoted solely for operating queries and this ensures that the entire out there sources are geared in the direction of executing these queries effectively. This implies you’ll be able to isolate queries from ingest or isolate totally different apps on totally different VIs to make sure scalability and efficiency.

We suggest doing load testing on at the very least two digital situations – with ingestion operating on the primary VI and on a separate question VI. This helps with deciding on a single or multi-VI structure.

Load testing helps us establish the boundaries of the chosen VI for our explicit use case and helps us decide an applicable VI dimension to deal with our desired load.

Instruments for load testing

In terms of load testing instruments, just a few common choices are JMeter, k6, Gatling and Locust. Every of those instruments has its strengths and weaknesses:

• JMeter: A flexible and user-friendly software with a GUI, excellent for numerous varieties of load testing, however could be resource-intensive.
• k6: Optimized for prime efficiency and cloud environments, utilizing JavaScript for scripting, appropriate for builders and CI/CD workflows.
• Gatling: Excessive-performance software utilizing Scala, finest for complicated, superior scripting eventualities.
• Locust: Python-based, providing simplicity and fast script growth, nice for easy testing wants.

Every software presents a novel set of options, and the selection is dependent upon the particular necessities of the load take a look at being performed. Whichever software you utilize, remember to learn via the documentation and perceive the way it works and the way it measures the latencies/response occasions. One other good tip is to not combine and match instruments in your testing – if you’re load testing a use case with JMeter, keep it up to get reproducible and reliable outcomes you can share along with your group or stakeholders.

Rockset has a REST API that can be utilized to execute queries, and all instruments listed above can be utilized to load take a look at REST API endpoints. For this weblog, I’ll concentrate on load testing Rockset with Locust, however I’ll present some helpful sources for JMeter, k6 and Gatling as nicely.

Establishing Rockset and Locust for load testing

Let’s say we now have a pattern SQL question that we wish to take a look at and our information is ingested into Rockset. The very first thing we often do is convert that question right into a Question Lambda – this makes it very straightforward to check that SQL question as a REST endpoint. It may be parametrized and the SQL could be versioned and stored in a single place, as an alternative of going forwards and backwards and altering your load testing scripts each time you’ll want to change one thing within the question.

Step 1 – Determine the question you wish to load take a look at

In our situation, we wish to discover the preferred product on our webshop for a selected day. That is what our SQL question appears to be like like (be aware that :date is a parameter which we will provide when executing the question):

--top product for a selected day
SELECT
    s.Date,
    MAX_BY(p.ProductName, s.Depend) AS ProductName,
    MAX(s.Depend) AS NumberOfClicks
FROM
    "Demo-Ecommerce".ProductStatsAlias s
    INNER JOIN "Demo-Ecommerce".ProductsAlias p ON s.ProductID = CAST(p._id AS INT)
WHERE
    s.Date = :date
GROUP BY
    1
ORDER BY
    1 DESC;

Step 2 – Save your question as a Question Lambda

We’ll save this question as a question lambda referred to as LoadTestQueryLambda which is able to then be out there as a REST endpoint:

https://api.usw2a1.rockset.com/v1/orgs/self/ws/sandbox/lambdas/LoadTestQueryLambda/tags/newest

curl --request POST 
--url https://api.usw2a1.rockset.com/v1/orgs/self/ws/sandbox/lambdas/LoadTestQueryLambda/tags/newest 
-H "Authorization: ApiKey $ROCKSET_APIKEY" 
-H 'Content material-Kind: software/json' 
  -d '{
    "parameters": [
      {
        "name": "days",
        "type": "int",
        "value": "1"
      }
    ],
      "virtual_instance_id": ""
  }' 
 | python -m json.software

Step 3 – Generate your API key

Now we have to generate an API key, which we’ll use as a means for our Locust script to authenticate itself to Rockset and run the take a look at. You possibly can create an API key simply via our console or via the API.

Step 4 – Create a digital occasion for load testing

Subsequent, we want the ID of the digital occasion we wish to load take a look at. In our situation, we wish to run a load take a look at towards a Rockset digital occasion that’s devoted solely to querying. We spin up an extra Medium digital occasion for this:

As soon as the VI is created, we will get its ID from the console:

Step 5 – Set up Locust

Subsequent, we’ll set up and arrange Locust. You are able to do this in your native machine or a devoted occasion (suppose EC2 in AWS).

$ pip set up locust

Step 6 – Create your Locust take a look at script

As soon as that’s achieved, we’ll create a Python script for the Locust load take a look at (be aware that it expects a ROCKSET_APIKEY setting variable to be set which is our API key from step 3).

We will use the script under as a template:

import os
from locust import HttpUser, job, tag
from random import randrange

class query_runner(HttpUser):
    ROCKSET_APIKEY = os.getenv('ROCKSET_APIKEY') # API secret is an setting variable

    header = {"authorization": "ApiKey " + ROCKSET_APIKEY}

    def on_start(self):
        self.headers = {
            "Authorization": "ApiKey " + self.ROCKSET_APIKEY,
            "Content material-Kind": "software/json"
        }
        self.shopper.headers = self.headers
        self.host="https://api.usw2a1.rockset.com/v1/orgs/self" # substitute this along with your area's URI
        self.shopper.base_url = self.host
        self.vi_id = '' # substitute this along with your VI ID

    @tag('LoadTestQueryLambda')
    @job(1)
    def LoadTestQueryLambda(self):
        # utilizing default params for now
        information = {
            "virtual_instance_id": self.vi_id
        }
        target_service="/ws/sandbox/lambdas/LoadTestQueryLambda/tags/newest" # substitute this along with your question lambda
        consequence = self.shopper.submit(
            target_service,
            json=information
        )

Step 7 – Run the load take a look at

As soon as we set the API key setting variable, we will run the Locust setting:

export ROCKSET_APIKEY=
locust -f my_locust_load_test.py --host https://api.usw2a1.rockset.com/v1/orgs/self

And navigate to: http://localhost:8089 the place we will begin our Locust load take a look at:

Let’s discover what occurs as soon as we hit the Begin swarming button:

1. Initialization of simulated customers: Locust begins creating digital customers (as much as the quantity you specified) on the price you outlined (the spawn price). These customers are situations of the person class outlined in your Locust script. In our case, we’re beginning with a single person however we’ll then manually improve it to five and 10 customers, after which go down to five and 1 once more.
2. Process execution: Every digital person begins executing the duties outlined within the script. In Locust, duties are sometimes HTTP requests, however they are often any Python code. The duties are picked randomly or primarily based on the weights assigned to them (if any). Now we have only one question that we’re executing (our LoadTestQueryLambda).
3. Efficiency metrics assortment: Because the digital customers carry out duties, Locust collects and calculates efficiency metrics. These metrics embrace the variety of requests made, the variety of requests per second, response occasions, and the variety of failures.
4. Actual-time statistics replace: The Locust internet interface updates in real-time, exhibiting these statistics. This consists of the variety of customers at the moment swarming, the request price, failure price, and response occasions.
5. Take a look at scalability: Locust will proceed to spawn customers till it reaches the whole quantity specified. It ensures the load is elevated regularly as per the required spawn price, permitting you to look at how the system efficiency modifications because the load will increase. You possibly can see this within the graph under the place the variety of customers begins to develop to five and 10 after which go down once more.
6. Consumer habits simulation: Digital customers will look forward to a random time between duties, as outlined by the wait_time within the script. This simulates extra life like person habits. We didn’t do that in our case however you are able to do this and extra superior issues in Locust like customized load shapes, and so forth.
7. Steady take a look at execution: The take a look at will proceed operating till you determine to cease it, or till it reaches a predefined period if you happen to’ve set one.
8. Useful resource utilization: Throughout this course of, Locust makes use of your machine’s sources to simulate the customers and make requests. It is essential to notice that the efficiency of the Locust take a look at can even rely upon the sources of the machine it is operating on.

Let’s now interpret the outcomes we’re seeing.

Deciphering and validating load testing outcomes

Deciphering outcomes from a Locust run includes understanding key metrics and what they point out concerning the efficiency of the system underneath take a look at. Listed below are a few of the foremost metrics supplied by Locust and the best way to interpret them:

• Variety of customers: The overall variety of simulated customers at any given level within the take a look at. This helps you perceive the load stage in your system. You possibly can correlate system efficiency with the variety of customers to find out at what level efficiency degrades.
• Requests per second (RPS): The variety of requests (queries) made to your system per second. A better RPS signifies a better load. Examine this with response occasions and error charges to evaluate if the system can deal with concurrency and excessive visitors easily.
• Response time: Often displayed as common, median, and percentile (e.g., ninetieth and 99th percentile) response occasions. You’ll seemingly take a look at median and the 90/99 percentile as this provides you the expertise for “most” customers – solely 10 or 1 % may have worse expertise.
• Failure price: The share or variety of requests that resulted in an error. A excessive failure price signifies issues with the system underneath take a look at. It is essential to investigate the character of those errors.

Beneath you’ll be able to see the whole RPS and response occasions we achieved underneath totally different hundreds for our load take a look at, going from a single person to 10 customers after which down once more.

Our RPS went as much as about 20 whereas sustaining median question latency under 300 milliseconds and P99 of 700 milliseconds.

We will now correlate these information factors with the out there digital occasion metrics in Rockset. Beneath, you’ll be able to see how the digital occasion handles the load when it comes to CPU, reminiscence and question latency. There’s a correlation between variety of customers from Locust and the peaks we see on the VI utilization graphs. You can too see the question latency beginning to rise and see the concurrency (requests or queries per second) go up. The CPU is under 75% on the height and reminiscence utilization appears to be like steady. We additionally don’t see any important queueing occurring in Rockset.

Aside from viewing these metrics within the Rockset console or via our metrics endpoint, you too can interpret and analyze the precise SQL queries that have been operating, what was their particular person efficiency, queue time, and so forth. To do that, we should first allow question logs after which we will do issues like this to determine our median run and queue occasions:

SELECT
    query_sql,
    COUNT(*) as depend,
    ARRAY_SORT(ARRAY_AGG(runtime_ms)) [(COUNT(*) + 1) / 2] as median_runtime,
    ARRAY_SORT(ARRAY_AGG(queued_time_ms)) [(COUNT(*) + 1) / 2] as median_queue_time
FROM
    commons."QueryLogs"
WHERE
    vi_id = ''
    AND _event_time > TIMESTAMP '2023-11-24 09:40:00'
GROUP BY
    query_sql

We will repeat this load take a look at on the primary VI as nicely, to see how the system performs ingestion and runs queries underneath load. The method could be the identical, we’d simply use a distinct VI identifier in our Locust script in Step 6.

Conclusion

In abstract, load testing is a vital a part of making certain the reliability and efficiency of any database resolution, together with Rockset. By choosing the precise load testing software and organising Rockset appropriately for load testing, you’ll be able to acquire helpful insights into how your system will carry out underneath numerous circumstances.

Locust is simple sufficient to get began with rapidly, however as a result of Rockset has REST API assist for executing queries and question lambdas, it’s straightforward to hook up any load testing software.

Bear in mind, the purpose of load testing is not only to establish the utmost load your system can deal with, but additionally to know the way it behaves underneath totally different stress ranges and to make sure that it meets the required efficiency requirements.

Fast load testing ideas earlier than we finish the weblog:

• All the time load take a look at your system earlier than going to manufacturing
• Use question lambdas in Rockset to simply parametrize, version-control and expose your queries as REST endpoints
• Use compute-compute separation to carry out load testing on a digital occasion devoted for queries, in addition to in your foremost (ingestion) VI
• Allow question logs in Rockset to maintain statistics of executed queries
• Analyze the outcomes you’re getting and evaluate them towards your SLAs – if you happen to want higher efficiency, there are a number of methods on the best way to sort out this, and we’ll undergo these in a future weblog.

Have enjoyable testing 💪

Helpful sources

Listed below are some helpful sources for JMeter, Gatling and k6. The method is similar to what we’re doing with Locust: you’ll want to have an API key and authenticate towards Rockset after which hit the question lambda REST endpoint for a selected digital occasion.