This is a step-by-step guide to get started with benchmarking on Vespa Cloud, based on the Vespa benchmarking guide, using the sample app. Overview:

Vespa Cloud Benchmarking

Set up a perf instance

Use an instance in the perf zone for benchmarks. To deploy an instance there, use the getting started guide, set the perf zone in the deploy step set:

$ vespa deploy -z --wait 600

Feed documents:

$ vespa feed -z ext/documents.jsonl

Query documents to validate the feed:

$ vespa query -z "select * from music where true"

Query documents using curl:

$ curl \
  --cert ~/.vespa/mytenant.myapp.default/data-plane-public-cert.pem \
  --key  ~/.vespa/mytenant.myapp.default/data-plane-private-key.pem \
  -H "Content-Type: application/json" \
  --data '{"yql" : "select * from music where true"}' \

At this point, the instance is ready, with data, and can be queried using data-plane credentials.

Test using vespa-fbench

The rest of the guide assumes the data-plane credentials are in working directory:

$ ls -1 *.pem

Prepare a query file:

$ echo "/search/?yql=select+*+from+music+where+true" > query001.txt

Test using vespa-fbench running in a docker container:

$ docker run -v $(pwd):/files -w /files \
    --entrypoint /opt/vespa/bin/vespa-fbench \
    vespaengine/vespa \
    -C data-plane-public-cert.pem \
    -K data-plane-private-key.pem \
    -T /etc/ssl/certs/ca-bundle.crt \
    -n 1 -q query001.txt -s 1 -c 0 \
    -o output.txt \ 443

-o output.txt is useful when validating the test - remove this option when load testing. Make sure there are no SSL_do_handshake errors in the output. Expect HTTP status code 200:

  Starting clients...
  Stopping clients
  Clients stopped.
  Clients Joined.
  *** HTTP keep-alive statistics ***
  connection reuse count -- 4
  ***************** Benchmark Summary *****************
  clients:                       1
  ran for:                       1 seconds
  cycle time:                    0 ms
  lower response limit:          0 bytes
  skipped requests:              0
  failed requests:               0
  successful requests:           5
  cycles not held:               5
  minimum response time:    128.17 ms
  maximum response time:    515.35 ms
  average response time:    206.38 ms
  25 percentile:            128.70 ms
  50 percentile:            129.60 ms
  75 percentile:            130.20 ms
  90 percentile:            361.32 ms
  95 percentile:            438.36 ms
  99 percentile:            499.99 ms
  actual query rate:          4.80 Q/s
  utilization:               99.03 %
  zero hit queries:              5
  http request status breakdown:
    200 :        5

At this point, running queries using vespa-fbench works well from local laptop.

Run queries inside data center

Next step is to run this from the same location (data center) as the perf zone. In this example, an AWS zone. Deduce the AWS zone from Vespa Cloud zone name. Below is an example using a host with Amazon Linux 2023 AMI (HVM) image:

  1. Create the host - here assume key pair is named key.pem. No need to do anything other than default.

  2. Log in, update, install docker:

    $ ssh -i key.pem
    [ec2-user]$ sudo yum update -y
    [ec2-user]$ sudo yum install -y docker
    [ec2-user]$ sudo service docker start
    [ec2-user]$ sudo usermod -a -G docker ec2-user
    [ec2-user]$ exit
  3. Copy credentials for endpoint access, log in and validate docker setup:

    $ scp -i key.pem data-plane-private-key.pem
    $ scp -i key.pem data-plane-public-cert.pem
    $ ssh -i key.pem
    [ec2-user]$ docker info
  4. Make a dummy query:

    [ec2-user]$ echo "/search/?yql=select+*+from+music+where+true" > query001.txt
  5. Run vespa-fbench and verify 200 response:

    [ec2-user]$ docker run -v $(pwd):/files -w /files \
                  --entrypoint /opt/vespa/bin/vespa-fbench \
                  vespaengine/vespa \
                  -C data-plane-public-cert.pem \
                  -K data-plane-private-key.pem \
                  -T /etc/ssl/certs/ca-bundle.crt \
                  -n 1 -q query001.txt -s 1 -c 0 \

At this point, you are able to benchmark using vespa-fbench in the same zone as the Vespa Cloud perf instance.

Run benchmark

Use the Vespa Benchmarking Guide to plan and run benchmarks. Also see sizing below. Make sure the client running the benchmark tool has sufficient resources.

Export metrics:

$ curl \
    --cert data-plane-public-cert.pem \
    --key data-plane-private-key.pem \


Making changes

Whenever deploying changes to configuration, track progress in the Deployment dashboard. Some changes, like changing requestthreads will restart content nodes, and this is done in sequence and takes time. Wait for successful completion in Wait for services and endpoints to come online.

When changing node type/count, wait for auto data redistribution to complete, watching the vds.idealstate.merge_bucket.pending.average metric:

$ while true; do curl -s \
    --cert data-plane-public-cert.pem \
    --key data-plane-private-key.pem \ | \
    grep idealstate.merge_bucket.pending.average; \
    sleep 10; done



Using Vespa Cloud enables the Vespa Team to assist you to optimise the application to reduce resource spend. Based on 150 applications running on Vespa Cloud today, savings are typically 50%. Cost optimization is hard to do without domain knowledge - but few teams are experts in both their application and its serving platform. Sizing means finding both the right node size and the right cluster topology:

Resize to fewer and smaller nodes

Applications use Vespa for their primary business use cases. Availability and performance vs. cost are business decisions. The best sized application can handle all expected load situations, and is configured to degrade quality gracefully for the unexpected.

Even though Vespa is cost-efficient out of the box, Vespa experts can usually spot over/under-allocations in CPU, memory and disk space/IO, and discuss trade-offs with the application team.

Using automated deployments applications go live with little risk. After launch, right-size the application based on true load after using Vespa’s elasticity features with automated data migration.

Use the Vespa sizing guide to size the application and find metrics used there. Pro-tips:

Rules of thumb: