Amazon EMR is the industry-leading cloud huge information answer, offering a set of open-source frameworks comparable to Spark, Hive, Hudi, and Presto, absolutely managed and with per-second billing. Amazon EMR on Amazon EKS is a deployment choice permitting you to deploy Amazon EMR on the identical Amazon Elastic Kubernetes Service (Amazon EKS) clusters that’s multi-tenant and utilized by different functions, enhancing useful resource utilization, lowering value, and simplifying infrastructure administration. EMR on EKS present you as much as 5.37 instances higher efficiency than OSS Spark v3.3.1 with 76.8% value financial savings. It additionally gives all kinds of job submission strategies, like an AWS API referred to as StartJobRun, or by means of a declarative manner with a Kubernetes controller by means of the AWS Controllers for Kubernetes for Amazon EMR on EKS.

This consolidation comes with a trade-off of elevated issue measuring fine-grained prices for showback or chargeback by group or utility. In line with a CNCF and FinOps Basis survey, 68% of Kubernetes customers both depend on month-to-month estimates or don’t monitor Kubernetes prices in any respect. And for respondents reporting energetic Kubernetes value monitoring, AWS Value Explorer and Kubecost have been ranked as the most well-liked instruments getting used.

At present, you possibly can distribute prices per tenant utilizing a tough multi-tenancy with separate EKS clusters in devoted AWS accounts or a tender multi-tenancy utilizing separate node teams in a shared EKS cluster. To cut back prices and enhance useful resource utilization, you should utilize namespace-based segregation, the place nodes are shared throughout completely different namespaces. Nevertheless, calculating and attributing prices to groups by workload or namespaces whereas taking into consideration compute optimization (like Saving Plans or Spot Occasion value) and the price of AWS providers like EMR on EKS is a difficult and non-trivial process.

On this put up, we current a value chargeback answer for EMR on EKS that mixes the AWS-native capabilities of AWS Value and Utilization Reviews (AWS CUR) alongside the in-depth Kubernetes value visibility and insights utilizing Kubecost on Amazon EKS.

Resolution overview

A job in EMR on EKS incur prices primarily on two dimensions: compute assets and a marginal uplift cost for EMR on EKS utilization. To trace the price related to every of the size, we use information from three sources:

• AWS CUR – We use this to get the EMR on EKS value uplift per job and for Kubecost to reconcile the compute value with any saving plans or reserved occasion used. The supporting infrastructure for CUR is deployed as outlined in Organising Athena utilizing AWS CloudFormation templates.
• Kubecost – We use this to get the compute value incurred by the executor and driver pods.

The price allocation course of contains the next parts:

• The compute value is offered by Kubecost. Nevertheless, with a purpose to do an in-depth evaluation, we outline an hourly Kubernetes CronJob on it that begins a pod to retrieve information from Kubecost and shops it in Amazon Easy Storage Service (Amazon S3).
• CUR recordsdata are saved in an S3 bucket.
• We use Amazon Athena to create a view and supply a consolidated view of the entire value to run an EMR on EKS job.
• Lastly, you possibly can join your most well-liked enterprise intelligence instruments utilizing the JDBC or ODBC connections to Athena. On this put up, we use Amazon QuickSight native integration for visualization functions.

The next diagram exhibits the general structure in addition to how the completely different parts work together with one another.

We offer a shell script to deploy our the monitoring answer. The shell script configures the infrastructure utilizing an AWS CloudFormation template, the AWS Command Line Interface (AWS CLI), and eksctl and kubectl instructions. This script runs the next actions:

1. Begin the CloudFormation deployment.
2. Create and configure an AWS Value and Utilization Report.
3. Configure and deploy Kubecost backed by Amazon Managed Service for Prometheus.
4. Deploy a Kubernetes CronJob.

Conditions

You want the next stipulations:

This put up assumes you have already got an EKS cluster and run EMR on EKS jobs. In the event you don’t have an EKS cluster prepared to check the answer, we advise beginning with an ordinary EMR on EKS blueprint that configures a cluster to submit EMR on EKS jobs.

Arrange the answer

To run the shell script, full the next steps:

1. Clone the next GitHub repository.
2. Go to the folder cost-tracking with the next command:

cd cost-tracking

3. Run the script with following command :

sh deploy-emr-eks-cost-tracking.sh REGION KUBECOST-VERSION EKS-CLUSTER-NAME ACCOUNT-ID

After you run the script, you’re prepared to make use of Kubecost and the CUR information to grasp the price related together with your EMR on EKS jobs.

Monitoring value

On this part, we present you how you can analyze the compute value that’s retrieved from Kubecost, how you can question EMR on EKS uplift information, and how you can mix them to have a single consolidated view for the price.

Compute value

Kubecost presents varied methods to trace value per Kubernetes object. For instance, you possibly can monitor value by pod, controller, job, label, or deployment. It additionally permits you to perceive the price of idle assets, like Amazon Elastic Compute Cloud (Amazon EC2) situations that aren’t absolutely utilized by pods. On this put up, we assume that no nodes are provisioned if no EMR on EKS job is working, and we use the Karpenter Cluster Autoscaler to provision nodes when jobs are submitted. Karpenter additionally does bin packing, which optimizes the EC2 useful resource utilization and in flip reduces the price of idle assets.

To trace compute value related to EMR on EKS pods, we question the Kubecost allocation API by passing pod and labels within the combination parameter. We use the emr-containers.amazonaws.com/job.id and emr-containers.amazonaws.com/virtual-cluster-id labels which might be at all times current in executor and driver pods. The labels are used to filter Kubecost information to get solely the price related to EMR on EKS pods. You possibly can assessment varied ranges of granularity on the pod, job, and digital cluster degree to grasp the price of a driver vs. executor, or of utilizing Spot Cases in jobs. You may as well use the digital cluster value to grasp the general value of a EMR on EMR when it’s utilized in a namespace that’s utilized by functions apart from EMR on EKS.

We additionally present the instance_id, occasion dimension, and capability sort (On-Demand or Spot) that was used to run the pod. That is retrieved by means of querying the Kubecost belongings API. This information will be helpful to grasp the way you run your jobs and which capability you utilize extra typically.

The information about the price of working the pods in addition to the belongings is retrieved with a Kubernetes CronJob that submits the request to the Kubecost API, joins the 2 information sources (allocation and belongings information) on the instance_id, cleans the information, and shops it in Amazon S3 in CSV format.

The compute value information has a number of fields which might be of curiosity, together with cpucost, ramcost (value of reminiscence), pvcost (value of Amazon EBS storage), effectivity of use of CPU and RAM, in addition to complete value, which represents the mixture value of all of the assets used, both at pod, job, or digital cluster degree.

To view this information, full the next steps:

1. On the Athena console, navigate to the question editor.
2. Select athenacurcfn_c_u_r for the database and cost_data for the desk.
3. Run the next question:

SELECT job_id,
vc_id,
sum(totalcost) as value
FROM "athenacurcfn_c_u_r"."compute_cost"
GROUP BY job_id, vc_id

The next screenshot exhibits the question outcomes.

To question the information about info on the pod degree, you possibly can run the next SQL assertion:

SELECT
split_part(title, '/', 1) as pod_name,
job_id,
vc_id,
totalcost,
instance_id,
"properties.labels.node_kubernetes_io_instance_type",
capacity_type
FROM "athenacurcfn_c_u_r"."compute_cost";

EMR on EKS uplift

The price related to EMR on EKS uplift is accessible by means of AWS CUT and is saved in an S3 bucket. The script you ran within the setup step created an Athena desk related to the information within the S3 bucket. The next steps take you thru how one can question the information:

1. On the Athena console, navigate to the question editor.
2. Select athenacurcfn_c_u_r for the database and cur_data for the desk.
3. Run the next question:

SELECT
split_part(line_item_resource_id, '/', 5) as job_id,
split_part(line_item_resource_id, '/', 3) as vc_id,
sum(line_item_blended_cost) as value
FROM athenacurcfn_c_u_r.automated
WHERE product_product_family='EMR Containers'
GROUP BY line_item_resource_id

This question gives you with the price per job. The next screenshot exhibits the outcomes.

You’ll have to wait as much as 24 hours for the CUR information to be out there. As such, it is best to solely run the previous question after the CUR information is accessible and you’ve got run the EMR on EKS jobs.

General value

To view the general value and carry out evaluation on it, create a view in Athena as follows:

CREATE VIEW emr_eks_cost AS
SELECT
split_part(line_item_resource_id, '/', 5) as job_id,
split_part(line_item_resource_id, '/', 3) as vc_id,
sum(line_item_blended_cost) as value,
'emr-uplift' as class
FROM athenacurcfn_c_u_r.cur_data
WHERE product_product_family='EMR Containers'
GROUP BY line_item_resource_id
UNION
SELECT
job_id,
vc_id,
sum(totalCost) as value,
'compute' as class
FROM "athenacurcfn_c_u_r"."compute_cost"
group by job_id, vc_id

Now that the view is created, you possibly can question and analyze the price of working your EMR on EKS jobs:

SELECT sum(value) as total_cost, job_id, vc_id
FROM "athenacurcfn_c_u_r"."emr_eks_cost"
GROUP BY job_id, vc_id;

The next screenshot exhibits an instance output of the question on the created view.

Lastly, you should utilize QuickSight for a graphical high-level view in your EMR on EKS spend. The next screenshot exhibits an instance dashboard.

Now you can adapt this answer to your particular wants and construct your customized evaluation.

Clear up

All through this put up, you deployed and configured the required infrastructure parts to trace value on your EMR on EKS workloads. To keep away from incurring extra fees for this answer, delete all of the assets you created:

1. Empty the S3 buckets cost-data-REGION-ACCOUNT_ID and aws-athena-query-results-cur-REGION-ACCOUNT_ID.
2. Delete the Athena workgroup kubecost-cur-workgroup.
3. Empty and delete the ECR repository emreks-compute-cost-exporter.
4. Run the script destroy-emr-eks-cost-tracking.sh, which can delete the AWS CloudFormation deployment, uninstall Kubecost, delete the CronJob, and delete the Value and Utilization Reviews.

Conclusion

On this put up, we confirmed how you should utilize Kubecost capabilities alongside Value and Utilization Reviews to carefully monitor the prices for Amazon EMR on EKS per digital cluster or per job. This answer permits you to obtain extra granular prices for chargebacks utilizing Athena, Amazon Managed Service for Prometheus, and QuickSight.

The answer introduced steps to arrange Value and Utilization Reviews and Kubecost, and configure a CronJob on an hourly foundation to get the price of working pods spun by EMR on EKS. You possibly can modify the introduced answer to run at longer intervals or to gather information on completely different EKS clusters. You may as well modify the Python script run by the CronJob to additional clear information or cut back the quantity of information saved by eliminating fields you don’t want. You should use the insights offered to drive value optimization efforts over time, detect any improve of prices, and measure the affect of recent deployments or explicit occasions on useful resource utilization and value efficiency. For extra details about integrating EMR on EKS in your present Amazon EKS deployment, consult with Design concerns for Amazon EMR on EKS in a multi-tenant Amazon EKS setting

Concerning the Authors

Lotfi Mouhib is a Senior Options Architect working for the Public Sector group with Amazon Net Providers. He helps public sector clients throughout EMEA notice their concepts, construct new providers, and innovate for residents. In his spare time, Lotfi enjoys biking and working.

Hamza Mimi Principal Options Architect within the French Public sector group at Amazon Net Providers (AWS). With an extended expertise within the telecommunications {industry}. He’s presently working as a buyer advisor on matters starting from digital transformation to architectural steering.