Introduction

Microsoft SQL Server 2019 introduced a groundbreaking data platform with SQL Server 2019 Big Data Clusters (BDC). Microsoft SQL Server Big Data Clusters are designed to solve the big data challenge faced by most organizations today. You can use SQL Server BDC to organize and analyze large volumes of data, you can also combine high value relational data with big data. This blog post describes the deployment of SQL Server BDC on a Kubernetes platform using Dell EMC PowerFlex software-defined storage.

PowerFlex

Dell EMC PowerFlex (previously VxFlex OS) is the software foundation of PowerFlex software-defined storage. It is a unified compute storage and networking solution delivering scale-out block storage service that is designed to deliver flexibility, elasticity, and simplicity with predictable high performance and resiliency at scale.

The PowerFlex platform is available in multiple consumption options to help customers meet their project and data center requirements. PowerFlex appliance and PowerFlex rack provide customers comprehensive IT Operations Management (ITOM) and life cycle management (LCM) of the entire infrastructure stack in addition to sophisticated high-performance, scalable, resilient storage services. PowerFlex appliance and PowerFlex rack are the preferred and proactively marketed consumption options. PowerFlex is also available on VxFlex Ready Nodes for those customers who are interested in software-defined compliant hardware without the ITOM and LCM capabilities.

PowerFlex software-define storage with unified compute and networking offers flexibility of deployment architecture to help best meet the specific deployment and architectural requirements. PowerFlex can be deployed in a two-layer for asymmetrical scaling of compute and storage for “right-sizing capacities, single-layer (HCI), or in mixed architecture.

Microsoft SQL Server Big Data Clusters Overview

Microsoft SQL Server Big Data Clusters are designed to address big data challenges in a unique way, BDC solves many traditional challenges through building big-data and data-lake environments. You can query external data sources, store big data in HDFS managed by SQL Server, or query data from multiple external data sources using the cluster.

SQL Server Big Data Clusters is an additional feature of Microsoft SQL Server 2019. You can query external data sources, store big data in HDFS managed by SQL Server, or query data from multiple external data sources using the cluster.

For more information, see the Microsoft page SQL Server Big Data Clusters partners.

You can use SQL Server Big Data Clusters to deploy scalable clusters of SQL Server and Apache Spark^TM and Hadoop Distributed File System (HDFS), as containers running on Kubernetes. 

For an overview of Microsoft SQL Server 2019 Big Data Clusters, see Microsoft’s Introducing SQL Server Big Data Clusters and on GitHub, see Workshop: SQL Server Big Data Clusters - Architecture.

Deploying Kubernetes Platform on PowerFlex

For this test, PowerFlex 3.6.0 is built in a two-layer configuration with six Compute Only (CO) nodes and eight Storage Only (SO) nodes. We used PowerFlex Manager to automatically provision the PowerFlex cluster with CO nodes on VMware vSphere 7.0 U2, and SO nodes with Red Hat Enterprise Linux 8.2.  

The following figure shows the logical architecture of SQL Server BDC on Kubernetes platform with PowerFlex.

Figure 1: Logical architecture of SQL BDC on PowerFlex

From the storage perspective, we created a single protection domain from eight PowerFlex nodes for SQL BDC. Then we created a single storage pool using all the SSDs installed in each node that is a member of the protection domain.

After we deployed the PowerFlex cluster, we created eleven virtual machines on the six identical CO nodes with Ubuntu 20.04 on them, as shown in the following table.

Table 1: Virtual machines for CO nodes

We manually installed the SDC component of PowerFlex on the worker nodes of Kubernetes. We then configured a Kubernetes cluster (v 1.20) on the virtual machines with three master nodes and eight worker nodes: 

$ kubectl get nodes

NAME   STATUS    ROLES                  AGE   VERSION

k8m1   Ready     control-plane,master   10d   v1.20.10

k8m2   Ready     control-plane,master   10d   v1.20.10

k8m3   Ready     control-plane,master   10d   v1.20.10

k8w1   Ready     <none>                 10d   v1.20.10

k8w2   Ready     <none>                 10d   v1.20.10

k8w3   Ready     <none>                 10d   v1.20.10

k8w4   Ready     <none>                 10d   v1.20.10

k8w5   Ready     <none>                 10d   v1.20.10

k8w6   Ready     <none>                 10d   v1.20.10

Dell EMC storage solutions provide CSI plugins that allow customers to deliver persistent storage for container-based applications at scale. The combination of the Kubernetes orchestration system and the Dell EMC PowerFlex CSI plugin enables easy provisioning of containers and persistent storage.

In the solution, after we installed the Kubernetes cluster, CSI 2.0 was provisioned to enable persistent volumes for SQL BDC workload.

For more information about PowerFlex CSI supported features, see Dell CSI Driver Documentation.

For more information about PowerFlex CSI installation using Helm charts, see PowerFlex CSI Documentation.

Deploying Microsoft SQL Server BDC on Kubernetes Platform

When the Kubernetes cluster with CSI is ready, Azure data CLI is installed on the client machine. To create base configuration files for deployment, see deploying Big Data Clusters on Kubernetes . For this solution, we used kubeadm-dev-test as the source for the configuration template.

Initially, using kubectl, each node is labelled to ensure that the pods start on the correct node:

$ kubectl label node k8w1 mssql-cluster=bdc mssql-resource=bdc-master --overwrite=true

$ kubectl label node k8w2 mssql-cluster=bdc mssql-resource=bdc-compute-pool --overwrite=true

$ kubectl label node k8w3 mssql-cluster=bdc mssql-resource=bdc-compute-pool --overwrite=true

$ kubectl label node k8w4 mssql-cluster=bdc mssql-resource=bdc-compute-pool --overwrite=true

$ kubectl label node k8w5 mssql-cluster=bdc mssql-resource=bdc-compute-pool --overwrite=true

$ kubectl label node k8w6 mssql-cluster=bdc mssql-resource=bdc-compute-pool --overwrite=true

To accelerate the deployment of BDC, we recommend that you use an offline installation method from a local private registry. While this means some extra work in creating and configuring a registry, it eliminates the network load of every BDC host pulling container images from the Microsoft repository. Instead, they are pulled once. On the host that acts as a private registry, install Docker and enable the Docker repository.

The BDC configuration is modified from the default settings to use cluster resources and address the workload requirements. For complete instructions about modifying these settings, see Customize deployments section in the Microsoft BDC website. To scale out the BDC resource pools, the number of replicas are adjusted to use the resources of the cluster.

The values shown in the following table are adjusted in the bdc.json file.  

Table 2: Cluster resources

The configuration values for running Spark and Apache Hadoop YARN are also adjusted to the compute resources available per node. In this configuration, sizing is based on 768 GB of RAM and 72 virtual CPU cores available per PowerFlex CO node. Most of this configuration is estimated and adjusted based on the workload. In this scenario, we assumed that the worker nodes were dedicated to running Spark workloads. If the worker nodes are performing other operations or other workloads, you may need to adjust these values. You can also override Spark values as job parameters.

For further guidance about configuration settings for Apache Spark and Apache Hadoop in Big Data Clusters, see Configure Apache Spark & Apache Hadoop in the SQL Server BDC documentation section.

The following table highlights the spark settings that are used on the SQL Server BDC cluster.

Table 3: Spark settings

The SQL Server BDC 2019 CU12 release notes state that Kubernetes API 1.20 is supported. Therefore, for this test, the image that was deployed on the SQL master pod was 2019-CU12-ubuntu-16.04.   A storage of 20 TB was provisioned for SQL master pod, with 10 TB as log space:

   "nodeLabel": "bdc-master",

                    "storage": {

                        "data": {

                            "className": "vxflexos-xfs",

                            "accessMode": "ReadWriteOnce",

                            "size": "20Ti"

                        },

                        "logs": {

                            "className": "vxflexos-xfs",

                            "accessMode": "ReadWriteOnce",

                            "size": "10Ti"

                        }

                    }

Because the test involved running the TPC-DS workload, we provisioned a total of 60 TB of space for five storage pods:

"storage-0": {

                "metadata": {

                    "kind": "Pool",

                    "name": "default"

                },

                "spec": {

                    "type": "Storage",

                    "replicas": 5,

                    "settings": {

                        "spark": {

                            "includeSpark": "true"

                        }

                    },

                    "nodeLabel": "bdc-compute-pool",

                    "storage": {

                        "data": {

                            "className": "vxflexos-xfs",

                            "accessMode": "ReadWriteOnce",

                            "size": "12Ti"

                        },

                        "logs": {

                            "className": "vxflexos-xfs",

                            "accessMode": "ReadWriteOnce",

                            "size": "4Ti"

                        }

                    }

                }

            }

Validating SQL Server BDC on PowerFlex

To validate the configuration of the Big Data Cluster that is running on PowerFlex and to test its scalability, we ran the TPC-DS workload on the cluster using the Databricks^® TPC-DS Spark SQL kit. The toolkit allows you to submit an entire TPC-DS workload as a Spark job that generates the test dataset and runs a series of analytics queries across it. Because this workload runs entirely inside the storage pool of the SQL Server Big Data Cluster, the environment was scaled to run the recommended maximum of five storage pods.

We assigned one storage pod to each worker node in the Kubernetes environment as shown in the following figure.

Figure 2: Pod placement across worker nodes

In this solution, Spark SQL TPC-DS workload is adopted to simulate a database environment that models several applicable aspects of a decision support system, including queries and data maintenance. Characterized by high CPU and I/O load, a decision support workload places a load on the SQL Server BDC cluster configuration to extract maximum operational efficiencies in areas of CPU, memory, and I/O utilization. The standard result is measured by the query response time and the query throughput.

A Spark JAR file is uploaded into a specified directory in HDFS, for example, /tpcds. The spark-submit is done by CURL, which uses the Livy server that is part of Microsoft SQL Server Big Data Cluster.

Using the Databricks TPC-DS Spark SQL kit, the workload is run as Spark jobs for the 1 TB, 5 TB, 10 TB, and 30 TB workloads. For each workload, only the size of the dataset is changed.

The parameters used for each job are specified in the following table.

Table 4: Job parameters

 We set the TPC-DS dataset with the different scale factors in the CURL command. The data was populated directly into the HDFS storage pool of the SQL Server Big Data Cluster. 

The following figure shows the time that is consumed for data generation of different scale factor settings. The data generation time also includes the post data analysis process that calculates the table statistics.

Figure 3: TPC-DS data generation

After the load we ran the TPC-DS workload to validate the Spark SQL performance and scalability with 99 predefined user queries. The queries are characterized with different user patterns. 

The following figure shows the performance and scalability test results. The results demonstrate that running Microsoft SQL Server Big Data Cluster on PowerFlex has linear scalability for different datasets. This shows the ability of PowerFlex to provide a consistent and predictable performance for different types of Spark SQL workloads.

Figure 4: TPC-DS test results

A Grafana dashboard instance that is captured during the 30 TB run of TPC-DS test is shown in the following figure. The figure shows that the read bandwidth of 15 GB/s is achieved during the tests. 

Figure 5: Grafana dashboard

In this minimal lab hardware, there were no storage bottlenecks for the TPC-DS data load and query execution. The CPU on the worker nodes reached close to 90 percent indicating that more powerful nodes could enhance the performance. 

Conclusion

Running SQL Server Big Data Clusters on PowerFlex is a straightforward way to get started with modernized big data workloads running on Kubernetes. This solution allows you to run modern containerized workloads using the existing IT infrastructure and processes. Big Data Clusters allows Big Data scientists to innovate and build with the agility of Kubernetes, while IT administrators manage the secure workloads in their familiar vSphere environment.

In this solution, Microsoft SQL Server Big Data Clusters are deployed on PowerFlex which provides the simplified operation of servicing cloud native workloads and can scale without compromise. IT administrators can implement policies for namespaces and manage access and quota allocation for application focused management. Application-focused management helps you build a developer-ready infrastructure with enterprise-grade Kubernetes, which provides advanced governance, reliability, and security.

Microsoft SQL Server Big Data Clusters are also used with Spark SQL TPC-DS workloads with the optimized parameters. The test results show that Microsoft SQL Server Big Data Clusters deployed in a PowerFlex environment can provide a strong analytics platform for Big Data solutions in addition to data warehousing type operations.

For more information about PowerFlex, see Dell EMC PowerFlex. For more information about Microsoft SQL Server Big Data Clusters, see Introducing SQL Big Data Clusters.

If you want to discover more, contact your Dell representative. 

References

• Dell Technologies Info Hub for PowerFlex documentation portal
• Microsoft SQL Server 2019 Big Data Clusters: A Big Data Solution Using Dell EMC Infrastructure
• Dell EMC PowerFlex product portal

Quick! The word has just come down. There is a new initiative that requires a massively parallel processing (MPP) database, and you are in charge of implementing it. What are you going to do? Luckily, you know the answer. You also just discovered that the Dell PowerFlex Solutions team has you covered with a solutions guide for VMware Tanzu Greenplum.

What is in the solutions guide and how will it help with an MPP database? This blog provides the answer. We look at what Greenplum is and how to leverage Dell PowerFlex for both the storage and compute resources in Greenplum.

Infrastructure flexibility: PowerFlex

If you have read my other blogs or are familiar with PowerFlex, you know it has powerful transmorphic properties. For example, PowerFlex nodes sometimes function as both storage and compute, like hyperconverged infrastructure (HCI). At other times, PowerFlex functions as a storage-only (SO) node or a compute-only (CO) node. Even more interesting, these node types can be mixed and matched in the same environment to meet the needs of the organization and the workloads that they run.    

This transmorphic property of PowerFlex is helpful in a Greenplum deployment, especially with the configuration described in the solutions guide. Because the deployment is built on open-source PostgreSQL, it is optimized for the needs of an MPP database, like Greenplum. PowerFlex can deliver the compute performance necessary to support massive data IO with its CO nodes. The PowerFlex infrastructure can also support workloads running on CO nodes or nodes that combine compute and storage (hybrid nodes). By leveraging the malleable nature of PowerFlex, no additional silos are needed in the data center, and it may even help remove existing ones.

The architecture used in the solutions guide consists of 12 CO nodes and 10 SO nodes. The CO nodes have VMware ESXi installed on them, with Greenplum instances deployed on top. There are 10 segments and one director deployed for the Greenplum environment. The 12th CO node is used for redundancy.

The storage tier uses the 10 SO nodes to deliver 12 volumes backed by SSDs. This configuration creates a high speed, highly redundant storage system that is needed for Greenplum. Also, two protection domains are used to provide both primary and mirror storage for the Greenplum instances. Greenplum mirrors the volumes between those protection domains, adding an additional level of protection to the environment, as shown in the following figure:

By using this fluid and composable architecture, the components can be scaled independently of one another, allowing for storage to be increased either independently or together with compute. Administrators can use this configuration to optimize usage and deliver appropriate resources as needed without creating silos in the environment.  

Testing and validation with Greenplum: we have you covered

The solutions guide not only describes how to build a Greenplum environment, it also addresses testing, which many administrators want to perform before they finish a build. The guide covers performing basic validations with FIO and gpcheckperf. In the simplest terms, these tools ensure that IO, memory, and network performance are acceptable. The FIO tests that were run for the guide showed that the HBA was fully saturated, maximizing both read and write operations. The gpcheckperf testing showed a performance of 14,283.62 MB/sec for write workloads.

Wouldn’t you feel better if a Greenplum environment was tested with a real-world dataset? That is, taking it beyond just the minimum, maximum, and average numbers? The great news is that the architecture was tested that way! Our Dell Digital team has developed an internal test suite running static benchmarked data. This test suite is used at Dell Technologies across new Greenplum environments as the gold standard for new deployments.

In this test design, all the datasets and queries are static. This scenario allows for a consistent measurement of the environment from one run to the next. It also provides a baseline of an environment that can be used over time to see how its performance has changed -- for example, if the environment sped up or slowed down following a software update.

Massive performance with real data

So how did the architecture fare? It did very well! When 182 parallel complex queries were run simultaneously to stress the system, it took just under 12 minutes for the test to run. In that time, the environment had a read bandwidth of 40 GB/s and a write bandwidth of 10 GB/s. These results are using actual production-based queries from the Dell Digital team workload. These results are close to saturating the network bandwidth for the environment, which indicates that there are no storage bottlenecks.

The design covered in this solution guide goes beyond simply verifying that the environment can handle the workload; it also shows how the configuration can maintain performance during ongoing operations.

Maintaining performance with snapshots

One of the key areas that we tested was the impact of snapshots on performance. Snapshots are a frequent operation in data centers and are used to create test copies of data as well as a source for backups. For this reason, consider the impact of snapshots on MPP databases when looking at an environment, not just how fast the database performs when it is first deployed.

In our testing, we used the native snapshot capabilities of PowerFlex to measure the impact that snapshots have on performance. Using PowerFlex snapshots provides significant flexibility in data protection and cloning operations that are commonly performed in data centers.

We found that when the first storage-consistent snapshot of the database volumes was taken, the test took 45 seconds longer to complete than initial tests. This result was because it was the first snapshot of the volumes. Follow-on snapshots during testing resulted in minimal impact to the environment. This minimal impact is significant for MPP databases in which performance is important. (Of course, performance can vary with each deployment.)  

We hope that these findings help administrators who are building a Greenplum environment feel more at ease. You not only have a solution guide to refer to as you architect the environment, you can be confident that it was built on best-in-class infrastructure and validated using common testing tools and real-world queries.

The bottom line

Now that you know the assignment is coming to build an MPP database using VMware Tanzu Greenplum -- are you up to the challenge?

If you are, be sure to read the solution guide. If you need additional guidance on building your Greenplum environment on PowerFlex, be sure to reach out to your Dell representative.  

Resources

• Dell PowerFlex
• VMware Tanzu Greenplum

Authors:

• Tony Foster – Dell Technologies, Twitter: @wonder_nerd
  LinkedIn
• Sue Mosovich – VMware