IBM Cloud and Portworx: Use Kubernetes to Run All Your Data Analytics Workloads
5 min read
By: Chris Rosen
Kubernetes and stateful apps
Kubernetes is the fastest growing open source project of all time, and for good reason. Developed initially at Google with learnings gained from running large-scale systems like YouTube in containers, Kubernetes enables IT teams to automatically manage the entire application lifecycle, no matter if those apps are running in the cloud or on-premises.
But Kubernetes itself can be difficult to install, update, and manage. That’s why IBM Cloud provides a managed Kubernetes offering, IBM Cloud Kubernetes Service. The Kubernetes Service delivers powerful container tools, an intuitive user experience, and built-in security for rapid delivery of applications that you can bind to cloud services related to IBM Watson, IoT, DevOps, and data analytics. As a certified Kubernetes provider, IBM Cloud Kubernetes Service provides intelligent scheduling, self-healing, horizontal scaling, service discovery and load balancing, automated rollouts and rollbacks, and secret and configuration management. The Kubernetes Service also has advanced capabilities around simplified cluster management, container security and isolation policies, the ability to design your own cluster, and integrated operational tools for consistency in deployment.
While the ability to automatically stand up and run a Kubernetes cluster is powerful, there has historically been a gap in the Kubernetes landscape: persistent storage and data management for stateful apps like databases, big data workloads, and AI/ML workflows. In other words, Kubernetes is great, but if you can only use it to manage stateless apps, then you are removing the bulk of enterprise applications that IBM has helped enterprises operate for decades.
IBM Cloud Kubernetes Service and Portworx
To provide a solution for these types of workloads, IBM has worked with Portworx, an IBM Business Partner, to offer a Kubernetes storage and data management solution that is designed to take advantage of IBM Cloud network-attached block storage as well as new software-defined-storage-enabled bare metal servers. Portworx is also a great option for IBM Cloud Private and Cloud Private for Data.
With integrations like Portworx and specialized bare-metal instance types to take advantage of cloud-native, software-defined storage (SDS), IBM has moved incredibly fast to secure a leading advantage among public cloud providers for Kubernetes workloads. As of this writing, none of the other cloud providers offer such bare-metal configuration options that are so uniquely tailored for DevOps-friendly cloud-native storage.
With the new SDS machine types for IBM Cloud Kubernetes Service and Portworx, customers can now best take advantage of Kubernetes along with IBM’s industry-leading data-centric set of applications and services, such as Watson Studio, Blockchain, MQ, DB2, and others. These and other applications can now run highly-available, across multiple zones, and with IBM’s highest degree of security built into its Kubernetes Service offering.
Adding all of this together demonstrates how well and how fast IBM moves in establishing leadership positions for its customers. IBM is thrilled to partner with Portworx, helping advance its cognitive solutions and cloud platforms.
Here is what Murli Thirumale, co-founder and CEO of Portworx, had to say about this joint offering: “IBM has been a leader in enterprise IT for decades, and with their moves into innovative technologies like containers, Kubernetes, and artificial intelligence, their customers have an enormous breadth of options for building and running their mission-critical applications. We’re proud to be an IBM Business Partner, enabling a growing number of customers to take advantage Kubernetes’ rich and mature ecosystem, including Portworx for data management of containerized applications.”
“Containers have redefined how today’s enterprises build and deploy applications, automate operations, and increase innovation. At the heart of these applications is data,” said Jason McGee, IBM Fellow, VP, and CTO, IBM Cloud. “As an industry-leading cloud-native storage solution, Portworx plays a key part in our customers’ ability to run mission-critical databases and analytics applications on Kubernetes in and across the IBM public and private clouds.”
Portworx’s flagship product, PX-Enterprise, is the cloud-native storage solution that enterprises, including 24 of the Fortune 2000, and government agencies rely on to dramatically cut costs and speed deployment for mission-critical containerized applications. With PX-Enterprise and container platforms like IBM Cloud Kubernetes Service, customers can run mission-critical containerized stateful services across multiple cloud and on-premises environments with dynamic provisioning, high-availability, class-of-service, encryption, snapshots and backups, and more, in a completely automated way.
In addition to powering stateful cloud-native applications, Portworx also delivers a significant ROI. With Portworx, enterprises can deploy stateful containerized applications while making far more efficient use of their hardware resources, cutting storage costs by 30 percent or more and compute costs by 40-60 percent compared to other containerized solutions. Recently, software company Aurea deployed PX-Enterprise to run 200 Kubernetes stateful pods per host—five times more than the limits imposed by Linux for traditional block storage devices. Because of these densities enabled by Kubernetes and Portworx, Aurea saved 60-90 percent on compute costs since they no longer had to over-provision VM instances to compensate for the limits of Linux storage networking.
Portworx PX-Enterprise 2.0 released on December 4, 2018.
If you have questions or concerns, engage our team via Slack. You can register here (https://bxcs-slack-invite.mybluemix.net) and join the discussion in the #general channel on https://ibm-container-service.slack.com.
Learn more about Portworx in the the IBM Marketplace