The Pure Storage team will soon be heading to the ‘Imagine Intuitive’ themed Cisco Live event, held over January 28–1 February in Barcelona. We’ve worked closely with Cisco ...
Today, every business is powered by software, and having a DevOps strategy is key to innovation and developer productivity. Pure can help. A data-centric architecture from Pure Storage accelerates the Continuous Integration/Continuous Delivery (CI/CD) pipeline, enabling you to run.faster(); You can accelerate time-to-market, deliver higher quality code more frequently and reliably, and gain a competitive advantage!
To get a glimpse into how Pure can add value to your DevOps organization, checkout this lightboard video:
Empower your developers and accelerate time-to-market. With your CI/CD pipeline powered by Virtual Machines or Containers on a Data-Centric Architecture, you can turbocharge every stage of the development pipeline. Your developers can code, build, test, deploy, and monitor – with everything being faster! Let’s dive into each stage.
IMPLEMENT CI/CD ON CONTAINERS AND FLASH
We’ve done the hard work of integrating with Docker, Kubernetes Flexdriver and more to seamlessly plug-in to your containerized environment.
And, now, we’ve introduced Pure Service Orchestrator to deliver container storage as-a-service experience so that your developers can leverage containers for DevOps and build persistent containerized apps. Learn more about Pure Service Orchestrator.
It’s easy to forget the importance of making a development and/or test environment available to developers instantly, so they’re not having to wait on infrastructure. Now, you can deliver the same level of agility with on-prem Infrastructure-as-a-Service and Platform-as-a-Service that your developers might expect they can only get from the public cloud.
If you’ve bet your infrastructure on VMware, you’ll love this: we’ve made it easy to build your VM-based cloud. We’ve done the hard work to integrate with VVOLs and vRealize, so you can build a fully automated, self-serviceable cloud.
FlashArray//X delivers the industry’s simplest VVOLs implementation, unlocking “per-VM control”. And with vRealize, you can truly have fully automated provisioning via a service catalog that you expose directly to your end-users: VMs as a service delivered!
We also introduced a reference architecture for all-flash, on-prem platform-as-a-service that combines the power of Pure with Red Hat OpenShift container platform. It’s simple to setup… You can go from zero to PaaS in less than an hour with a fully automated setup. And, it empowers your developers and DBAs. They can provision data stores within seconds! Learn more about OpenShift PaaS.
Your developers no longer have to wait for a database, virtual machine or container to be spun up on their behalf. Reduce time to provision development environments from hours to seconds! And enable your developers to code.faster();
You can also build.faster(); by speeding up concurrent builds with parallelism. During the build and test stage, there are lots of file open operations, reads and simultaneous writes that occur. And, we can help. For example an auto manufacturer sped up its software builds by 3X, and its developers are now able to run 3X more concurrent builds by putting its software build farm on FlashBlade. With 1000+ developers, the flexibility to test more versions of software in parallel unlocks the path to faster innovation. Thanks to all-flash speed for making this possible.
You can start small, and have plenty of headroom to scale performance as your concurrent builds grow. As an example of performance scalability, here’s a performance snapshot of the last 30 days of a real-world multi-blade FlashBlade system at a customer site. Scaling to multi-GB/s read bandwidth with simultaneous writes is now possible.
The next step where the Pure Data-Centric Architecture adds value is testing faster. You can refresh real-world production data into development instantly, so developers are working on current data. With our snapshots and replication, delivering production data to developers is effortless. You can create a library of zero-cost point-in-time snaps that can be delivered to developers using a variety of automation tools.
For example, a financial services company has built a ‘1-click’ pre-live environment for its developers, enabling them to test with latest production data, and thereby have greater confidence in their code quality when deployed into production.
Another important use case that Purity Snapshots enables is rapid spin-up of troubleshooting environments that would otherwise take hours to provision. By doing this, several customers have dramatically reduced mean-time-to-troubleshoot and fix customer issues. And, that means happier customers.
The massive parallelism of FlashBlade also enables a novel use of real-time analytics with technologies like Kafka and Spark for DevOps. Automated testing is integral to the Continuous Integration cycle and, as the code base scales, it’s common to end up with 1000’s of automated tests that run continuously. Of course, failures are inevitable, but the time required for manual failure analysis by developers can quickly add up and chew up precious cycles that would otherwise have been spent on developing code. Our own internal developer FlashBlade team used real-time analytics to dramatically reduce the distinct test failures that require a developer’s time. Now, a single developer can perform the root cause analysis, which the FlashBlade team estimated would have otherwise taken approximately 20 software developers’ time. Checkout this blog by Ivan Jibaja from our FlashBlade team.
Continuous Delivery for stateful apps requires data to be attached automatically as new virtual machines with new software code are deployed into production. And developers need the flexibility to access infrastructure as code to make this happen.
At Pure, we have an API-first strategy, and we’ve invested in pre-built integrations and SDKs with the common orchestration tools including Ansible, Puppet, SaltStack, ServiceNow, Python, PowerShell and more, so your developers can truly build automated and repeatable processes for continuous delivery.
An online gaming company that depends on continuous delivery deploys 3000+ new virtual machines to Dev and Test on a daily basis for their stateful app, and its developers use Pure as infrastructure-as-code via Ansible to attach data. Watch this video to learn more.
Last, but not least, you can monitor.faster(); Turbocharge analytic tools for service reliability monitoring. Not only can you build a full analytic pipeline, the massive all-flash speed means you can achieve near real-time analytics. We even offer a Splunk App to enable seamless monitoring of your fleet of FlashArrays.
We’re super excited about engaging with DevOps professionals! With Pure Data-Centric Architecture, we can help you empower your developers and deliver a competitive advantage for your business:
With Pure, your development organization can run.faster();