No Walk in the Woods: 10 Challenges— and Solutions—for DevOps in the Cloud

[Cloud Volumes ONTAP, Cloud Volumes Service for AWS, Advanced, 7 minute read]

In this entry of our series of DevOps on the AWS cloud, we take look at building a strong DevOps culture. 

The cloud is a great opportunity for DevOps to speed up go-to-market, control costs, and improve quality control over the lifetime of a software product.

However, positioning DevOps in public clouds like Amazon Web Services is no slam dunk. Cloud storage is not optimal out of the box, and teams need to optimize their infrastructure-as-a-service (IaaS) environment and adopt software tools to take full advantage of cloud benefits. Although this blog post discusses solutions for DevOps on AWS, similar challenges exist across the public cloud.

Challenge 1: Hire or train cloud experts

It’s not necessary to turn developers and IT into cloud architects, but configuring AWS IaaS takes some doing.

The three most popular options are to use consultants, to hire cloud experts to join the DevOps team, and to train existing team members. These approaches aren’t mutually exclusive: A single team might hire a consultant to facilitate the move from on the premises to the cloud, hire a cloud architect to manage the cloud environment, and cross-train operations IT to administer the cloud.

  • Hire consultants. This is an excellent option for DevOps teams that are doing their first migration to IaaS. AWS offers a list of partners on its Consulting Partners NetApp is an AWS consulting partner for DevOps and enterprise workloads.
  • Hire cloud specialists. This is a good option with multi-project, sophisticated continuous integration/continuous delivery (CI/CD) pipelines in the cloud, especially when containers are part of the project.
  • Train existing team members. AWS offers training and certification digitally and in person. There are also numerous independent training services for DevOps moving to AWS.

Challenge 2: Adapt to the IaaS environment

The cloud is highly flexible and configurable, but DevOps also needs to work with the realities of IaaS in the cloud. For example, planning AWS Availability Zones is a critical choice for global availability and disaster recovery. Load balancing for DevOps is another important issue, because it differs from on-premises load balancing. Cloud DevOps balancing depends on factors like protocol awareness, Kubernetes support, and APIs. NetApp® Cloud Volumes ONTAP® helps DevOps with these configuration issues by automating the virtual storage environment for each workload.

It’s also important to understand AWS service level agreements (SLAs). Understandably, AWS writes SLAs in its favor. Enterprises that are used to hammering out SLAs with corporate IT might assume agreement levels that are not, in fact, there. That’s  why, with Cloud Volumes Service for AWS, NetApp offers guaranteed SLAs for performance, availability, and durability

Challenge 3: Align DevOps practices to the realities of cloud storage

High-performance storage is crucial for DevOps in the cloud. For example, SQL developers need to provide sufficient performance and capacity for a wide range of SQL database services. On top of that, the team needs to manage high-performance/low-latency cloud storage and control the high cost of running workloads on AWS’ more expensive storage tiers.

Even if DevOps hires cloud specialists, they’ll be hard pressed to keep up with this level of manual oversight and maintenance. Cloud Volumes ONTAP comes with a large set of storage efficiency tools, including deduplication, compression, and data tiering to inexpensive object storage like Amazon S3. ONTAP can also manage persistent storage allocation for containers. And Cloud Volumes Service offers three dynamic tiers of service levels to support cost-effective performance and scale-up/scale-down.

Challenge 4: Continuously test in the cloud

Continuous testing is feasible only with automatic and rapid provisioning. Cost effectively spinning up data clones within seconds supports a wide variety of tests, including development, continuous integration, quality assurance, staging, and production. As soon as the test is finished, the same automation tools immediately deprovision the clones.

However, cloud providers, including AWS, don’t offer built-in cloning functionality. NetApp designed FlexClone® specifically for high-speed, efficient volume cloning.

Challenge 5: Improve cloud storage efficiencies

Cloud storage apps enable DevOps to optimize IaaS for large CI/CD pipelines. AWS offers several tools to help its users. For example, AWS AppSync simplifies application development with a flexible API that combines data from multiple sources.

However, AWS doesn’t always provide native data protection and backup for its services. For example, to back up Amazon Elastic File System (EFS), you need to subscribe to additional AWS services

A better option is to use NetApp Cloud Volumes ONTAP to protect your data. Also, the NetApp Cloud Volumes Service API lets you extend the service to other AWS automation services such as AWS Lambda, which lets users provision persistent storage for cloud workloads.

Challenge 6: Effectively move workloads to the cloud

Options for moving workloads to the cloud include the “six Rs”: rehost, replatform, refactor or rearchitect, repurchase, retire, and retain.

The most popular options are to rehost or replatform, where IT moves workloads to the cloud with no or minimal changes. DevOps should expect to make some changes to optimize AWS IaaS to their development environment. However, those changes can be relatively painless with NetApp Cloud Volumes Service NFS/SMB and multiprotocol support. (As NetApp puts it, there’s no need to fit a round app into a square cloud.)

Challenge 7: Ensure high performance for cloud workloads

Cloud storage performance is a concern for DevOps teams that own high-performance on-premises storage. However, with the right tools and optimization, cloud storage can attain the same or higher performance levels. At the same time, DevOps needs cost-effective solutions to avoid overspending on the cloud.

Cloud Volumes Service enables high-performance storage at low cost. The service achieves predictable performance of over 470K IOPS and 4.4 GB/s throughput from a single cloud volume. And storage efficiencies built into ONTAP can cut down on cloud storage footprint and costs by up to 70%.

Challenge 8: Low TCO, high ROI

Both technology and culture contribute to lower costs and higher returns. One fundamental cultural change for DevOps is to establish a lean cost culture through governance, adapting user behavior, and adopting software tools to optimize CI/CD pipelines.

Technology also contributes to low TCO and high ROI. DevOps teams moving to the cloud initially need to invest in training, configuration, and optimization. When the infrastructure is operational, monitoring and automated software cost effectively manage compute and storage.

NetApp Cloud Volumes Service supports low TCO and high ROI with low prices and storage efficiencies. For example, NetApp delivers dynamic service levels to suit DevOps’ changing needs. When the team needs basic performance and inexpensive storage, they can use the Standard service level at $1.10/GB per month in all AWS regions. When they need higher performance for a project, they can immediately switch to the Premium storage tier—and switch right back again at will. And when they need a high-performance environment, they can change to Extreme for the life of the instance.  

Challenge 9: Adopting tools for DevOps

DevOps is a fast-growing market for cloud providers and partners. Companies like Jenkins, Ansible, and GitHub support the compute side. For the storage management side, Cloud Volumes Service supports development and monitoring tools by integrating them with its RESTful API.

NetApp Cloud Services ONTAP also automates data movement between AWS tiers to balance performance with cost. Cloud Volumes Service does the same, and its storage-as-a-service feature provisions NAS-like shares and volumes, and mounts on demand. This relieves DevOps from having to build high-availability and highly scalable storage by offloading the task to NetApp.

Challenge 10: Innovate cloud apps

AWS promotes innovation for modern application development. The AWS framework includes creating an ownership culture within DevOps, using microservices and managed services, purpose-building databases for specific tasks, automating as many processes as possible, and prioritizing security.

NetApp Cloud Volumes ONTAP builds the critical foundation to support the AWS framework:

  • Make data requirements transparent to the application layer.
  • Support high availability with ONTAP native services.
  • Create a secure single sign-on to efficiently manage cloud data services.
  • Provide data protection and compliance with cloud data services.
  • Manage cloud costs with NetApp Snapshot™ copies, thin provisioning, data compression, deduplication, and data tiering.

In Summary

Moving DevOps to the cloud is not a stroll in the park. It takes will, training, and investment, now and in the future. The good news is that the strong benefits of cloud-based DevOps more than outweigh the cost, and NetApp makes it possible to achieve those benefits early on and for the lifetime of the data.

This is the final blog in our DevOps Series, here is our previous blog. For more detailed information on DevOps with NetApp on AWS you can view this webcast titled; Running Cloud-Native Applications On AWS? Also you can read about Cloud Volumes Service or start a free trial of Cloud Volumes ONTAP.

New call-to-action

Product Marketing Manager, Cloud Data Services

-