BlueXP Blog

Enterprise Workloads with Cloud Volumes ONTAP on Google Cloud

Written by Yifat Perry, Technical Content Manager | Jul 21, 2020 10:59:23 AM

According to the Flexera 2020 State of the Cloud Report, 57% of enterprise workloads and 53% of enterprise data will be in a public cloud within the next twelve months, representing YoY increases of 9% and 8% respectively. In addition, companies are expressing a greater willingness to migrate sensitive consumer and corporate data to the cloud.

These trends mean that more and more organizations are facing the challenges of enterprise-grade data storage management using cloud-native services, including:

  • Migration on-premises workloads to the cloud
  • Managing application dependencies in the cloud
  • Data mobility across the multicloud and hybrid environments that predominate the enterprise landscape
  • Agility, which is critical for continuous delivery of new features, products, and services
  • Performant file-sharing, given that ~80% of enterprise data is stored in files
  • Cost management, with optimizing costs cited for the fourth year in a row as the top initiative for cloud-mature enterprises

In this blog post we explore in more detail the requirements for enterprise-grade data in the cloud, and describe how NetApp’s Cloud Volumes ONTAP data management platform integrates with Google Cloud Platform (GCP) cloud-native services to provide mature data storage for enterprise workloads.

Enterprise-Grade Data Requirements in Google Cloud

Enterprise stakeholders—from executives and employees to customers and partners—require data in general and file shares in particular to be continuously available, with high-latency throughput, and securely protected. While Google Cloud’s services and offerings can help address many of these concerns, achieving enterprise data management requirements can still pose a challenge, for the following reasons:

  • Availability: Ensuring undisrupted data and file availability on GCP is the user’s responsibility, not Google’s. Meeting zero data loss (RPO=0) and minimal downtime (RTO<60 seconds) SLAs involves complex and costly configurations that require high levels of cloud expertise and can be error-prone.
  • Protection: When it comes to data protection on Google Cloud users get the building blocks to assemble their custom data protection solution. This might be a considerable task in enterprise deployments, which require the strictest security and protection.
  • Accessibility: Structured enterprise data such as files and databases are used by Linux/Unix and Windows workloads, and require support for both NFS and SMB/CIFS protocols. The Google Cloud Filestore service does not provide multi-protocol access.
  • Cost-performance optimization: High-latency data storage resources are performant but costly, while low-latency resources are cost-effective but slow. Google Cloud offers both, but doesn’t provide automated data tiering between those tiers.
  • Monitoring and control: In order to address governance and compliance requirements, it is important that data policies and best practices be applied and monitored consistently throughout the enterprise ecosystem. The enterprise must establish uniform monitoring and control planes across cloud interfaces.
  • Cost management: As a cloud vendor, Google Cloud makes it less expensive to operate than in traditional data centers. However, enterprise-level deployments can find themselves facing considerable bills at the end of each month for these services.

Getting More from Google Cloud: Enterprise Workloads with Cloud Volumes ONTAP

NetApp’s Cloud Volumes ONTAP is a proven enterprise-grade cloud data management solution that provides single-pane monitoring and control of enterprise storage, and mobility across complex multicloud/hybrid architectures. Cloud Volumes ONTAP runs as an instance on Google Cloud Compute Engine and Persistent Disks, but is also available for use with Azure and AWS.

In this section we’ll see how Cloud Volumes ONTAP complements GCP-native services to provide mature, enterprise-grade data and file management.

Migration and Mobility

Migrating enterprise workloads to the cloud continues to present significant challenges. Figure 1 shows how the 750 respondents to the Flexera 2020 State of the Cloud survey ranked those challenges. The overarching message is that, if not planned and executed properly, cloud migration can result in performance issues as well as significant cost overruns.

Figure 1: Cloud Migration Challenges. Source: Flexera 2020 State of the Cloud Report 

One of the biggest migration challenges is adapting legacy enterprise workloads to object-based cloud storage. Rearchitecting applications consumes considerable time and budget, with the risk that the modified application will not meet its performance SLAs once in the cloud. Because Cloud Volumes ONTAP for GCP abstracts the file and data storage infrastructure from the application layer, it supports seamless lifting and shifting of enterprise workloads to the cloud.

For existing NetApp ONTAP users, NetApp SnapMirror® data replication provides an efficient, seamless enterprise mobility solution that is infrastructure-agnostic—from flash to disk to cloud. SnapMirror only synchronizes data blocks that have changed since the last synchronization cycle.

And for new NetApp users, Cloud Volume ONTAP for GCP can rely on NetApp Cloud Sync service to replicate and synchronize data transfers seamlessly, quickly, and securely between any data repository and Google Cloud storage.

High Availability

Undisrupted data availability is critical to business continuity in general and to compliance in particular. In some cases, enterprises are subject to regulatory frameworks that apply the same stringent data loss (RPO) and downtime (RTO) metrics to cloud-based data as to on-premises data storage.

Like any infrastructure, cloud platforms experience downtime due to disk, instance, network, or availability zone failures.

In an upcoming release, Cloud Volumes ONTAP for GCP will be able to be deployed in a high availability configuration in which dual redundant nodes, which can be located in different availability zones. Data is kept in constant sync between these two nodes, and should a failure occur in one node, failover to the other is automatic with zero data loss (RPO = 0) and minimal downtime (RTO<60 seconds). When the failure has been remediated, automated failback processes re-synchronize and activate the recovered node.

Enterprise Data Protection

NetApp Snapshot™ copies are point-in-time, read-only views of Google Persistent Disk volumes.  Based on technology that manipulates block pointers rather than copying data blocks, NetApp Snapshot copies are very fast, and have near-zero storage footprints.

Snapshot copies are the backbone of Cloud Volumes ONTAP for GCP’s built-in enterprise-grade data protection capabilities, including backup/restore and disaster recovery (DR) secondary sites with automated failover and failback processes.

Accessibility

Cloud Volumes ONTAP for GCP supports both the NFS and SMB/CIFS file server network protocols. This protocol flexibility is important because many enterprises use both NFS or SMB/CIFS with their Linux/UNIX and Windows file servers.

Unified Management and Automation

NetApp Cloud Manager is a centralized control and monitoring platform for agile and reliable data management across hybrid and multicloud architectures, as well as across multiple GCP accounts. Cloud Manager provides both a simple wizard-based interface and a RESTful API to walk you through the set up and automation of provisioning volumes and disk aggregates, setting up new storage systems, creating snapshot schedules, configuring automated backup restore and DR failover/failback processes, provisioning clone environments, and creating new file shares.

Scalability

Cloud Volumes ONTAP for GCP has three licensing tiers (Explore, Standard, Premium) as well as an option to design your own license (BYOL). The Premium license as well as the BYOL model support up to 368 TB of storage on high-performance instance types.

In addition, Cloud Volumes ONTAP supports automated data tiering in order to cost-effectively manage dynamic data storage and access requirements. For example, Cloud Volumes ONTAP automatically detects active and inactive data and switches seamlessly between high-performance and capacity storage tiers.

Costs

In addition to the automated active/inactive data storage tiering and near-zero-footprint snapshots described above, Cloud Volumes ONTAP for GCP lowers enterprise storage costs through data deduplication, compression and compaction as well as through thin provisioning that allocates storage only when required

Depending on the use case, Cloud Volumes ONTAP for GCP can save as much as 70% of cloud storage costs.

Kubernetes Integration

Kubernetes is an open-source container orchestration platform that was originally designed by Google to meet its own internal container orchestration needs. Today it is the leading solution for deploying and managing containerized applications at scale.

Stateful containerized applications require the mounting of persistent volumes on Kubernetes clusters. NetApp Trident is an open-source enterprise Kubernetes storage provisioner that allows native Kubernetes manifests to provision persistent volumes via Cloud Volumes ONTAP. This gives your Kubernetes persistent volumes all the scalability, data protection, and powerful storage efficiencies Cloud Volumes ONTAP offers GCP.

Conclusion

Cloud Volumes ONTAP addresses many of the key challenges that GCP users face when they migrate and deploy enterprise workloads to the cloud.

Cloud Volumes ONTAP for GCP delivers an array of benefits while significantly lowering cloud storage footprints and costs. To find out how all of this works in practice, read this selection of case studies on enterprise deployment on Google Cloud with Cloud Volumes ONTAP.