There’s good reason why NetApp is trying to fight outside its weight class and stake claim to an area where it has little relevance—the cloud. The old storage business is fighting a losing battle of trying to sell expensive storage boxes, while both new and established companies are shifting applications and workloads to the cloud. And with all those apps and services running in the cloud, there is a huge market opportunity in cloud data management, as we described in this recent post.
But to be sure, storing data is not the same as managing data. To make data available from a wide range of sources, for use by cloud apps and services, purchasing additional storage (virtual or otherwise) is not the answer. Over the past two years, more than 75 percent of businesses have been unable to surface the right data, says a 2017 survey by Forrester.[i] Why? Because they’ve been looking to legacy data storage vendors to solve the problem. But those vendors are not cloud-native, never think cloud-first, and strive to sell you more and more data repositories, not data management.
To judge the value of NetApp for managing cloud data, just take a look at what storage vendors like NetApp have done in the enterprise storage market. For example, the average enterprise today has more than nine copies of most pieces of information. Each copy is intended to support a singular purpose, such as disaster recovery, DevOps, incident response, archive, etc.[ii] That’s not a data management strategy—that’s a ‘sell more storage’ strategy. Now, imagine what old-line storage vendors are planning as the world moves toward the zettabyte era wherein, says the Cisco Global Cloud Index, the cloud data centers of 2021 will have over 13 workloads and compute instances available per server, using exponentially more data than today.
This is why purpose-built, cloud-native data management from Kmesh SaaS is superior to anything NetApp or its Cloud Volumes Service has to offer. To understand that fact more deeply, let’s look at the four core areas necessary for optimizing data management in the cloud to support cloud-based apps and workloads.
Comparison Area #1: Data Access & Orchestration
Kmesh SaaS transforms centralized data (file system data, unstructured data, NoSQL data) into distributed data, operating over multiple clouds, countries, and edges as a single global namespace. That single namespace operates without having to explicitly replicate the data, which means line-of-business leaders and application developers can leverage Kmesh SaaS to grant apps access to any data, in real time and as needed. Moreover, the data may continue to reside in its original location—no replication or additional storage purchases required. Users simply establish data access policies in the Kmesh dashboard and watch their apps and services access whatever data they need to get the job done.
In contrast, the NetApp Cloud Volumes Service has very little notion regarding data management and orchestration. It’s just storage replication in a different place. Rather than letting your data reside in place and access it where it lives (like Kmesh does), instead, you cannot touch that data unless you pay to have NetApp replicate it first. Imagine how much that pay-as-you-go service will cost as your apps and services use exponentially more cloud data and far more third-party Big Data sources over the next few years.
Comparison Area #2: Multi-Cloud Data Mobility and Management
When you use Kmesh Dataflows, you gain instant access to data sources across multiple clouds, freeing you to leverage the best of what each cloud vendor has to offer. Let’s say you want to run enterprise apps on Azure while leveraging the superior analytics capabilities of Google Cloud Platform. Simply register all your clouds in Kmesh, and your data gets moved and managed across different cloud vendors and cloud locations—that can include your own data center as well.
Don’t try multi-cloud with NetApp Cloud Volumes Service, because it doesn’t exist. If you read the Cloud Volumes Service FAQ carefully, Cloud Volumes Service can only perform its data replication on a single-cloud basis. It is simply not a multi-cloud service. How will that enable you to take advantage of the latest innovations in IoT and other multi-cloud technologies? Won’t you want to tap data sources and capabilities from different clouds? With NetApp, you would have to pull data from each cloud down separately, then perform your own data aggregation. Those are additional steps that will dramatically slow the business, lead to data gaps, and cost you three times more in data replication.
Cross-cloud data mobility is proving quite popular and is already a core requirement for many developers of apps and services. In fact, Kmesh recently published several customer case studies on our Resources page that demonstrate the use of Kmesh SaaS by customers that expressly need to orchestrate data in a multi-cloud fashion.
Comparison Area #3: Storage-Agnostic Data Orchestration
NetApp’s Cloud Volumes Service is a closed system that only works within its own technology stack, and it does not allow you to tap into other vendors’ storage systems…or even NetApp’s own storage systems. That’s right. The NetApp FAQ document for Cloud Volumes Service says this: “Currently non-disruptive migration from a NetApp storage system to Cloud Volumes is not supported.”
You know what that means—now you must pay to move everything to Cloud Volumes. That would make you dependent on a single vendor, cost some serious money to make the initial move, and leave you locked into one vendor moving forward.
Kmesh SaaS can tap into any type of data source, regardless of the location, hardware, storage system, etc. As a result, users of the Kmesh service instantly gain access to any data source simply by naming it in the Kmesh dashboard. If you have data stored on EMC, HP, NetApp or any other vendors’ systems, you can always tap into it with Kmesh.
Comparison Area #4: Performance for Compute-Intensive Workloads
NetApp touts the performance of their Cloud Volumes Service. Yet, the service cannot provide the performance needed to run compute-intensive workloads. A NetApp cloud volume is an NFS file system, and NFS files systems cannot provide the throughput necessary for workloads such as EDA, Big Data analytics, Cloud Bursting from on-premises, Machine Learning, etc. Again, a quick scan of the FAQ documents provided by NetApp reveals that their three levels of IOPS/throughput are all low-performance. In fact, their highest-performance tier, which they name “Extreme” tops out at a mere 128MB of throughput per TB. That won’t cut it for modern, compute-intensive workloads.
The Kmesh founders, knowing that many cloud workloads will be compute-intensive, chose to build Kmesh SaaS on top of the world’s highest performing file system, Lustre. Using Lustre, Kmesh can provide throughput ten times that of NetApp Cloud Volumes. In fact, Oak Ridge National Laboratory’s supercomputer runs Lustre and has more than 25,000 clients, over 10PB of storage, and aggregate IO throughput of 250 GB/sec. The performance of Lustre is unmatched, which is why Amazon recently launched their own Amazon FSx for Lustre service.
Kmesh versus NetApp in the Cloud: It’s No Contest
You want to move your apps and workloads to the cloud to take advantage of the flexibility, capability, and cost savings that cloud services have to offer. When it comes to managing and moving your data in the cloud, only Kmesh frees you to do it as fast and flexibly as possible. Go with an old-line storage vendor, and you’ll be paying the price in more ways than one.