At GO 2019, Commvault Systems announced the availability of Metallic, a new SaaS-based solution for data protection. The company will be hoping that the Metallic service can expand the business into the mid-market and bring in a whole new set of customers with diverse and hybrid data protection needs.
The Metallic offering is billed as a new service built on Commvault intellectual property. Potential customers can sign up online and get protection for three main categories of data:
- Metallic Core – traditional virtual machines, SQL Server databases, file servers and virtual instances in either AWS or Azure.
- Metallic Office 365 – protection for SaaS-based applications that includes Exchange Online, SharePoint and OneDrive.
- Endpoint – laptops and desktops running Linux, macOS or Windows.
Pricing is based on the volume of data protected per month for Core and per user for both Office 365 and Endpoint (inclusive of all data volumes). Core users can either pay for inclusive storage capacity on Azure (where Metallic currently runs) or use their own cloud storage. Core can also write to local on-premises storage (such as an object store or file server) for fast restores.
Commvault has taken the interesting approach of making Metallic a channel-led sale through partners. Clearly, getting traction and awareness for a new SaaS-based solution isn’t easy, so selling through the channel speeds this process up. Potential customers looking to trial the solution will be directed to a partner once the trial completes. Note that the solution is currently for US-based businesses only.
Metallic has been brought to market in only nine months. I chatted to David Ngo and other members of the team at Commvault GO to find out how this had been achieved. Metallic was run as a separate “start-up” project within the company, with access to Commvault IP, while effectively developing the solution from scratch. It’s clear that significant work was put into building a user interface and understanding customer workflow – essential for any SaaS solution.
The Metallic business will be run as a separate P&L within Commvault, which should mean we get to see the sales results in financial reporting. SaaS applications have distinct advantages over traditional licensing because the customer generally comes back month after month to use the service. Sales teams only have to find new business to add growth, however, in backup there is almost always automatic growth in demand because customers expand their primary storage footprint on a regular basis.
- Unpicking the Commvault Acquisition of Hedvig
- Backup is Your Responsibility – Even in Public Cloud
- Backup as a Service
The “stickiness” of backup makes the SaaS model even more attractive, as customers typically don’t want to invest in the pain of moving to a new backup solution. Of course, this can work in reverse, because customers may be slow in adopting Metallic for the same reasons.
I’m always interested in understanding how solution providers build and operate SaaS-based solutions. While we don’t need to know how the “sausage factory” makes the sausages (as long as the end product is good enough), the operational efficiencies (and ingredients) of the factory are important.
This applies equally to SaaS-based data protection. Simply lifting and shifting Commvault Complete into the public cloud would be a terrible approach for the company to take. Complete was designed for on-premises deployments and simply wouldn’t scale operationally to hundreds of thousands of installations.
This model wouldn’t be practical for Commvault either, as it would mean running virtual instances permanently and not taking advantage of cloud-native infrastructure applications (such as key-value stores and structured databases). Fixed infrastructure removes any ability to easily deal with transient demand, such as peaks or troughs in backup traffic. Not using native services precludes the ability to offload some infrastructure maintenance tasks to the cloud service provider.
In analyst briefings, Commvault representatives were at pains to highlight that this wasn’t just a “lift and shift” approach with a nice GUI front-end. The solution has been developed using Commvault intellectual property but isn’t simply Complete running in virtual instances.
The proof of this will come as the service scales out. Commvault will want to onboard as many customers as fast as possible, so efficient back-end support of the solution will be essential. This will undoubtedly be an area to watch.
For more details on Metallic today, this Tech Field Day presentation recorded at GO provides some great additional background.
Work in Progress
Of course, a new solution to the market can’t provide all features from day one. Commvault has chosen the most likely subset of data protection capabilities that meets mid-market requirements. This subset is essentially VMware customers running traditional Microsoft applications and some aspect of SaaS services.
The next challenge for Commvault will be to widen the appeal of the offering without creating an overly complex solution. The IP is all there to be used, but there are some pieces outside of that which need to be solved.
- Client-side de-dupe – currently de-duplication is done on the back-end. There are bound to be some efficiencies to be gained with client-side de-duplication.
- Initial seeding – some customers will want the option to seed the backup solution using the “Snowball” type approach of shipping data on physical media. This will especially be the case for larger customers.
- DR capability – this could be recovery into the public cloud or to a partner co-location environment.
- Analytics – it’s not clear if Commvault Activate will initially support Metallic (I think it doesn’t) but this will be an opportunity for the company to upsell.
Possibly the biggest long-term benefit for existing Commvault customers could be the ability to support existing data backed up in on-premises solutions. As businesses move towards a more diversified landscape, we will see data assets in SaaS, IaaS, edge and on-premises locations. Today, no vendors offer the ability to ingest legacy or old backups into SaaS offerings. If Commvault could deliver the option to backup and restore from existing on-premises data (and reduce the on-premises footprint), then Metallic could be made very attractive indeed.
Let’s also not forget about the Hedvig acquisition. Commvault is touting the integration of primary and secondary storage as one of the reasons for purchasing Hedvig. With Metallic as an integration point, Commvault could offer automatic data protection on the Hedvig Distributed Storage Platform without the need to deploy additional hardware or software. This would literally be a fully integrated solution.
The Architect’s View
Commvault is by no means first to the market with SaaS-based backup. Vendors are offering SaaS services using proprietary data centres or the public cloud, with DR capabilities too. The Metallic team has a lot of work to do, in order to lead the pack in this market.
With that in mind, let’s not forget this solution was developed in nine months and has the benefit of 20+ years of IP behind it. The success of Metallic will be based on the ability for Commvault to bring new features to the platform quickly and without adding complications.
One last thought. How does Microsoft figure in all of this? Metallic runs on Azure and focuses on Microsoft applications. Although Commvault hasn’t been forthcoming about whether Metallic could be “white labelled”, I can see a scenario where Commvault becomes the preferred data protection integrated solution within Azure and on-premises for Microsoft customers. There’s an idea to think about in the future…
Copyright (c) 2007-2019 Brookend Ltd, no reproduction without permission, in part or whole. Post #fb1f.
Disclaimer: Chris was invited to Commvault GO 2019, with Commvault and GestaltIT covering travel and accommodation. There is no requirement to blog or produce any content from the event and no content is reviewed before publication.