Object storage slinger Scality has announced an open-source public cloud gateway and controller called Zenko.
Why "Zenko" as a name? It's allegedly Japanese for a benevolent, celestial fox that has occasional supernatural powers. Yes, we're none the wiser, either. Essentially, Zenko is based on Scality's implementation of the Amazon S3 API, aka S3 Server, from 2016. That open-source code has had 605,000 downloads from Github in Docker container form, and some fifty customers have asked Scality to develop it further, apparently.
Think of it as an S3 data mover, taking in information sent to it in S3 form from apps and pumping it out to on-premises storage resources – such as Scality's own RING – or off to the AWS and Azure public clouds. On that score, Zenko is a cloud storage gateway, but Scality prefers to call it an S3 storage controller, a data backbone or even a data fabric.
Data pushed to the public cloud is stored in the destination service's native format, such as the Azure Blob storage or, of course, Amazon S3. That means it could be accessed and used by in-cloud applications.
Zenko has a namespace that spans its backends, and you can search it, filter and select files, form groups of documents that match a filter term, and do something with the group, such as sending it somewhere using policies. This feature could be used to trigger actions based on the data, such as sending file copies to a second destination for disaster recovery, or aged files off to an archive. An enterprise's branches could use Zenko to automatically push local data meeting some criteria to a central site.
A Backbeat addition to Zenko will come in September and provide policy-based data workflows for things such as replication and migration.
September will also see the addition of a Clueso metadata search facility, using Apache Spark, to provide, yes, search functions.cCustomers can add their own searchable metadata tags to Zenko and so customise it for their own workflows and processes.
Scality has added governance and security aspects to the first version of Zenko. The search is Athena-style. Development plans include adding NAS backend target support using NFS and SMB. An enterprise version of the product could be developed with paid-for functionality above the base Zenko feature set – search and analytics, for example. For now, Scality has no monetisation plans.
The company will also package Zenko with its RING product in future.
There are no plans to add support for IBM's SoftLayer or the Oracle Cloud as target backends, but the Google Cloud Platform could be added. We're told Zenko users can write plugins to support their backend of choice.
The starting support for AWS and Azure plus coming GCP support shows that Zenko could be used to prevent lock-in to a single public cloud.
Word of mouth
Other products with some overlapping functionality include Primary Data – with its metadata engine for managing, tagging, placing and searching files – and SpectraLogic, with its BlackPearl product and S3 gateway used in archiving.
Rubrik and Cohesity have functionality similar to some Zenko features. However, as Zenko is free, customers who already use one or more competing and overlapping paid-for products could download it anyway; for text and dev use perhaps.
In El Reg's view the main acquisition and IPO phase of pure object storage startup players is more or less over. For a player like Scality, probably the leading object storage startup, the idea of adding a second string to its bow, or arrow to its quiver, and using the S3 API to provide a potentially one-stop shop for unstructured data management, access and search across the on-premises and public cloud worlds seems a good bet.
Look at this from Scality's point of view: 605,000 downloads says word-of-mouth regard for the software is high and there is a real need out there for Zenko's functionality. So let's fly this kite and see how strong the wind is up there, in the open-source, data moving and management atmosphere. ®