WoodITWork.com

Cloud Field Day 2, part of the Tech Field Day family of events is happening in San Francisco and Silicon Valley, from 26-28 July and I’m super excited to be invited as a delegate.

We are hearing from a number of companies about how they cloud!

Scality has been a previous Tech Field Day presenter.

Scality is one of the new storage companies leveraging the AWS S3 storage API standard to create new enterprise storage options beyond your typical block and file store. S3 is object storage which is all about scale, built to store billions of objects or massive petabyte sized files or stores.

Scality already provides an open source implementation of the AWS S3 API called Scality S3 Server. Interestingly it is packaged as a Docker container so can leverage the benefits of Docker such as the same deployment mechanism from a developers laptop to being deployed in production and further scaled out via Docker Swarm.

Scality RING is the enterprise friendly version of S3 Server for more critical workloads with the usual enterprise feature requirements of security, support, availability, etc.

AWS S3 is all great but some enterprises aren’t willing to store everything in a public cloud. There may be (often unfounded) security concerns or more valid concerns about bandwidth usage, data gravity and cost. If you have PBs of on-prem storage for your media files, x-rays, satellite images etc. you would love the ease of use of the S3 API but accessed locally. Scality can provide this S3 API on-prem as well as the replicated, highly available storage infrastructure running on standard x86 underneath. Having S3 locally also allows your developers to test functionality locally for things that may eventually access AWS S3.

Zenko

Scality has now announced Zenko which is an open source multi-cloud controller and this is what I expect we’ll hear more about at Cloud Field Day.

Zenko is a multi-cloud object storage management solution. This allows you to now connect your on-prem S3 storage with native public cloud storage options. If S3 Server was about providing your on-prem version of AWS S3, Zenko is about creating an abstraction level across multiple private and public clouds at the object storage layer. If a file is sent to Zenko to be stored as an S3 object, Zenko can send it onto an on-prem S3 Server or out to AWS or Azure based on whatever policies you need saying where you want your objects stored.

Importantly the public cloud storage is still stored in the public cloud’s native format so you can read it directly without having to go through Zenko from the public cloud. This shows it is not a gateway. For Azure this adds an interesting proposition. You can store objects in Azure Blob storage but access it using AWS S3 API. This could be very useful as you still have native Blob storage access from other Azure services but now also S3 access from Amazon or anywhere else.

Zenko provides a single namespace that spans whatever cloud is underneath so you search and filter requests across all clouds, private and public. Nice. This is very important as due to it being an object store you don’t have logical folders to group files so filters are what can be used to group objects which you use to search. I gather the index and metadata to do the search are stored as part of Zenko which is itself stored in S3 Server so can benefit from the speed and scale of the distributed object store.

Zenko can also act as an object bit shifter for cross cloud lifecycle management, so you can use policies to move data between clouds. Think of a retention policy to move stale data somewhere else or enterprises wanting to sync shared data between offices or to have a locally accessed copy or back things up into a central location. With Zenko you are no longer “limited” to using your own object store but can now splash and play in the big pools but all visible via that single namespace.

I’m interested in how Scality sees the growth and use case of non public cloud object storage. Other than massive media archives, what use cases are they coming across. Are they seeing enterprises move away from file/SMB/NFS access to object? What interesting benefits do object stores give for enterprise use above being a massively scalable file dumping ground? How do they see the future playing out in a multi-cloud world and how do enterprises decide on which cloud to store their data?

Looking forward to hearing more from Scality.

Gestalt IT is paying for travel, accommodation and things to eat to attend Cloud Field Day but isn’t paying a penny for me to write anything good or bad about anyone.