back to article All right, pet? Getting owlish about Hedvig

Trying to position Hedvig's software-defined storage in a market function way is next to impossible as the software is so flexible. And that could be a brilliant position for Hedvig to be in. The startup's distributed storage platform can provide storage for traditional workloads, new-style containerised workloads, hyper- …

  1. Yaron Haviv

    Nice marketing, but ..

    Today its hard to tell with all the Marketing spins

    but AFAIK Hedvig is not a scale-out NAS or object solution rather a scale-out Block storage, no different than EMC ScaleIO with standard Linux FS installed on that LUN

    Correct me if i'm wrong, what they do is implement (a single) NAS or Object protocol head over a distributed virtual block device (LUN), this means the FS or Object layer doesn't scale, and there is no (distributed) Metadata awareness, or built-in metadata search capabilities, which are very fundamental for modern object or scale-out NAS solutions


    1. HedvigKranz

      Re: Nice marketing, but ..

      Hey Yaron, thanks for the comment. Honesty first, I am a Hedvig employee, but on the technical side!

      So we're not a scale-out NAS for user file data, but we are a scale-out NAS for applications. I'm a massive fan of NFS for VMware and we work great there, in a proper scale-out way. We also do scale-out Object and as you say we also do block. But there's no standard Linux FS installed on top of it either.

      Our metadata is very much distributed, but yes the value of metadata is limited in a block world, it's more for our own operations than anything. File and Object metadata is fully distributed and responds how you expect it.

      In terms of the Scale-IO comparison, we offer some additional things such as deduplication, compression and something that I honestly think is killer, being active in multiple locations. We’re a clear path for customers doing hybrid cloud.

      1. Yaron Haviv

        Re: Nice marketing, but ..

        not sure whats the diff between "user" and "app" data, thought there is one definitions for a scale-out FS

        it boils down to simple questions:

        Do you process the (heavy) NFS protocol and Metadata ops in a single node or its fully distributes ?

        same for object protocols like S3, is metadata and protocol handling fully sharded ?

        if you get an object query to find the ones with attr1=X & attr2=Y how many nodes will participate in looking up the matching objects ? 1, 2, fully distributed ? is the Metadata & search co-located with data ?

        IMO doesn't make sense to run object like S3 on a virtual/scale-out block storage since sharding and transactions are in object boundaries and not sector boundaries, why all native object solutions use DAS

        you may have a wonderful product, and indeed the DR feature is a differentiation, but i think storage vendors should exercise transparency, having an object API option doesn't make one a true object storage


        1. HedvigKranz

          Re: Nice marketing, but ..

          I think this discussion could go on for some time as we appear to have different views on the subject and the Internet is an infamous platform for disagreements! :-)

          But simple answer to your simple question, we process the heavy NFS protocol and metadata operations across multiple nodes, as many as we have in the cluster. The same for object and block. And yes metadata is fully distributed across the cluster, I don't think sharding is the right term, but that's a pedantic technicality and our implementation of distribution has some similarities.

          I think we are being fully transparent, at no point do we say that we are a true object based storage system. In fact we try to actively point out that we are not as we think that introduces some compromises for block and file based requirements, but that's maybe another conversation. We have an object API and we don't think this makes us an object based storage system, but we do think it means our customers don't need to buy a dedicated object storage.

          1. Yaron Haviv

            Re: Nice marketing, but ..

            your answer somewhat contradicts the architecture your team presented at TFD10:


            you can see how the FS/Object "proxy" is placed in a VM and mounted on your distributed "virtual disk" (i.e. limited iops, metadata ops & space per namespace), all references to "metadata" are in the context of vDisk, not file/object metadata.

            wont make much sense to place a distributed file/object layer in those Proxies over a distributed block layer, that means doing clustering traffic and consistency twice, not to mention the proxies are stateless so every file/object IO commit need to issue multiple block layer IOs over the network (for journaling, data, and metadata ..).

            this design is cool for Block/VMware focused customers, small NAS or Object deployment or many such small file volumes, but is not a scale-out file solution, those are much harder to build.

            BTW re the "Hyper-Scale" slogan, non of those guys use a distributed "virtual disk" approach, they all use a distributed Object/file/NoSQL over DAS for the reasons i outlined, just like Cassandra :)

            we can continue the discussion offline if you want


  2. Anonymous Coward
    Anonymous Coward

    When people get OWNED in a polite way.

    Hey Yaron,

    I think you just got owned in a really nice way.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019