Hybrid Cloud Storage Solutions Resources

For the cloud storage infrastructure reviewers on G2 who have worked on bolting cloud capacity onto an existing SAN: which hybrid cloud storage providers give the most seamless integration with an existing on-prem SAN so a company can extend storage capacity to the cloud without replacing its current infrastructure?

I want to start a discussion on this, because nobody wants to rip out working infrastructure just to get more headroom. I've been reading through the hybrid cloud storage solutions category and three routes stand out, each with a different answer:

• AWS Storage Gateway is the vendor-neutral route: a virtual appliance that sits beside your SAN, presents iSCSI volumes to your servers, keeps hot data cached locally, and pushes the rest to S3. My question: how did cache sizing hold up once production traffic hit it?
• NetApp ONTAP is the route if your existing arrays are already NetApp: tiering cold data to cloud object storage is built into the array itself, no extra box. For those running it: how transparent is the recall, really, when someone suddenly needs data that's been tiered away?
• CTERA works at the file level rather than block, so it's less a SAN extension and more a way to move your file shares to a cached edge appliance backed by cloud object storage. If that's the part of your estate that's actually full: did it let you retire file servers, or just add another layer?

One honest flag: Azure StorSimple still appears in this category's data, but Microsoft retired it at the end of 2022, so I wouldn't start there.

Which route did I get wrong, and what's missing that should be on this list?

Hoping to get a read from G2 software researchers in this space. The question is which hybrid cloud storage tools best handle large unstructured data, meaning the files, images, and logs that pile up fast and never fit tidily in a database. 

Structured data has an obvious home, but once you're into millions of objects and media files, the tool choice really shapes cost, retrieval speed, and how much time you lose to management overhead. Object storage keeps coming up as the answer, so I looked at what reviewers say about the platforms built for exactly this.

• NetApp StorageGRID: Reviewers describe it as software-defined object storage built for large volumes of unstructured data, with tiering across NVMe, SSD, S3, Azure Blob, and Google Cloud, a single namespace that can span multiple data centers, and metadata-driven lifecycle policies that move data by cost or compliance. The common gripes are a high overall cost and some tasks feeling clunky to set up.
• Cloudian HyperStore: This is the one software and IT reviewers point to for keeping massive object data on their own terms, S3-compatible so apps connect without changes, and built to scale toward exabytes as IoT and AI workloads pile up, with object lock and encryption for protection. A couple note the monitoring interface could be more intuitive.
• zStorage (by Zadara): Reviewers like that it handles object, file, and block from one fully managed, pay-as-you-go platform, with strong read and write speeds and the option to keep data on a dedicated array instead of public cloud. It's managed 24/7, so for a team buried in file and log growth, is offloading the storage operations worth leaning on a managed service, or would you rather keep hands-on control?

For teams that already live in this kind of data, which of these kept retrieval fast and costs sane once you were past a few hundred terabytes? And has anyone run images and logs on the same object platform happily, or did you end up splitting hot log data off onto something separate?

Hoping to get a read from G2 software researchers in this space. The question is which hybrid cloud storage tools best handle large unstructured data, meaning the files, images, and logs that pile up fast and never fit tidily in a database. 

Structured data has an obvious home, but once you're into millions of objects and media files, the tool choice really shapes cost, retrieval speed, and how much time you lose to management overhead. Object storage keeps coming up as the answer, so I looked at what reviewers say about the platforms built for exactly this.

• NetApp StorageGRID: Reviewers describe it as software-defined object storage built for large volumes of unstructured data, with tiering across NVMe, SSD, S3, Azure Blob, and Google Cloud, a single namespace that can span multiple data centers, and metadata-driven lifecycle policies that move data by cost or compliance. The common gripes are a high overall cost and some tasks feeling clunky to set up.
• Cloudian HyperStore: This is the one software and IT reviewers point to for keeping massive object data on their own terms, S3-compatible so apps connect without changes, and built to scale toward exabytes as IoT and AI workloads pile up, with object lock and encryption for protection. A couple note the monitoring interface could be more intuitive.
• zStorage (by Zadara): Reviewers like that it handles object, file, and block from one fully managed, pay-as-you-go platform, with strong read and write speeds and the option to keep data on a dedicated array instead of public cloud. It's managed 24/7, so for a team buried in file and log growth, is offloading the storage operations worth leaning on a managed service, or would you rather keep hands-on control?

For teams that already live in this kind of data, which of these kept retrieval fast and costs sane once you were past a few hundred terabytes? And has anyone run images and logs on the same object platform happily, or did you end up splitting hot log data off onto something separate?