Longhorn Block Storage is well suited for most Kubernetes workloads where data storage is required, but when very high storage speed is essential, Longhorn Block Storage might not be the best solution. For those rare situations, we use local storage mounts. Longhorn Block Storage's ability to easily create/restore volume snapshots is a very frequently used feature among our dev teams because they can easily play multiple scenarios with the same data - modify data, restore it and modify it again.
When you need speed, it's FAST, especially for MS SQL databases. If you are having bottlenecks, you can spend your time finding it in the code because it's not Pure Storage. Love the data reduction and duplication. We can store so much more on this unit than the base physical size. Love the data snapshots. When you need to complete copy your entire environment for testing etc. Nothing is easier, a few clicks and you are done. Extremely helpful for testing and training.
Support - I've never had a call passed off to a different party that I then have to wait to get back to me. The person responding to my support ticket has been able to help me every time.
Non-disruptive upgrades. We were able to transition from a SAS based //m series array to a NVMe based //x series array with zero down time. This gave us greater performance and capacity with minimal disruption.
Simplicity - Pure Storage aims to automate the complicated steps of storage management with superior software. You know they are still happening under the surface, but the administrator does not have to deal with keeping the process just right with a given vendor's command set.
Licensing - All-in base licensing instead of expensive add-on licenses for additional features. There are very few exceptions to this rule.
ReadWriteMany Longhorn volumes are still using NFS (file-based) protocol in the core.
Using iSCSI as main protocol instead of FC ties Longhorn to Ethernet-based LAN which is in most architectures much slower that FC-based SAN.
Longhorn could implement S3 as alternative access protocol to its volumes.
Backups, and snapshots configuration could be configured at each volume-level by administrators (maybe from additional CRD object?), because currently is configured at storage-class level which is not granular enough.
It has so far been a very fast and stable product. We have had wonderful support when we have needed it and the account manager and engineer attached to our organization have been very responsive with any questions and concerns we have had.
Longhorn is mature software defined storage solution that is still developed and receive new functionalities. From the beginning every Longhorn volume have multiple (at least two) replicas, can leverage manual or automatic snapshots and backup to external S3 volume. Longhorn provides nice and clear GUI for administrators, but also can be managed from CLI.
From the day our first array was put in (2017) we have had very little issues using it the way we wanted to and none of the setup or processes we incorporate this into have ever been complex. All of the interfaces in the GUI are very intuitive and do not require any CLI experience to do what we need it to do.
We have never had an outage with Purestorage. Yes we have had 2 drives go out but replacements were installed so fast there wasn't any issues. We have had a NVRAM module replaced but were noticed from Pure they were sending a replacement because they were predicting a failure so pure was proactive preventing any issues. Knowing Pure is monitoring our device 24x7 gives us peace of mind.
The corporation has a very diverse data load and when we migrate from a hybrid storage to pure all-flash, significant gains have already been observed. Latency always remains below 1 ms regardless of the load and volumetry used. Application performance also depends heavily on whether or not the code is more permormatic and network access infrastructure. The product has been meeting expectations
Pure Support is timely and communicative. They are always ready to assist and very skilled at what they do, no matter the time of day. Even at 12:10 a.m. on New Years Day when your array has a hiccup that causes some errors in your environment, they will be on the phone with you in an instant. If they are unable to determine the cause right away, because things appear to be normal, they will take your environment's configuration, lab it up in their support environment and then research until they figure out what caused the problem. Even if it takes 3 months. True story.
Excellent training and really you don't need that much. We received training at the Pure offices and also on site when the product was installed. Simple things like how to login to the GUI interface, how to setup users, how to create volume and mount it. How to make a snapshot, how to copy it and mount it. Really is pretty self explanatory on how to do things with the GUI interface. We were expecting complicated base on other vendor products, we got super simple.
Great videos and documentation. There is a common theme with PureStorage, "Keep it elegantly simple". They have great support network and great support user groups. Documentation has been very helpful with providing auditors with methods and procedures on security and other ways the product works. Great documentation on setting PureStorage to be the most effective with VM's, SQL Server, and other products.
GlusterFS was first Persistent Storage solution used in our Kubernetes-based clusters. It is file-based what in some usages led us to many data corruptions. CEPH is object-based persistent storage which can be used as file-based Persistent Storage in Kubernetes. It is also is much more resource-hungry than other solutions including Longhorn. Dell PowerScale (or Isilon) is a hardware-software solution, that provides volumes that can be accessed by file-based NFS and CIFS protocols. Recently was added access to its volumes with object-based S3 protocol. Longhorn is in the middle. It is block-based, it is build on industry standards like iSCSI, performs very well on 10Gbit or faster LAN and commodity hardware (or in virtual machines)
Pure Storage FlashArray came up as being the most cost-effective of the lot plus the extras added like call-home support were very welcome. The competitors' options were a lot more expensive and the call-home support added a lot more to the already high costs.
We run what seems to me to be a crazy number of hosts (800), backed by 2 pure arrays. Around 400 hosts per array, including database servers, render farm, application servers, etc.
Longhorn is fully open source. One can try and/or use Longhorn for free even in enterprise and buy subscriptions only for environments that must be fully supported. We started with Longhorn in our lab environment and followed on through reference platform to non-production and production environments.
Longhorn subscriptions are not cheap, but its biggest advantage is that price-to-functionality ratio is very reasonable.
We have volumes with hundreds of gigabytes of data and these on Longhorn perform well in solutions where file-based GlusterFS volumes were corrupting data.
