Home
Join

12 Replies

  • Hey OP! Hopefully, I may be able to help provide some possible suggestions. I will say, after reading through your post, replacing this kind of infrastructure is coincidentally what the Scale team built Scale Computing for. With Scale clusters at both sides, you would be able to radically streamline your processes, eliminate the SAN arrays entirely, eliminate your layer2 iSCSI SAN subnets, utilize the built in replication between the sites and do it all for less than you would spend on a refresh of his infra. Likewise, you could also leverage Acronis to send additional backups to literally ANY cost effective storage location (up to 5 simultaneous storage locations).

    If you happen to be interested, I'm happy to help get you connected with the team. Just send over a private message my way!

    Or, if you'd like to contact directly, you can do so here: https://www.scalecomputing.com/contact

    Good luck in your search! And, feel free to reach out.


    1 of 2 found this helpful thumb_up thumb_down
  • Kelly For Scale Computing wrote:

    With Scale clusters at both sides, you would be able to radically streamline your processes, eliminate the SAN arrays entirely

    What is this? Hyperconverged nodes? By "eliminate the SAN arrays" you'd be referring to the two brand new 100 TB usable Nimble units that are already bought and paid for? I'm not sure that these are what I would've chosen had I been here when that decision was made, but that's what I have to work with for now. Do you offer a solution that sits on top of our existing technology?

    Spice (1) flagReport
    1 found this helpful thumb_up thumb_down
  • If you have a SAN already (Nimble) just use it for shared storage and configure SQL Server FCI, guest VM cluster for your file server and some HA with your proprietary app.

    Here are some links:

    https://docs.microsoft.com/en-us/sql/sql-server/failover-clusters/windows/always-on-failover-cluster...

    https://docs.microsoft.com/en-us/previous-versions/windows/it-pro/windows-server-2012-r2-and-2012/dn...

    https://docs.vmware.com/en/VMware-vSphere/7.0/vsphere-esxi-vcenter-server-703-setup-wsfc.pdf

    For DR and VM backup stick with Veeam.

    Hope this helped :)

    Spice (1) flagReport
    1 found this helpful thumb_up thumb_down
  • KOOLER (StarWind) wrote:

    If you have a SAN already (Nimble) just use it for shared storage and configure SQL Server FCI, guest VM cluster for your file server and some HA with your proprietary app.

    Thank you. Though, I suppose I should've specified that I am wanting failover clusters to be stretched between the primary and secondary sites. It wouldn't be just DR sitting at the secondary data center. This way, our primary datacenter could go offline with little more than a hiccup. We wouldn't necessarily be in a disaster situation unless something affected data at both sites (e.g. ransomware).

    Was this post helpful? thumb_up thumb_down
  • SoggyFrog wrote:

    Kelly For Scale Computing wrote:

    With Scale clusters at both sides, you would be able to radically streamline your processes, eliminate the SAN arrays entirely

    What is this? Hyperconverged nodes? By "eliminate the SAN arrays" you'd be referring to the two brand new 100 TB usable Nimble units that are already bought and paid for? I'm not sure that these are what I would've chosen had I been here when that decision was made, but that's what I have to work with for now. Do you offer a solution that sits on top of our existing technology?

    Hey OP - Just asked the Scale team, and they mentioned (with your existing Nimble equipment) you could use Acronis’ Convert to VM and send the conversions to a Scale target, either single node or cluster - Simple and quick failover process. Any chance you'd be interested in chatting with the team further on the matter?

    Was this post helpful? thumb_up thumb_down
  • I'm doing some more digging into the capabilities of these Nimble units. As an alternative to snapshot based replication, they have synchronous replication, which I think would work for stretching failover clusters. We'd need to upgrade our intersite links first.

    And I need to verify that snapshot-based backups (e.g. via Veeam) are compatible with synchronous replication enabled. It's apparently possible, though I'm not sure if it's supported by Veeam yet (plain synchronous vs "Peer Persistence").

    If I could stretch clusters that way, we wouldn't have to license any virtual SAN technology. Then I suppose I would pull primary backups off the secondary site, pull CDP for DR off the primary site.

    Was this post helpful? thumb_up thumb_down
  • SoggyFrog wrote:

    I'm doing some more digging into the capabilities of these Nimble units. As an alternative to snapshot based replication, they have synchronous replication, which I think would work for stretching failover clusters. We'd need to upgrade our intersite links first.

    And I need to verify that snapshot-based backups (e.g. via Veeam) are compatible with synchronous replication enabled. It's apparently possible, though I'm not sure if it's supported by Veeam yet (plain synchronous vs "Peer Persistence").

    If I could stretch clusters that way, we wouldn't have to license any virtual SAN technology. Then I suppose I would pull primary backups off the secondary site, pull CDP for DR off the primary site.

    Not sure what is there specifically on this topic but I definitely recommend you check the InfoSight documentation library. Lots of stuff there and I bet there are technical white papers answering what you're trying to figure out.  Here's a link (assuming you already have access to InfoSight). 

    In addition, there's is deep integration between Veeam and HPE Nimble Storage that could help address what you're trying to do. 

    And yes, Nimble Storage has Sync replication that we call Peer Persistence (came from the name we called it with 3PAR). Not sure how the integration with that is with Veeam but asking my tech marketing lead that does Veeam and HPE Storage integration. Here's a good lightboard video showing Peer Persistence. https://youtu.be/-IFg7iG1F8I

    Spice (1) flagReport
    2 found this helpful thumb_up thumb_down
  • SoggyFrog wrote:

    KOOLER (StarWind) wrote:

    If you have a SAN already (Nimble) just use it for shared storage and configure SQL Server FCI, guest VM cluster for your file server and some HA with your proprietary app.

    Thank you. Though, I suppose I should've specified that I am wanting failover clusters to be stretched between the primary and secondary sites. It wouldn't be just DR sitting at the secondary data center. This way, our primary datacenter could go offline with little more than a hiccup. We wouldn't necessarily be in a disaster situation unless something affected data at both sites (e.g. ransomware).

    1) You can do stretched cluster with DFS-R. While it's not recommended (and it's pretty horrible...) for the every day use, DR scenario when second replica is never touched most of the time is OK with DFS-R. You have to watch out the "split brain" scenario however.

    https://docs.microsoft.com/en-us/windows-server/storage/dfs-replication/dfsr-overview

    2) Alternatively you can use StarWind vSAN (even free edition of it...) to handle some shared storage for your file server. It's a fully supported scenario.

    3) It's still better to use async replication with SQL Server (read-only remote replica). In such a case FCI isn't what you want, but you'd better stick with Availability Groups (AGs). See:

    https://docs.microsoft.com/en-us/sql/database-engine/availability-groups/windows/getting-started-wit...

    4) There might be a way to do the same with Nimble's built-in stuff, but I'm not aware of the details. Adding SGF.

    Spice (2) flagReport
    1 found this helpful thumb_up thumb_down
  • HPEStorageGuy (HPE) wrote:

    And yes, Nimble Storage has Sync replication that we call Peer Persistence (came from the name we called it with 3PAR). Not sure how the integration with that is with Veeam but asking my tech marketing lead that does Veeam and HPE Storage integration.

    Yes, I think for now I can't pursue an implementation with Peer Persistence (synchronous replication + Automatic Switchover), and perhaps not even with plain synchronous replication by itself. I don't mind a lack of support for Peer Persistence; I'm not sure we would implement that in any case.

    I came across this regarding backups from Nimbles with synchronous replication:

    https://helpcenter.veeam.com/docs/backup/vsphere/system_requirements.html?ver=110

    https://forums.veeam.com/vmware-vsphere-f24/hpe-nimble-veeam-storage-discovery-jobs-causing-vcenter-...

    I probably need to devise a temporary DR setup until Veeam B&R v12 drops (assuming support for synchronous replication volumes is in the final release).

    Spice (1) flagReport
    1 found this helpful thumb_up thumb_down
  • SoggyFrog wrote:

    HPEStorageGuy (HPE) wrote:

    And yes, Nimble Storage has Sync replication that we call Peer Persistence (came from the name we called it with 3PAR). Not sure how the integration with that is with Veeam but asking my tech marketing lead that does Veeam and HPE Storage integration.

    Yes, I think for now I can't pursue an implementation with Peer Persistence (synchronous replication + Automatic Switchover), and perhaps not even with plain synchronous replication by itself. I don't mind a lack of support for Peer Persistence; I'm not sure we would implement that in any case.

    I came across this regarding backups from Nimbles with synchronous replication:

    https://helpcenter.veeam.com/docs/backup/vsphere/system_requirements.html?ver=110

    https://forums.veeam.com/vmware-vsphere-f24/hpe-nimble-veeam-storage-discovery-jobs-causing-vcenter-...

    I probably need to devise a temporary DR setup until Veeam B&R v12 drops (assuming support for synchronous replication volumes is in the final release).

    Yeah, my HPE Storage and Veeam integration engineer replied to me that it comes in v12. 

    If you worked with Zerto and Veeam in your previous job, I'm sure you have a very good handle on both of those. Sync replication can be taxing on Nimble. Currently, our "go to" solution for DR is with Zerto. They have been doing CDP with 0 RPO for a LONG time. But obviously it comes down to what you need and your budget. Happy to answer questions on the HPE side of the equation. 

    Spice (3) flagReport
    2 found this helpful thumb_up thumb_down
  • You might want to check out our recent post about Backup Exec 22. It includes points about:

    -New BE Simple Subscription Licensing
    -New Built-in protection of Microsoft 365 OneDrive and Exchange Online
    -Support for V4 authentication of S3 service
    0 of 1 found this helpful thumb_up thumb_down
  • My philosophy (not sure it's right) has always been to use application level clustering, such as Exchange DAG or SQL Clustering, etc. even if I am augmenting it with VMware level clustering locally. I think in general it takes more effort and complexity to cluster across regions/sites and why Veeam and Zerto work so well and are popular for doing DR fail-over (and fail-back). 

    Has the organization defined the RPO?  Do you really require 10 second RPO?  How many servers require that RPO and how many don't?  We have a handful of customers that use both Veeam and Zerto to replicate to our DRaaS. Works great, though fail-over, fail-back is a slightly different process for each.

    Was this post helpful? thumb_up thumb_down

Read these next...

  • Snap! Reporting phishing in Teams, State of IT, Arc A770 graphics card, Optimus

    Snap! Reporting phishing in Teams, State of IT, Arc A770 graphics card, Optimus

    Spiceworks Originals

    Your daily dose of tech news, in brief. How is it already Monday? Actually, how is it already October 2022? It felt like SpiceWorld was just starting, and already it's over (for this year). But don't worry, the fun continues as this month is Cybers...

  • Can you run a print server on windows 10

    Can you run a print server on windows 10

    Windows

    I have many clients with no servers.  Looking for the best way to manage printers.If I share the printer then that machine that shares the printer needs to be online to print correct?   Having to always download the print driver and lookup the printers IP...

  • Lost of VDI rights on Windows 10 Enterprise E3

    Lost of VDI rights on Windows 10 Enterprise E3

    Software

    I am migration from standalone Windows 10 Enterprise E3 online subscription to Microsoft 365 E3 online subscription. However, I am seeing this in the terms.https://www.microsoft.com/licensing/terms/productoffering/Microsoft365/MOSADoes this mean I am losi...

  • Spark! Pro Series - 3rd October 2022

    Spark! Pro Series - 3rd October 2022

    Spiceworks Originals

    Welcome to Monday. I am currently sat at home in isolation, with my wife and I having succumbed to the dreaded Covid-19 after managing to avoid it for the last two and a half years! Still, it gives me the chance to indulge my fancy in yet another Spark! t...

  • What security best practices would you suggest?

    What security best practices would you suggest?

    Spiceworks

    It's Cybersecurity Awareness month, which if people took seriously, would likely be a great boon to all of ITkind. :)  The National Cybersecurity Alliance suggests some specific things that people can do to increase cybersecurity: Enabling multi-facto...