Some time ago, I published here comparison of SQL Server Failover Cluster Instances (FCI) and SQL Server Availability Group (AG) performance while having them run on top of StarWind Virtual SAN (https://www.starwindsoftware.com/blog/hyper-v/can-sql-server-failover-cluster-instance-run-twice-fast-sql-server-basic-availability-groups-2-node-cluster-part-2-studying-fci-performance/). Today, I measure SQL Server AG performance on Storage Spaces. The next part sheds light on SQL Server FCI performance on S2D. Can I squeeze two times more TPM out of SQL Server FCI on S2D than SQL Server AG can provide on Storage Spaces?

It is the second part of my research on SQL Server Availability Groups (AG) and SQL Server Failover Cluster Instances (FCI) performance. Before, I measured SQL Server AG performance on Storage Spaces (https://www.starwindsoftware.com/blog/hyper-v/can-sql-server-failover-cluster-instance-run-s2d-twice-fast-sql-server-availability-groups-storage-spaces-part-1-studying-ag-performance/). Today, I study the performance of SQL Server FCI on S2D, trying to prove that this thing can run 2 times faster than SQL Server AG on Storage Spaces

  Some time ago, I wrote an article about backup storage media. Today, I’d like to talk about secondary storage. Before I move on, I want to clarify what I mean by “secondary storage” here, just to make sure that we are on the same page. Secondary storage is the storage where the actively used data resides. It can be both some local storage like SAN or NAS, or some public cloud hot tier. Well, it’s absolutely true that you can use disk arrays too, but let’s think of them today just as NAS-like servers packed with many disks, ok? That’s entirely up to you “which side you are on”, and there’s no “one-size-fits-all” solution. NAS, SAN, and public cloud storage… Whatever secondary storage you choose, it has own pros and cons. I discuss them in this article.

How does it work? Being triggered, the Node Fairness service determines the current workload on the cluster hosts. If needed, it automatically migrates VMs to the less loaded host. After the start, the cluster service checks the host for two factors: If the average host CPU load exceeds the tolerated threshold If the average host memory utilization is higher than the tolerated threshold In case either of these criteria is met, the VM gets live migrated to the less loaded host. Neither guest system nor services do experience any downtime.

So, we all know about Microsoft’s Storage Spaces Direct (S2D to put it simple) by now. It’s the feature introduced in Microsoft Server 2016 (Datacenter Edition) that pools together server’s storage allowing to build…that’s right: highly available and easily scalable software-defined storage systems. In this article, I’m gonna talk about not as much about its fault-tolerance characteristics themselves, but some hands-on experience, namely: how to replace a failed disk.

For that matter, I’ll create a file server of two types: “File server for general use” and Scale-Out File Server. For the latter, I’ll have two scenarios: VMs running and off. After that, I’m gonna evaluate the deduplication ratio using a free tool – StarWind Deduplication Analyzer. And finally, run data deduplication using the Windows Server 2016 Data Deduplication feature. We also need our storage shared, so for this purpose, I’ve decided on StarWind Virtual SAN Free.

While some OS-s built on Linux kernel support NVMe-oF, Windows just does not. No worries, there are some ways to bring this protocol to a Windows environment! In this article, I investigate whether presenting an NVMe drive over RDMA with Linux SPDK NVMe-oF Target + Chelsio NVMe-oF Initiator provides you the perfomance that vendors of flash list in their datasheets.

In my previous article, I measured SQL Server Basic Availability Groups (BAG) performance. This, as it comes from the name, addresses SQL Cluster Failover Cluster Instance (FCI) performance. I expect SQL Server FCI to exhibit two times higher performance than BAG. Before I start, I’d like to tell you one important thing about this measurement. SQL Server FCI database resides on a StarWind virtual device. Why did I choose StarWind? Because I got their NFR license some time ago and decided to give this software-defined storage solution a shot. Let’s just hope that it won’t limit SQL Server FCI performance.