ZFS Managability

In this blog post, we will focus on ZFS from a manageability perspective. We will cover following topics

• ZFS Storage Status Monitoring
• Physical Disk Monitoring
• ZFS Capacity Monitoring
• ZFS Performance Monitoring
• Periodic and E-mail notifications

ZFS Storage Performance in my homelab

In my homelab I have Dell PowerEdge R620 Server with FreeBSD 14.3 and ZFS 2.2.7. I want to use this server for BHyVe server virtualization and run Virtual Machines on top of BHyVe hypervisor. 

In virtualized environment, the typical average I/O size differs based on workload running in virtual machines. Different applications generate distinct I/O patterns. 

• Databases and transactional systems: These often produce a large number of small, random I/O requests (e.g., 4KB, 8KB, or 16KB). This is because they frequently read and write small chunks of data to update records, log transactions, and access indexes.
• Virtual Desktop Infrastructure (VDI): VDI workloads are notoriously random and write-heavy, with an average I/O size often falling in the 24KB to 32KB range.
• File servers and data backups: These workloads typically generate large, sequential I/O requests (e.g., 64KB, 128KB, 256KB, or larger) as they read or write large files in a continuous stream. 

When I look at a typical enterprise cloud datacenter, where the types of workloads are not under your control, I usually observe the average I/O size between 40 KB/s and 64 KB/s. That's the reason why I typically test 32 KB I/O size, however, if you know the specific type of workload you are interested, you should test application specific I/O size.

LACP on FreeBSD

LACP stands for Link Aggregation Control Protocol. It’s a network protocol used to combine multiple physical network links into a single logical link to increase bandwidth and provide redundancy. It’s part of the IEEE 802.3ad standard (now 802.1AX). 

Here’s a breakdown of what it does and why it’s useful:

• Increases Bandwidth
• By bundling multiple links (like two or more Ethernet cables) between switches or between a switch and a server, the total throughput can be higher than a single link.
• Provides Redundancy
•  If one physical link fails, traffic is automatically rerouted over the remaining links, so the connection stays up.
• Dynamic Configuration
• LACP allows devices to automatically detect and configure link aggregation groups, making it easier to manage than static link aggregation.
• Load Balancing
• Traffic can be distributed across the aggregated links based on rules like source/destination IP, MAC addresses, or TCP/UDP ports. 

Let's configure and test it in my homelab.

LLDP on FreeBSD

LLDP stands for Link Layer Discovery Protocol. It’s a vendor-neutral Layer 2 protocol (defined in IEEE 802.1AB) that allows network devices (switches, routers, servers, firewalls, access points, phones, etc.) to advertise information about themselves to directly connected devices and to learn information about their neighbors.

In this short blog post we will install, enable and test LLDP on FreeBSD.

Home LAN (Cat5e vs WiFi) network throughputs

While testing Wi-Fi quality and network throughput on FreeBSD 14.3 drivers, I realized that before running any benchmarks, it’s important to document my home LAN topology and the network capacity across its zones. It’s essential to understand how different network technologies work, including the gap between their theoretical throughput and the actual achievable performance. 

For example, a Wi-Fi 5 (802.11ac) connection might advertise speeds up to 1.3 Gbps, but real-world performance is typically much lower due to factors like signal interference, channel width, and protocol overhead. Similarly, a 1 Gbps Ethernet link theoretically provides 1,000 Mbps, but after accounting for TCP/IP overhead and other factors, the actual throughput is closer to 940 Mbps. Another significant factor impacting real-world throughput is the use of Wi-Fi Mesh with wireless backhaul. While mesh systems improve coverage, they often introduce additional latency and bandwidth reduction because each hop between nodes consumes part of the available wireless spectrum for backhaul traffic. This means that, in practice, a device connected to a secondary mesh node (Extender) might experience only half or even less of the primary link’s bandwidth. Knowing these differences helps set realistic expectations and troubleshoot performance issues effectively. 

A picture is worth a thousand words, so here is a diagram illustrating both the theoretical and real-world throughput values in my home network setup.

Home LAN zones and Network Throughput

 You can find all the details in the remainder of this blog post.