10GbE speed expectations

10GbE File Transfer Speed: What Should You Actually Get?

Ten gigabits per second converts to 1,250 MB/s before protocol overhead, but File Explorer, Finder and NAS dashboards report application throughput after Ethernet, TCP, SMB/NFS and storage constraints. Those are different measurements.

Run the diagnostic See matched hardware

Quick answer

Start by isolating the layer, not by buying hardware

Current Amazon listings

Multi-gig hardware matched by exact model

Checking the dedicated multi-gig catalogue…

Technical decision

The troubleshooting decision

Interactive troubleshooting tool

10GbE File Transfer Speed Estimator

Enter measured values where possible. The result is a diagnostic priority, not a substitute for vendor support, cable certification, security policy or controlled testing.

Multi-gig checklist

Four checks before changing or replacing anything

Find the slowest link

End-to-end throughput cannot exceed the slowest NIC, switch port, uplink, cable path, host bus or storage endpoint in the flow.

Separate line rate from file speed

10GbE is 10 gigabits per second at the Ethernet layer, not a promise that every SMB copy will show 1.25 GB/s.

Test the network without storage

Use iPerf3 or an equivalent network-only benchmark before concluding that SMB, RAID or a disk is the reason the Ethernet path is slow.

Measure storage as well as network

A single HDD, small RAID, SATA SSD, NVMe pool and RAM cache can produce very different ceilings even on the same 10GbE link.

Start with the decimal line-rate conversion

10Gbps divided by eight equals 1.25GB/s, or 1,250MB/s, before Ethernet and protocol overhead. It is a useful upper-bound conversion, not a normal file-copy promise.

For 10GbE File Transfer Speed: What Should You Actually Get?, checkpoint 1 records start with the decimal line-rate conversion. Keep 10GbE File Transfer Speed: What Should You Actually Get? conditions fixed. Change one variable inside 10GbE File Transfer Speed: What Should You Actually Get?, then remeasure start with the decimal line-rate conversion. If start with the decimal line-rate conversion moves, 10GbE File Transfer Speed: What Should You Actually Get? stays on this layer. If start with the decimal line-rate conversion remains stable, 10GbE File Transfer Speed: What Should You Actually Get? advances. That 10GbE File Transfer Speed: What Should You Actually Get? discipline protects 10GbE File Transfer Speed: What Should You Actually Get? from unrelated hardware changes.

Use network-only throughput as the next ceiling

QNAP currently cites about 9.4Gbps, around 1.17GB/s, for a healthy 10GbE iPerf3 path. That shows the difference between raw line rate and usable TCP throughput.

Compare the source storage read rate

A client that reads at 500MB/s cannot send files at 1GB/s no matter how fast the NIC is. Benchmark the actual source medium and workload.

For 10GbE File Transfer Speed: What Should You Actually Get?, checkpoint 3 records compare the source storage read rate. Keep 10GbE File Transfer Speed: What Should You Actually Get? conditions fixed. Change one variable inside 10GbE File Transfer Speed: What Should You Actually Get?, then remeasure compare the source storage read rate. If compare the source storage read rate moves, 10GbE File Transfer Speed: What Should You Actually Get? stays on this layer. If compare the source storage read rate remains stable, 10GbE File Transfer Speed: What Should You Actually Get? advances. That 10GbE File Transfer Speed: What Should You Actually Get? discipline protects 10GbE File Transfer Speed: What Should You Actually Get? from unrelated hardware changes.

Compare the destination write rate

NAS writes can be limited by RAID parity, cache policy, disk zones, filesystem sync behavior or background work. Sustained write speed is often the decisive number.

Account for protocol efficiency

SMB and NFS add semantics and CPU work beyond TCP. Signing, encryption and metadata operations can reduce application throughput further.

Treat HDD, SATA SSD and NVMe differently

A single HDD is far below the full 10GbE payload. SATA SSD arrays can approach higher rates, while NVMe can exceed the network and make Ethernet the clearer bottleneck.

For 10GbE File Transfer Speed: What Should You Actually Get?, checkpoint 6 records treat hdd, sata ssd and nvme differently. Keep 10GbE File Transfer Speed: What Should You Actually Get? conditions fixed. Change one variable inside 10GbE File Transfer Speed: What Should You Actually Get?, then remeasure treat hdd, sata ssd and nvme differently. If treat hdd, sata ssd and nvme differently moves, 10GbE File Transfer Speed: What Should You Actually Get? stays on this layer. If treat hdd, sata ssd and nvme differently remains stable, 10GbE File Transfer Speed: What Should You Actually Get? advances. That 10GbE File Transfer Speed: What Should You Actually Get? discipline protects 10GbE File Transfer Speed: What Should You Actually Get? from unrelated hardware changes.

Separate burst speed from sustained speed

Caches can produce impressive short transfers that fall once dirty data must be committed to slower media. Test files larger than the relevant cache when sustained performance matters.

Understand RAID read and write asymmetry

Parity and mirror layouts can have different read and write characteristics. Do not assume a benchmark in one direction predicts the other.

Expect small files to underuse the link

File count, directory updates and metadata round trips make small-file workloads latency-sensitive. Network utilization can be low even when there is no defect.

For 10GbE File Transfer Speed: What Should You Actually Get?, checkpoint 9 records expect small files to underuse the link. Keep 10GbE File Transfer Speed: What Should You Actually Get? conditions fixed. Change one variable inside 10GbE File Transfer Speed: What Should You Actually Get?, then remeasure expect small files to underuse the link. If expect small files to underuse the link moves, 10GbE File Transfer Speed: What Should You Actually Get? stays on this layer. If expect small files to underuse the link remains stable, 10GbE File Transfer Speed: What Should You Actually Get? advances. That 10GbE File Transfer Speed: What Should You Actually Get? discipline protects 10GbE File Transfer Speed: What Should You Actually Get? from unrelated hardware changes.

Check CPU and security overhead

High-speed file protocols can become CPU-bound, especially with encryption or weak single-core performance. Monitor both endpoints during the test.

Use a range rather than one magic number

Real deployments vary by operating system, NIC, protocol, storage and workload. A modeled range is more useful than claiming every 10GbE system should display the same MB/s.

Upgrade only after comparing all ceilings

If storage peaks at 600MB/s, a faster switch will not create 1GB/s file copies. Upgrade the layer that sits below the performance target.

For 10GbE File Transfer Speed: What Should You Actually Get?, checkpoint 12 records upgrade only after comparing all ceilings. Keep 10GbE File Transfer Speed: What Should You Actually Get? conditions fixed. Change one variable inside 10GbE File Transfer Speed: What Should You Actually Get?, then remeasure upgrade only after comparing all ceilings. If upgrade only after comparing all ceilings moves, 10GbE File Transfer Speed: What Should You Actually Get? stays on this layer. If upgrade only after comparing all ceilings remains stable, 10GbE File Transfer Speed: What Should You Actually Get? advances. That 10GbE File Transfer Speed: What Should You Actually Get? discipline protects 10GbE File Transfer Speed: What Should You Actually Get? from unrelated hardware changes.

Questions people ask

Common questions about 10GbE file speed

Is 10GbE equal to 1,250 MB/s?

That is the raw decimal conversion before protocol overhead and endpoint limits.

What is a healthy 10GbE iPerf3 result?

QNAP currently uses roughly 9.4Gbps, around 1.17GB/s, as a healthy network-only reference.

Why do I get 700 MB/s instead of 1,000+?

Storage, SMB/NFS overhead, CPU, RAID and workload behavior can all reduce application throughput.

Can one SATA SSD saturate 10GbE?

A single SATA SSD usually cannot sustain the entire practical 10GbE payload by itself.

Can NVMe saturate 10GbE?

Many NVMe drives can exceed 10GbE network throughput, making the network a clearer limit.

Why are writes slower than reads?

RAID parity, cache policy and destination media behavior can make writes more expensive.

Why are small files so much slower?

Metadata and latency dominate, so the workload cannot use the full streaming bandwidth.

Does SMB signing count as overhead?

It adds integrity work and can affect CPU use, although the effect varies by platform.

Should I use a huge file for testing?

Use a file large enough to expose sustained behavior rather than only cache bursts.

How do I predict my speed?

Take the minimum of network payload, source read and destination write ceilings, then account for protocol and workload efficiency.

Primary references and methodology

Verify the exact port, cable, host and storage path

Use the slowest real layer as the expected ceiling: network payload, source read speed or destination write speed. Then discount further for protocol and workload behavior rather than expecting the raw 1,250 MB/s conversion. In 10GbE File Transfer Speed: What Should You Actually Get?, start with the decimal line-rate conversion is measured before account for protocol efficiency. 10GbE File Transfer Speed: What Should You Actually Get? then checks expect small files to underuse the link before upgrade only after comparing all ceilings. The 10GbE File Transfer Speed Estimator keeps 10GbE File Transfer Speed: What Should You Actually Get? measurements ordered. In 10GbE File Transfer Speed: What Should You Actually Get?, a passing layer is left untouched. A replacement purchase enters 10GbE File Transfer Speed: What Should You Actually Get? only after the measured diagnosis points there.

As an Amazon Associate, Cloudzat may earn from qualifying purchases. Ethernet line rate is not the same as application throughput. Product revisions, chipsets, firmware, operating-system drivers, cable quality, thermals, switch configuration and storage performance can change results; verify the exact hardware revision before purchase.