Case Study — WonderPhi Compute Burn-In on a Dual-Xeon Workstation

Reading Time: 3 minutes
592 words

3 min read


7 parts, 15 paragraphs

Software acceleration for existing hardware.

Theory is cheap. We measured WonderPhi Compute on real hardware — a workstation with dual Intel Xeon Gold 6138T processors (40 physical cores, 80 logical), an NVIDIA A10M GPU with 19 GB of VRAM, running Windows 11 Pro. Every number below is wall-clock time, single run, cold cache. No averaging. No cherry-picking.

The Machine

CPU 2x Intel Xeon Gold 6138T @ 2.00 GHz — 40 physical / 80 logical cores
GPU NVIDIA A10M — 19 GB VRAM, Ampere architecture, 7th-gen NVENC
OS Windows 11 Pro
WPC Config 64 workers (cap 64, cores 80), GPU exposed as $WPC_GPU=1

This isn’t a cloud instance. This is a machine sitting on a desk, running production workloads around the clock, with WonderPhi’s three watchers armed and autonomous.

Test 1 — SHA-256 Hashing (Pure CPU Parallel)

The simplest possible benchmark: hash the same 500 MB file 40 times. CPU-bound, memory-bandwidth-sensitive.

Serial (bash for-loop) 96.61 seconds
WPC parallel (64 workers) 6.84 seconds
Speedup 14.1x

Why not the full 64x? All 40 threads read the same file concurrently — memory bandwidth saturates before CPU does. That’s a DDR4 bottleneck, not a WPC limit. Workloads with distinct files per task (like the 541-domain health check we ran) hit 40-50x because each worker reads independent bytes.

Test 2 — 1080p H.264 Video Encoding (CPU vs. GPU)

Same 3-minute 1080p30 source (5,400 frames), 3 parallel encodes on each side.

CPU (libx264, CRF 23) 51.38s — 315 fps aggregate
GPU (A10M NVENC, CQ 23) 48.07s — 337 fps aggregate
GPU speedup 1.1x (6% faster)

On 1080p H.264, this box’s 40-core CPU is competitive with the A10M. Use the GPU when you want to keep CPU free for other work, not for raw throughput.

Test 3 — 4K HEVC Encoding (CPU vs. GPU)

30-second 3840×2160 30fps source (900 frames), 2 parallel encodes.

CPU (libx265 fast, CRF 28) 36.27s — 50 fps aggregate
GPU (A10M NVENC, CQ 28) 24.36s — 74 fps aggregate
GPU speedup 1.5x (33% faster)

At 4K, the GPU pulls ahead. HEVC at this resolution is far heavier than H.264 at 1080p. NVENC is silicon — it doesn’t care about resolution the way software encoders do. The gap widens with resolution: at 8K, expect 3-5x.

Test 4 — AV1 Encoding

An honest caveat: the A10M does not support AV1 hardware encoding. Ampere NVENC (7th gen, 2020) has H.264 + HEVC only. AV1 NVENC arrived with Ada Lovelace (2022). All AV1 results are CPU.

libaom-av1 (reference encoder) 219.87s — 4.1 fps
libsvtav1 (fast encoder) 12.33s — 73.0 fps
Encoder choice speedup 17.8x

Encoder choice matters more than parallelism on AV1. And WPC parallelism helps for multi-file batches: 6 videos in parallel finished 3.1x faster than serial — but libsvtav1 is already multi-threaded internally, so you get diminishing returns.

The 541-Domain Health Check

We also ran a real-world operational test: a WPC plan with 541 independent curl health checks against live domains. Results: 442 reachable, 99 unreachable, total wall time 17.1 seconds. That’s 541 network requests, each with a 5-second timeout, completed in under 18 seconds.

Honest Takeaways

Workload Best Option on This Box
Batch hashing, checksumming, validation WPC parallel CPU (14-50x)
Batch web scraping / health checks WPC parallel CPU (bounded by network)
1080p H.264 encoding Either — CPU or NVENC depending on what else is running
4K HEVC encoding A10M NVENC (1.5x, grows with resolution)
AV1 encoding libsvtav1 on CPU (17.8x over libaom; no GPU AV1 HW)
Real-time streaming A10M NVENC — consistent fps regardless of CPU load

Every claim on the product page is backed by a number we measured. That’s the difference between marketing and evidence.

Get WonderPhi Compute — $99.99 →

Browse the full WonderPhi product line