Benchmarks Hardware Qualcomm Centriq vs. Intel Xeon Skylake/Broadwell CPU comparison

eva2000 · Nov 12, 2017

Interesting server processor developments are happening with Qualcomm Centriq 48 core ARM based processor announcements Qualcomm Launches 48-core Centriq for $1995: Arm Servers for Cloud Native Applications.

Cloudflare has early access to Qualcomm Centriq 2452 46 core model and ran some comparison benchmarks against Intel Xeon Skylake and Broadwell based server cpus ARM Takes Wing: Qualcomm vs. Intel CPU comparison

At Cloudflare we run a complex software stack that consists of many integrated services, and running each of them efficiently is top priority.

On the edge we have the NGINX server software, that does support ARMv8. NGINX is written in C, and it also uses several libraries written in C, such as zlib and BoringSSL, therefore solid C compiler support is very important.

In addition our flavor of NGINX is highly integrated with the lua-nginx-module, and we rely a lot on LuaJIT.

Finally a lot of our services, such as our DNS server, RRDNS, are written in Go.

The good news is that both gcc and clang not only support ARMv8 in general, but have optimization profiles for the Falkor core.

Go has official support for ARMv8 as well, and they improve the arm64 backend constantly.

As for LuaJIT, the stable version, 2.0.5 does not support ARMv8, but the beta version, 2.1.0 does. Let’s hope it gets out of beta soon.
Click to expand...

OpenSSL benchmarks

single core

all cores - very nice potential for Qualcomm Centriq !

Falkor is at a disadvantage here. First, in a single core benchmark, the turbo is engaged, meaning the Intel processors run at a higher frequency. Second, in Broadwell, Intel introduced two special instructions to accelerate big number multiplication: ADCX and ADOX. These perform two independent add-with-carry operations per cycle, whereas ARM can only do one. Similarly the ARMv8 instruction set does not have a single instruction to perform 64-bit multiplication, instead it uses a pair of MUL and UMULH instructions.

Nevertheless, at the SoC level, Falkor wins big time. It is only marginally slower than Skylake at an RSA2048 signature, and only because RSA2048 does not have an optimized implementation for ARM. The ECDSA performance is ridiculously fast. A single Centriq chip can satisfy the ECDSA needs of almost any company in the world.

It is also very interesting to see Skylake outperform Broadwell by a 30% margin, despite losing the single core benchmark, and only having 20% more cores. This can be explained by more efficient all-core turbo, and improved hyper-threading.
Click to expand...

full benchmarks ARM Takes Wing: Qualcomm vs. Intel CPU comparison

eva2000 · Nov 12, 2017

Also interesting article on Intel frequency scaling On the dangers of Intel's frequency scaling

To keep power in check Intel introduced something called dynamic frequency scaling. It reduces the base frequency of the processor whenever AVX2 or AVX-512 instructions are used. This is not new, and has existed since Haswell introduced AVX2 three years ago.

The scaling gets worse when more cores execute AVX-512 and when multiplication is used.

If you only run AVX-512 code, then everything is good. The frequency is lower, but your overall productivity is higher, because each instruction does more work.

OpenSSL 1.1.1dev implements several variants of ChaCha20-Poly1305, including AVX2 and AVX-512 variants. BoringSSL implements a different AVX2 version of ChaCha20-Poly1305. It is understandable then why BoringSSL achieves only 1.6GB/s on a single core, compared to the 2.89GB/s OpenSSL does.

So how does this affect you, if you mix a little AVX-512 with your real workload? We use the Xeon Silver 4116 CPUs, with a base frequency 2.1GHz, in a dual socket configuration. From a figure I found on wikichip it seems that running AVX-512 even just on one core on this CPU will reduce the base frequency to 1.8GHz. Running AVX-512 on all cores will reduce it to just 1.4GHz.

Now imagine you run a webserver with Apache or NGINX. In addition you have many other services, performing some real, important work. What happens if you start encrypting your traffic with ChaCha20-Poly1305 using AVX-512? That is the question I asked myself.
Click to expand...

Interesting comment in blog by Patrick from STH

See the Centriq testing for an offer to get code on more machines for you.

The Intel Xeon Silver line specifically is de-tuned for AVX-512. First, it does not have dual port FMA which limits throughput per clock. Second, clocks for AVX-512 are lower.

Your Silver 4116:
Non-AVX Base = 2.1GHz
AVX2 Base = 1.7Ghz
AVX-512 Base = 1.1GHz

Non-AVX Turbo with 12 cores = 2.4GHz
AVX2 Turbo with 12 cores = 2.1GHz
AVX-512 Turbo with 12 cores = 1.4GHz

So when you run AVX-512 on an Intel Xeon Silver you are getting the double-whammy of a low clock speed plus only "half" the benefit of a dual port Gold 6100 / Platinum SKU. Here is a good example where 2x 8 core Xeon Gold 6134's are over 2x the speed of 2x 10 core Xeon Silver 4114's in GROMACS which uses AVX-512
Click to expand...

Log in / Register

Forums

Join the community today

Benchmarks Hardware Qualcomm Centriq vs. Intel Xeon Skylake/Broadwell CPU comparison

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.

Log in / Register

Useful Searches

Join the community today

Benchmarks Hardware Qualcomm Centriq vs. Intel Xeon Skylake/Broadwell CPU comparison

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.