Which system is faster?

UNIX for Beginners Q & A UNIX for Dummies Questions & Answers
1

i'm trying to decide if to move operations from one of these hosts to the other. but i cant decide which one of them is the most powerful.

each host has 8 cpus.

HOSTA

processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 44
model name	: Intel(R) Xeon(R) CPU           E5630  @ 2.53GHz
stepping	: 2
cpu MHz		: 1600.000
cache size	: 12288 KB
physical id	: 0
siblings	: 8
core id		: 0
cpu cores	: 4
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes

bogomips	: 5066.69
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management:

and

HOSTB

processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 15
model name	: Intel(R) Xeon(R) CPU           E5345  @ 2.33GHz
stepping	: 11
cpu MHz		: 1999.998
cache size	: 4096 KB
physical id	: 0
siblings	: 4
core id		: 0
cpu cores	: 4
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 10
wp		: yes
flexpriority
bogomips	: 4999.8
clflush size	: 64
cache_alignment	: 64
address sizes	: 38 bits physical, 48 bits virtual
power management:
2

Consider picking the CPU with the higher Bogomips.

3

thank you for this. can i ask you to please explain why? i was gonna go by the Hz but i wasn't sure if that's an accurate gauge.

4

In general, the more Hz the better the performance. But not always. MIPS is more meaningful because it measures actual instructions carried out (million instructions per second). Depending on the CPU, an instruction may take several cycles.

PassMark Software - CPU Benchmark Charts uses "passmarks" to measure CPU performance. I think is perhaps as valid as anything else I've seen.

E5630 @ 2.53GHz is 4991 passmarks, listed at $558
E5345 @ 2.33GHz is 2710 passmarks, listed at $196

So by both measures the first host will likely perform better. It's also a more expensive CPU, and that's an indication that it's likely better.

5

The only way to know which system will be faster at running your workload is to benchmark them using your workload.

For what it's worth (which isn't much), the first thing that jumps out at me from the OP is the cache differential. Cache misses are awfully expensive.

It's been over 15 years since I first encountered it, but the word "bogomips" still makes me chuckle a bit.

Regards,
Alister

6

Also I noticed that E5630 has 4 cpu cores and 8 siblings.

If the number of cores = the number of siblings for a given physical processor, then hyper-threading is OFF.

So if hyper-threading is ON, it may help achieve better function unit utilization by letting more than one program execute concurrently.

7

thanks everyone. looks like HOSTA is winning out here. and yoda thanks for the hyperthreading information!!!