• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 642
  • Last Modified:

Server comparaison

Hello

I have 2 servers and I was wondering if someone could acknowledege my findings.

First Server

Windows 2008 R2 Enterprise
Xeon X5560
8 GB Ram DDR3 1333 Mhz
136Gb ST3146356SS

Second Server

Windows 2008 R2 Standard
I5 750
8 GB Ram DDR3 533 Mhz
1 TB ST31000524AS


We use an single thread application and its doing a lot of calculations and intrapolation. The first server is considerably more faster than the second. A set of calculation 30 mins on the first server vs 4 hours on the second server.

Is it true, to say just with the processor, a 1200$ processor vs a 150$ processor, there is something there that the second server cant handle. The low speed ram is another indicator. Finally the hard drives (15K RPM vs 7K RPM).
0
metraon
Asked:
metraon
1 Solution
 
Frosty555Commented:
Think about exactly what kind of processing the server is doing and what is involved at the hardware level to make that happen. You say it is a single-threaded application doing a lot of number crunching.

That tells me it will be

- Reading/writing a small or insignificant amount of data from the hard disks (so hard disks don't really matter)
- Lots of RAM access, swapping bits of data in and out of the CPU's cache (so quantity of RAM and ram speed and timing is very important)
- Executing lots of arithmetic / floating point instructions (so the efficient design and optimization that went into the ALUs, floating point processors and all the other complicated stuff that we take for granted on the CPU are critical)
- No multi-core optimizations, so it's just raw single-core speed (so overall clock speed of the CPU is critical)

So basically the CPU and RAM, and the communication between these two components is the most significant factor here. The CPU's ability to crunch numbers, read instructions, and move data inbetween it's internal caches, registers, and the RAM on the motherboard is what will affect the overall speed of the calculations you're doing.

The Xeon has different technology inside than the i5. It looks like it has a much faster QPI, slightly faster clock speed, and most likely the rest of the hardware on the server will be more highly tuned and designed with enterprise usage in mind, so it will overall run much faster than the second server.

But this kind of functionality is difficult to measure, benchmark and publish specs for, and usually the differences involve complicated optimizations and quality control that are not really very easy to market. So even though the published specs of the enterprise grade hardware in your first server are pretty similar to the consumer grade hardware in your second server..... you see a big difference in performance when you're doing certain specific CPU intensive tasks.
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now