Is mutliprocessing worthless???

wacki

Senior member
Oct 30, 2001
881
0
76
According to anandtechs own: "AMD's Athlon XP: Great performance, poor marketing" article the AMD Athlon XP 1.53GHz (1800+) scored 213 on Internet Content Creation SYSMark 2001 bechmarks. Yet 2 Athlon MP's 1.2 Ghz could only score 208. (found in Tyan's Tiger MP: Affordable Multiprocessing - Sep 26th, 2001). According to these benchmarks, multiprocessing is twice as expensive and 1% slower than going the conventional route. I know I'm missing something but I cannot find one benchmark where 2 Athlon MP's beat 1 Athlon XP. I'm going to be building some servers for my website yet I cannot make an educated decision on Multiprocessing through benchmarks. Is Multiprocessing worthless? Are there any accurate benchmarks out there because I know this cannot be right. Also what benchmarks are best to look at for a simple website that gets alot of hits.

Please help,

The seriously confused (me)
 

interchange

Diamond Member
Oct 10, 1999
8,022
2,872
136
That was comparing a 1.53Ghz to two 1.2Ghz, which really doesn't say anything too entirely useful. SMP is good in certain cases, particularly in server environments. Server type software can be a big CPU drain when there are a lot of users, but each task for handling a user is likely completely independent of a task another user is running. As long as the data doesn't need to be shared, and you have an architecture that doesn't choke memory bandwidth like Intel's GTL+ bus, you will get double performance if both processors are maxed out.

Realistically, this never happens for a single-user gaming system. Games that want to take advantage of SMP have to be threaded properly and set up to do so (i.e. Quake III engine) for any practical benifit. Not only that, but common game environments are much more dependent on video cards and memory bandwidth. Some benchmarks may prove better than others for SMP.

One thing you will notice, however, is that SMP systems are often a lot more responsive, due to the fact that processing typically comes in short bursts, but this is still limited by software and OS design.

It depends on what kind of server you are going to be running whether or not SMP is right for you. In many cases, you may be simply limited by the amount of RAM you have and not even touch your second CPU.
 

Nothinman

Elite Member
Sep 14, 2001
30,672
0
0
SMP has clear advantages in some areas and none in others, you have to decide if what you'll be doing can take advantage of SMP.

You mention a web server which is something that can take advantage of SMP well, atleast Apache can which is where my knowledge lies, because it forks a seperate process for each connection so all the connections will be spread across the CPUs. If you start doing scripting with things like CGI or mod_perl it can be very helpfull because more scripts can run at once as long as they don't fight over resources (shared memory, files, etc). But the main consideration is how heavy a load you expect, a lot of times disk or network I/O becomes the bottleneck instead of CPU time so multiple machines in a load balanced cluster may be more desirable than one big SMP box.
 

wacki

Senior member
Oct 30, 2001
881
0
76
I am running a dedicated server to host a information database website. For instance the end user will only be looking at pictures and text. nothing else. The only services the server will provide will be e-mail, and a search-engine that will search the website itself. I expect several hundred hits per hour and high bandwidth will be needed to a transfer a very large amounts of tables, charts, and graphs (all in gif and only gif format). Will SMP benefit this and if so by how much. Also would there be a advantage of 2 gig ECC DDR RAM over 1 gig ECC DDR RAM in this scenerio??? Are there any benchmarks out there which I can make an educated decision on these Factors???


THX wacki
 

Nothinman

Elite Member
Sep 14, 2001
30,672
0
0
You still never stated the OS or Webserver you plan on using, that can make a difference.

In general it sounds like the most CPU intensive tasks will be the search functions, as long as each search is a seperate process or thread and they don't run into contention (and I doubt they will unless you use a lot of shared global variables or explicitely lock things yourself) you can run twice as many at a time with 2 CPUs. Of course other things like the speed of what you're searching are big too, if your disks are too slow you'll end up wating on disk I/O no matter how many searches can run at once, but having a lot of memory for disk cache should help alleviate that.

It's hard to say anything authoritative without knowing how CPU bound the search function is, how many files your searching, etc. At the very least I would get a SMP capable board, even if you only put in 1 CPU at first, with 1G memory and run it through some tests and see if you can tell where it's choking at, if it chokes at all.

Apache comes with a tool that specifically says:

ab is a tool for benchmarking the performance of your
Apache HyperText Transfer Protocol (HTTP) server. It does
this by giving you an indication of how many requests per
second your Apache installation can serve.


I'm not sure if you're describing the site as a database or if you're planning on running a real database on the box too. If you are running a database you probably want to run that on a seperate box if the web server is plenty busy.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |