Server MFG reliability poll.

Who makes the most reliable server/board?

  • Dell

  • HP

  • IBM

  • Cisco

  • Lenovo

  • Asus

  • ASRock

  • Gigabyte

  • Intel

  • Supermicro


Results are only viewable after voting.
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Patriot

Moderator
Apr 18, 2011
1,450
789
113
I have lost <5 HP boxes over the course of 10 years. I have seen a DL380gen8 be miss-racked and take a nose dive from chest height landing on a corner. Chassis stayed straight, plastic corner clip broke but the box was fine and was not part of the lost in action. Not sure how many boxes were in that lab... but about 40 racks worth. Had one procurve bit the dust after a hard power fault.

IDK with that many racks, replacing servers every other gen give or take... that is going to be well over 500 boxes. So sub percent failure rate.

Disclaimer 1, I work for them, Disclaimer 2 proto deaths don't count, though they tend to be quirky not die young.
 
  • Like
Reactions: Patrick

Blinky 42

Active Member
Aug 6, 2015
615
232
43
48
PA, USA
We had HPs and real IBMs (before Lenovo) slog along happy and healthy for 10+ years in all sorts of remote locations with only the normal disk-related issues in general. Other than human and natural causes (heavy stuff falling on server and killing it, lightning strikes) we had probably < 10 boxes out of 300+ deployed that had an issue, and the HP or IBM service peeps would be able them fix it right up. The servers lasted longer than the companies that originally purchased them ;)

After the IBM era we have had ~100 various supermicro based boxes that held up in the same crazy remote conditions for 10+ years across 3 companies and multiple resale and relocation of hardware.
 

cheezehead

Active Member
Sep 23, 2012
723
175
43
Midwest, US
So far only have dealt with quantities (100+ of each) of Dell, HP, and IBM. The only issue I've had was with Dell's small business line of servers...over multiple generations of boxes burning out the onboard nics after <200TB transmitted (imaging servers).


Really the question should be split into SMB servers (aka glorified desktop boxes) and enterprise gear. Enterprise gear regardless of the vendor tends to be pretty stable (kind of a key requirement). SMB servers are where companies look at the enterprise gear and what can we do to make it significantly cheaper, one thing that often comes up in short QA cycles and sometimes shoddy component suppliers.
 

Jon Massey

Active Member
Nov 11, 2015
339
82
28
37
Worked with many racks of HP and Dell servers and blades over the years, all on fixed 5 year lifespans, not had significant failure rates from either. Even the old Dell SC boxes we used to use as warehouse terminals in filthy conditions seemed to survive 5 years without issue.
 

Diavuno

Active Member
Really the question should be split into SMB servers (aka glorified desktop boxes) and enterprise gear. Enterprise gear regardless of the vendor tends to be pretty stable (kind of a key requirement). SMB servers are where companies look at the enterprise gear and what can we do to make it significantly cheaper, one thing that often comes up in short QA cycles and sometimes shoddy component suppliers.
That would be interesting to see, I know the low end dells 70% of the time are optiplex chassis/boards some are vostro or even the dreaded inspiron.
 

cheezehead

Active Member
Sep 23, 2012
723
175
43
Midwest, US
That would be interesting to see, I know the low end dells 70% of the time are optiplex chassis/boards some are vostro or even the dreaded inspiron.
The ones I was dealing with at the time had the feeling of the "cheap" optiplexes...they seem to go in good and bad generations on the low end models.