Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › Decent RAID5 controller?
New Posts  All Forums:Forum Nav:

Decent RAID5 controller? - Page 7

post #61 of 75
Quote:
Originally Posted by the_beast View Post
Not if their HDDs will be bottlenecking them it isn't - a better disk array will help more than extra RAM or faster CPUs if what you're doing is I/O bound.

The point is that you have to gear the hardware precisely to the type of usage it will see. You can't just say one setup is always faster than another (or cheaper for a given performance level), because it just isn't true.
That was the same point I was trying to make. Situational situations are situational. lolwut.
Money Pit
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q6600 3.9ghz (433x9) eVGA 780i FTW PNY GTX 480 + eVGA 9800gtx+ (physx) Corsair XMS2 4x2gb 833mhz 
Hard DriveOSMonitorKeyboard
Seagate 2x750gb RAID0 | 5x1.5tb RAID5 Win 7 Ult. x64 22" Samsung 2233rz | 22" Acer 223wx Logitech G15 
PowerCaseMouseMouse Pad
Corsair HX750w + FSP GPU PSU 450w MX-Alien Razer Copperhead Thermaltake LCD Pad 
  hide details  
Reply
Money Pit
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q6600 3.9ghz (433x9) eVGA 780i FTW PNY GTX 480 + eVGA 9800gtx+ (physx) Corsair XMS2 4x2gb 833mhz 
Hard DriveOSMonitorKeyboard
Seagate 2x750gb RAID0 | 5x1.5tb RAID5 Win 7 Ult. x64 22" Samsung 2233rz | 22" Acer 223wx Logitech G15 
PowerCaseMouseMouse Pad
Corsair HX750w + FSP GPU PSU 450w MX-Alien Razer Copperhead Thermaltake LCD Pad 
  hide details  
Reply
post #62 of 75
Thread Starter 
Quote:
Originally Posted by the_beast View Post
People don't go with RAID5 because it's cheaper, they go with RAID5 because it's faster. In certain circumstances. Or at least those people who know what they're doing choose either RAID5 or RAID10 depending on the situation. Cost is a non-issue - you don't drop $20k on a server then worry about another few hundred on a drive if it will boost your performance by 50%, which is easily possible with higher spindle counts for read-heavy uses...

It's worth noting (regarding your pros/cons) that a RAID5 array can take hours or even days to rebuild (especially if the array remains in service during the failure) - so relying on a hotspare is not ideal, especially as a rebuild triggers heavy workload on the remaining drives, so a second failure is more likely to happen during a rebuild as the drives take a beating.

Remember when you run your tests you need to replicate the exact usage you want to use the array for - which is all but impossible. If you set up the test wrong (like doing sequential read tests for an array which is used for small writes) then your benchmarks could throw you off in completely the wrong direction - kinda like when people point to a 4-drive RAID0 array of mechanical drives being faster than a C300 for sequential reads, so therefore it must be a better OS drive...
I can see RAID5 on a good controller being faster with reads vs a RAID10, once you get past using 4 drives for the array. RAID5 can effectively read from all but one HDD at the same time, whereas RAID10 is limited to reading from half of them. It's just a matter of whether the parity calculations take longer than the difference between the RAID5 and RAID10 read speeds.

Yes - I've experienced having to rebuild a RAID5 a couple of times. Once was on a mailserver, and I had 120 people asking me why they still couldn't access their email, and when it would be back. I do see your point about the hotspare as well - it will take time to rebuild on a spare drive, and it will definitely increase the possibility of a second drive failure while it is rebuilding. Of course, this could be said of a RAID10 as well - it just won't take as long to rebuild a RAID10, and it would only be taxing one HDD instead of all of them.

I like CrystalMark for a HDD benchmark. I'll usually let it run through all of the ranges, then compare. The 512KB read/write would probably be the most applicable to my situation. It's certainly not going to be sequential, but the vast majority of files updated will be 4MB each. Any better benchmarks to use?
post #63 of 75
CrystalDiskMark has always been ok for me! You can also try HDTune.
     
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R MSI HD5970 @ 950MHz & 1250MHz Mem 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingCooling
SanDisk Xtreme 240GB Who Cares? Custom Loop Black Ice GT Stealth 360mm Rad 
CoolingCoolingCoolingCooling
Black Ice GT Stealth 240mm Rad Laing DDC-1+ Pump EK Waterblocks 10335 EK-Supreme Highflow CPU Bl... EK-FC5970  
OSMonitorKeyboardPower
Win 7 Ultimate X64 Samsung Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff 
CaseMouse
Xigmatek Elysium Steelseries Xai 
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R XFX HD 5970 @ 950MHz 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingOS
SanDisk Xtreme 240GB Who Cares? Corsair H80 Win 7 Ultimate X64 
MonitorKeyboardPowerCase
Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff CoolerMaster CM690 
Mouse
Logitech G500 
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R XFX HD 5970 @ 950MHz & 1250MHz Mem 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingOS
SanDisk Xtreme 240GB Who Cares? Custom Loop Win 7 Ultimate X64 
MonitorKeyboardPowerCase
Samsung Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff HAF 922 
Mouse
Steelseries Xai 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R MSI HD5970 @ 950MHz & 1250MHz Mem 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingCooling
SanDisk Xtreme 240GB Who Cares? Custom Loop Black Ice GT Stealth 360mm Rad 
CoolingCoolingCoolingCooling
Black Ice GT Stealth 240mm Rad Laing DDC-1+ Pump EK Waterblocks 10335 EK-Supreme Highflow CPU Bl... EK-FC5970  
OSMonitorKeyboardPower
Win 7 Ultimate X64 Samsung Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff 
CaseMouse
Xigmatek Elysium Steelseries Xai 
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R XFX HD 5970 @ 950MHz 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingOS
SanDisk Xtreme 240GB Who Cares? Corsair H80 Win 7 Ultimate X64 
MonitorKeyboardPowerCase
Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff CoolerMaster CM690 
Mouse
Logitech G500 
CPUMotherboardGraphicsRAM
I7 920 D0 @ 4.2Ghz Gigabyte X58A-UD3R XFX HD 5970 @ 950MHz & 1250MHz Mem 12GB Corsair Vengeance 8-8-8-24 
Hard DriveOptical DriveCoolingOS
SanDisk Xtreme 240GB Who Cares? Custom Loop Win 7 Ultimate X64 
MonitorKeyboardPowerCase
Samsung Syncmaster 2494 Steelseries 6Gv2 XFX Pro 850W Silver 88%Eff HAF 922 
Mouse
Steelseries Xai 
  hide details  
Reply
post #64 of 75
Thread Starter 
Quote:
Originally Posted by Mattyd893 View Post
CrystalDiskMark has always been ok for me! You can also try HDTune.
HDTune is pretty much worthless unless you want to see random access times. It only does sequential read/write tests, so it doesn't tell you much of anything about how the drive will perform with real-world usage. The random access time tests can be interesting, but I would expect those to be similar between both RAID variations. I could be wrong on that though.

Random writes of various sizes, like what CrystalMark does, is a much better test.
post #65 of 75
Nothing beats IOMeter for options - you can test pretty much anything you can dream up. But with that flexibility comes a fairly hefty degree of complexity...
post #66 of 75
Thread Starter 
Quote:
Originally Posted by the_beast View Post
Nothing beats IOMeter for options - you can test pretty much anything you can dream up. But with that flexibility comes a fairly hefty degree of complexity...
Thanks, that's a very cool utility.
post #67 of 75
Quote:
Originally Posted by SgtSpike View Post
If I have two devices plugged into my router, and they are communicating with each other, they will be using router resources (CPU/RAM). I already have periods of time where my RAM usage is 70%, and this is using an ASUS RT-N16 with 128MB RAM.
I find that strange. Any traffic passing across the switch shouldn't even involve the router CPU or use any significant amount of RAM. I have a D-Link 24 port GigE switch, and I very much doubt it comes with 128MB of RAM.
Ryzen
(12 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Gigabyte GA-AB350M Gaming 3 Palit GT-430 Corsair Vengeance LPX CMK16GX4M2B3000C15 
Hard DriveCoolingOSMonitor
Samsung 850 EVO AMD Wraith Spire Linux Mint 18.x Dell UltraSharp U2414H 
KeyboardPowerCaseMouse
Apple Basic Keyboard Thermaltake ToughPower 850W Lian-Li PC-A04B Logitech Trackman Wheel 
  hide details  
Reply
Ryzen
(12 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Gigabyte GA-AB350M Gaming 3 Palit GT-430 Corsair Vengeance LPX CMK16GX4M2B3000C15 
Hard DriveCoolingOSMonitor
Samsung 850 EVO AMD Wraith Spire Linux Mint 18.x Dell UltraSharp U2414H 
KeyboardPowerCaseMouse
Apple Basic Keyboard Thermaltake ToughPower 850W Lian-Li PC-A04B Logitech Trackman Wheel 
  hide details  
Reply
post #68 of 75
Thread Starter 
Quote:
Originally Posted by parityboy View Post
I find that strange. Any traffic passing across the switch shouldn't even involve the router CPU or use any significant amount of RAM. I have a D-Link 24 port GigE switch, and I very much doubt it comes with 128MB of RAM.
Well, I was just making an assumption there, that it would consume router resources. How much it would consume might be difficult to determine. I just don't want to have another potential factor that could lag the server<->outside communications when it can be easily avoided by not going the NAS route. If the NAS can read/write as fast as a gigabit connection can handle, I'd have to deal with packet prioritization, etc, and I really don't want to do that.

How could traffic not involve the CPU? Do switches not have a CPU?
post #69 of 75
Not in the same way that a router does. I believe switches do have a processor of some kind, and a certain amount of memory allocated per port, but I believe that on a router, the embedded switch will only pass traffic to the CPU whose destination is not on the switch.

I think they use ARP tables to make this determination. As an example, I believe that if you set up two BitTorrent clients (one as a seed) on your LAN, the traffic will not pass through the router, since it wouldn't actually be involved.
Ryzen
(12 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Gigabyte GA-AB350M Gaming 3 Palit GT-430 Corsair Vengeance LPX CMK16GX4M2B3000C15 
Hard DriveCoolingOSMonitor
Samsung 850 EVO AMD Wraith Spire Linux Mint 18.x Dell UltraSharp U2414H 
KeyboardPowerCaseMouse
Apple Basic Keyboard Thermaltake ToughPower 850W Lian-Li PC-A04B Logitech Trackman Wheel 
  hide details  
Reply
Ryzen
(12 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Gigabyte GA-AB350M Gaming 3 Palit GT-430 Corsair Vengeance LPX CMK16GX4M2B3000C15 
Hard DriveCoolingOSMonitor
Samsung 850 EVO AMD Wraith Spire Linux Mint 18.x Dell UltraSharp U2414H 
KeyboardPowerCaseMouse
Apple Basic Keyboard Thermaltake ToughPower 850W Lian-Li PC-A04B Logitech Trackman Wheel 
  hide details  
Reply
post #70 of 75
Thread Starter 
Quote:
Originally Posted by parityboy View Post
Not in the same way that a router does. I believe switches do have a processor of some kind, and a certain amount of memory allocated per port, but I believe that on a router, the embedded switch will only pass traffic to the CPU whose destination is not on the switch.

I think they use ARP tables to make this determination. As an example, I believe that if you set up two BitTorrent clients (one as a seed) on your LAN, the traffic will not pass through the router, since it wouldn't actually be involved.
Cool, well that's good to know. Thanks.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: RAID Controllers and Software
Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › Decent RAID5 controller?