SSD vs HD = HD

hi guys ignore this title i dont suggest HD is better

heres situation i currently have
i7 2600k 16gb random access memory 2x1tb HD 1gbit vent from worldstream

and i choose to get the exact same cpu specs i7 2600k
but this place is using 2x240gb SSD by using hardware raid 0

then my spouse and i get my different server today and attempt to install everything i have around my old hosts like (modules, or anything else, etc)
ive fully installed 100% replicated even the OS are the same
the only new will be SSD with components raid, hardware raid should be faster than that software raid, but my old the first is not using just about any raid config,

here’s the thing i dont fully grasp my old Box
are nonetheless faster than our new SSD server
this new i7 need to be faster right becaused this is certainly SSD drives which should be good for the web server since im utilizing a database site (wordpress), and so i should look atleast 1. 5x improvement for this… im estimating the speed using the site load of my webpage, yeah women and men old server even now do better by using page speed (Miliseconds) concerning a tool similar to yslow to discover the page launching time.. so my partner and i cant be completely wrong… this new SSD server isn’t good

please explain to me anyone possess a experiece with raid 0 with regard to SSD
please say to me if i’m right or incorrect.. becaused i realize this new SSD server need to be faster,

for facts
im implementing wordpress
about 150mb database
using that modules
memcached
APC (Alternative php caching)
Varnish Cache
w3 whole cache plugin within wordpress( faster cache plugin)

old equipment load time is definitely: 0. 40 milliseconds (stable)
new node load time us: 0. 60-70 and sometimes 100-250ms

its unique!

please virtually any guide or comment will be appreciate

how do anyone measure the web site load time

i employ a addon in our browser like yslow

and it is cant be completely wrong i tried 20-30x but result are the same the previous server are still faster

About the only time your computer software stack touches prolonged storage significantly will be when MySQL needs to create a temporary table on computer. It doesn’t occur often (heavy select/join queries). You probably assumed that SSD always ends up in performance improvement no matter what the bottleneck.

SSD is not going to help if your own IOPS is underneath capacity.

Regards
Joe / UNIXY

so my personal old server employing HD 2x1tb remains better
can a person explaine more regards!

Quote: About really the only time your software package stack touches on going storage significantly can be when MySQL must create a technical staffing , table on disk. It doesn’t materialize often (heavy select/join queries). It is likely you assumed that SSD always contributes to performance improvement no matter what the bottleneck.

SSD will not only help if a person’s IOPS is down below capacity.

Regards
Joe / UNIXYSeek time continues to be reduced significantly and that could be apparent even when IOPS isn’t a difficulty.

You should often be seeing some degree of a performance development, even if not a terrific difference. Are you sure absolutely nothing is misconfigured

It is fairly easy that for some applications there isn’t a benefit to using SSDs, but it is rare a typical site will not only see an improvement for some reason.

Quote: Seek time continues to be reduced significantly and that could be apparent even in the event IOPS isn’t a difficulty. He’s heavy on memory and opcode caching. He / she barely touches the actual disk. Not also to seek/load a php file or maybe static assets.

Regards

Quote: They’ve heavy on recollection and opcode caching. He / she barely touches the actual disk. Not also to seek/load a php file or maybe static assets.

RegardsIn this kind of situation (especially with a great small database that is easily cached) We can see your issue.

However it’s still strange how the new system is usually -slower- which actually strongly suggests of which something is misconfigured…

Quote: However it’s still strange how the new system is usually -slower- which seriously strongly suggests this something is misconfigured… Arranged, it should be on par while using old system at worse.

Regards

Are you in a different host now
What brand ssd is he Some hosts utilize some pretty weak quality ssd of which dont even perform together with sata disks.

Quote: we’ve a addon with my browser such as yslow

and their cant be completely wrong i tried 20-30x but result include the same the outdated server are even now fasterSo, the measurement is about the client side, not the node side What actually does it assess The page era time or the particular page loading time May seem like it’s the second item (page loading time)

Are both the servers located inside same data center

Quote: They’ve heavy on recollection and opcode caching. He / she barely touches the actual disk. Not also to seek/load a php file or maybe static assets.

RegardsIf the 2 servers are identical, save the SATA compared to SSD drives, along with the two servers usually are configured identically, you ought to expect that the fresh server is speedier.

Doesn’t matter whether the site is heavy in memory or possibly not. Obviously…

Your repository size is 150MB, that is certainly extremely tiny. Likely your entire data set was in memory, in which case SSD wouldn’t make any difference as long as you’re not write heavy

This is not really related to your question but if you used a raid0 for any install then You have to keep very excellent backups. You will certainly need them eventually when one belonging to the drives is dropped because of the controller.

You can skip to the end and leave a response. Pinging is currently not allowed.
Custom Search

Leave a Reply