How to increase LAN throughput?? utilities? tweaks?

[zapp22]

gang I have a small lan that works well for small tasks but the throughput seems awfully slow for how simple and straightforward it is. I need help understanding troubleshooting, any utilities that can help identify issues [I'm not a linux geek so a bit wary of NDT's rigor], and common things to look for.
This net has two simple routers, with one serving as a WIFI Client Bridge to the other. The one server is on CAT5 connected to the primary router. All the clients save one [a combo HTPC and utility mule] are via WIFI, with the AP being common everywhere, same credentials etc.

any/all links, utilities, methodology, common speedup tweaks would be greatly appreciated.

z

 

[bawldiggle]

I am a beginner, just started playing around with NAS/WiFi and inadequate LAN cabling.
Noticed you are using CAT 5 cable, have you considered CAT6, I think it is x10 faster than CAT5
(Somebody might correct me if I am wrong)

 

[Golden]

Comparison between CAT5 and 6:
Ask LH: What's The Difference Between Cat5, Cat5e And Cat6 Ethernet Cables? | Lifehacker Australia

 

[zapp22]

well, my answer will illustrate just how ignorant I really am!

I just ran some -to-internet-and-back tests... from my server [hardwired] and from my main win7 daily driver, via WIFI, thinkpad with its own embedded [which doesn't have a speed-demon reputation btw]. using ookla speedtest, the server got a hair over 22.2mbps download speed. I'm paying for 20. the notebook fared a little better [sunspots..] at 23. My 'droid [G Note 1] got about 18.

Where I notice the slowdown is when I transfer large files from the server to one of the clients or vice-versa [though I need to compare this more closely]. last time I watched one, what windows was reporting as the transfer speed was ~40mbps/5+MB/s. But is the windows 'meter' accurate? I would hope I could get closer to the 100mbps mark.

And, does this client/server 'speed' translate directly when doing media streaming? My assumption is "NO"... all bets off. I assume that one reason why specific "Media Server" 'wares emerged was to bias the system soft settings to favor large blocks at a fairly linear throughput, reliably. Am I close?

So back to the first issue: what really is a good way to MEASURE this stuff to start with? My main Router, and its subordinated Wireless-bridged-client, are ostensibly 150's, not 300's. And, FWIW, After a bunch of reading on supposed "dual band" setup in auto-select mode, I've nixed that and stick with 2.4ghz on the wifi side.

As to cable, I'll spring for CAT6 in a heartbeat if it is meaningful. the connection to the server is direct: its plugged directly into the main CSCO/Linksys router port, via a cat5 cable. The HTPC is plugged directly to port on the sub-tending router a couple of rooms away. all the other devices, most of the time, are WIFI only [three ipads, several 'droids, winxpp, win8.1, win7, no macs!]

 

[chev65]

well, my answer will illustrate just how ignorant I really am!

I just ran some -to-internet-and-back tests... from my server [hardwired] and from my main win7 daily driver, via WIFI, thinkpad with its own embedded [which doesn't have a speed-demon reputation btw]. using ookla speedtest, the server got a hair over 22.2mbps download speed. I'm paying for 20. the notebook fared a little better [sunspots..] at 23. My 'droid [G Note 1] got about 18.

Where I notice the slowdown is when I transfer large files from the server to one of the clients or vice-versa [though I need to compare this more closely]. last time I watched one, what windows was reporting as the transfer speed was ~40mbps/5+MB/s. But is the windows 'meter' accurate? I would hope I could get closer to the 100mbps mark.

And, does this client/server 'speed' translate directly when doing media streaming? My assumption is "NO"... all bets off. I assume that one reason why specific "Media Server" 'wares emerged was to bias the system soft settings to favor large blocks at a fairly linear throughput, reliably. Am I close?

So back to the first issue: what really is a good way to MEASURE this stuff to start with? My main Router, and its subordinated Wireless-bridged-client, are ostensibly 150's, not 300's. And, FWIW, After a bunch of reading on supposed "dual band" setup in auto-select mode, I've nixed that and stick with 2.4ghz on the wifi side.

As to cable, I'll spring for CAT6 in a heartbeat if it is meaningful. the connection to the server is direct: its plugged directly into the main CSCO/Linksys router port, via a cat5 cable. The HTPC is plugged directly to port on the sub-tending router a couple of rooms away. all the other devices, most of the time, are WIFI only [three ipads, several 'droids, winxpp, win8.1, win7, no macs!]

As you found out, network bridges cut the bandwidth in half, this is why you only see 150Mpbs though the bridge. This is why a wired access point is better than a non wired network bridge.

Dual band has it's limitations with the 5GHz band not being capable of going through walls like 2.4GHz does. The 5GHz band is made for large open areas outdoors or to cover large rooms indoors, the wider bandwidth loses much more signal strength going through walls.

 

[bawldiggle]

Thanks Golden and Chev65
I learnt something too.
I will post my questions in a new thread
ie, http://www.sevenforums.com/network-...wifi-switch-do-i-need-bridge.html#post2633777