Multi CPU motherboard question

[numerobis]

What is your transfer rate without Link Aggregation? I reach 102MB/s top with DS412+ (with the same WD Red)

 

Link aggregation can not increase the overall transferrate of Gbit lan for one connection - with a theoretical max of roughly 125MB/s (practically maybe ~110-115MB/s). It can only increase the overall throughput if more than one client is connecting to the switch at the same time or with multiple copy actions if your workstation also has link aggregation.

[Jose Negrete]

Part 2:

I started building my nodes as dual E5-2680 v2, first two should be completed by Friday. I'll post some picks and benchmarks, how they compare to my regular 4930k. I had to already go with xeons because I already have 5x4930k as workstations/ workstation-nodes and I simply do not with to clutter my room with any more boxes to manage. Comfort of silent lower-clocked xeons seemed what I need now. So yay, my first 2p systems :- )

 

Juraj,

 

Did you happen to post these somewhere? I tried search but couldnt find post/thread. Thanks.

[RyderSK]

Juraj,

 

Did you happen to post these somewhere? I tried search but couldnt find post/thread. Thanks.

 

Hi Jose,

 

if you're interested, here is full gallery from my builder : {I think I posted some stuff on here and there, so I apologize if this is seen as spam}

 

 

 

Comp list: {listed in that link, but shortened here}

 

Fractal Design Define XL R2 + ASUS Z9PE-D8 WS, 2x Intel® Xeon® E5-2680 v2 (25M Cache, 2.80 GHz) + 2x Corsair Vengeance 32GB (Kit 4x8GB) Low Prof. 1600MHz DDR3,CL10, XMP + Samsung SSD SSD840 EVO 250GB SATAIII 2.5'', (540MB/s; 520MB/s), 7mm, Basic + Sapphire Radeon HD 6450, 2GB DDR3 (64 Bit) + Super Flower SF-1000P14PE Golden King + 2x Noctua NH-U14S + 3x Noiseblocker NB-BlackSilent PRO PK1

 

Overall, they are extremely silent, and CPUs run 50C during rendering (and could even less if more powerful coolers or 2 fans per cpu coolers were used, but I went for silence).

They overclock to 3100MHz during rendertime for all 20cores/40threads, and achieve 29.5+/- Cinebench 11.5 Score.

To compare it to my workstation, which is 4930k, at overclocked 4.2 (I do not run higher mostly), has 12.3 +/- Score.

Both CoronaA7 and Vray3 scale quite linearly within this performance, so it is almost 3 times faster.

 

Overall, I am very satisfied, and will be building V3 version in October hopefully, + 5960X for workstation, so I will post builds and comparison again.