Benefits of SATA Express?

[Blueprint]

I was looking at the new Z97 motherboards and noticed one had SATA Express up to 10gbit/s.

 

Where would you benefit with 10gbit/s as opposed to 6gbit/s?

Doesn't your drive's speed limit the data transfer rate anyway? How would you use the 10gbit/s speed to its maximum potential?

 

Thanks,

Blueprint

[Freaky_spider]

there will be SSD's that can go up to that speed, ssd's where never stuck at 550 mb/s, there just wasnt a reason to go higher because the interface couldnt go that high

[jscho]

I personally think SATA Express is silly. It's such an unnecessarily large connector for 10Gbps. M.2 looks to be way better.

I'd just RAID two SSDs together and be done with it if I wanted moar speed out of my SSDs (the drawbacks of RAID 0 don't worry me) because I'd imagine that SATAEX SSDs would come at a rather hefty premium...

[Rheinwasser]

i dont think that there will be much of an real world perfomance increase. maybe if you often copy 100gb big files to your ssd, but otherwise i dont see me upgrading.

[alpenwasser]

SSDs have been bottlenecked by the 6 Gbit/s limit for quite a while

now. If you look at existing PCI-E SSDs, the technology has the

potential to be quite a lot faster than that.

But, that's only really relevant with large sequential transfers,

you're not going to be seeing much of an improvement for random

small file access as far as I can tell.

[Guest]

Zero. Literally zero. SSD's aren't all that fast for the majoirty of use cases. Sure they're fine for sequential reads and writes however random read and random write is no where near taking advantage of the existing SATA 6 gigabit infrastructure. Intel are just trying to throw features at the platform so they can release a refresh. I was at an Intel conference tonight and one of the things we covered was the changes to Z97/H97 over Z87/H87 and besides the addition of SATA Express, mildly faster CPU's and a new security structure that many consumers won't take advantage of, there was nothing new. Intel are scratching in the dirt for new innovative things to bring to market and joining two SATA lanes together to give more bandwidth for those sequential speeds was what their R&D team came up with. 

[wpirobotbuilder]

Until SSD controller complexity increases, we will be stuck with 10-40 MB/s for random operations (most operations are random read/write), which is way slower than SATA III.

 

I can see SATA express being useful in enterprise environments where the load on the SSDs is high enough to be bottlenecked by SATA III, since the only other option is (very expensive) PCI-E cards from the likes of FusionIO. But I will give exactly zero cares until the aforementioned controller complexity increases and delivers us 50-100 MB/s in random operations. That would be a noticeably snappy SSD compared to current SSDs, and would end the debate about choosing one SSD over another.