M.2 SSD vs Regular SSD

[Ozymandias]

M2 is a physical form factor. SATA is both a physical form factor, and a communication protocol. NVMe is a communication protocol.
Some M2 drives talk SATA. And they suck.
M.2 slots can use both SATA and PCI-E. Some motherboards have support for both types others only one type.

10gb/s is interface chip-rate for M2-PCIE. It is higher limit for designers of hard drive controllers.

SATA based M.2 drives aren't any faster than the 2.5" versions. PCIe based ones are faster, but are also more expensive.

M.2 is a form factor, physical standardisation being as heterogeneous, as the one present in drives using the SATA bus. NVMe is to M.2, what AHCI was to SATA, a new way of using the bus, that allows it to stretch its legs. The best M.2 drives, using the SATA bus, suck only when compared to the best ones using PCIe.

Those are the main buses for M.2, but not the only ones, USB comes to mind (I think there're more on the way).

When you talk about gb/s, I guess you mean gigabits per second (Gbit/s, Gb/s, or Gbps), in which case, I don't know about that speed limit.

Agreed, they don't suck more or less than their 2.5" counterparts; as for price differences, you'd be surprised as to how small they can be.

[Ozymandias]

Probing 6-Men with M.2 SSD is faster than regular.

However is the difference enough to be affordable?

M.2 | 10gb/s

Sata III | 6gb/s

PCIe. Not any doubt the way to go.

What about an M.2 drive, sitting on a PCIe slot, using an adapter?

[Ozymandias]

So how much difference gain in terms tablebase probing?

Do TCEC take advantage of PCIe SSDs?

I hope it'll be sizeable, I ordered one yesterday.

Endgame Tablebases
For Season 8 Stage 1-3, tablebases that are available for the engines are: 5-men Syzygy, 5-men Nalimov and 5-men Gaviota (cp2), depending on which each specific engine can use. For some engines specifying more than one type is possible, but here only one is allowed. Nalimov is preferred. For Stage 1 they were hosted in a RAM-drive, for Stage 2 onwards they are hosted on an SSD drive. For the Superfinal, 6-men Syzygy hosted on an SSD drive are available. When watching a game you can see the type of the tablebases (if any) that is in use for the engines currently playing by clicking the gears next to the engine logo.

I doubt that SSD drive was PCIe based.

[Ozymandias]

50-60% faster than the top of line SSD. They are almost 2.5x faster than standard SATA SSD. Thruput can be almost 820MB under ideal conditions. Figure $600 per TB so also 50% more expensive. Just for chess tb probing - a waste of money. But people with more money than brains will buy them for tb probing anyway. Now if used them to backup your data - yes it could be worth it since you always want to minimise your backup and restore times.

50% sounds about right for tb probing. Whether that's a waste of money or not, that's up to consumer, even if they plan on using the drive exclusively for chess.

[Ozymandias]

5-Men with PCIe SSD can be good alternative in engine matches right?

6-Men usually slow down the total speed somewhat.

A senseless waste. Why buy a top of the line drive, if you're only going to use a set of files, that are less than 400MB in size (DTZ can literally be anywhere)?
Not with such a drive, or at least, not in a way that tuns out to be detrimental.

[kgburcham]

I have been using a PCIE SSD for several months now. It is extremely fast in a benchtest. Using Windows 8.1 with about 35 gig cpu and 16 gigs ram
most SSD are about 500 in ATTO bench

http://www.newegg.com/Product/Product.a ... 85V3R91984

[kgburcham]

here is an ATTO bench on my Intel SATA SSD where my EGTB are stored.

[yurikvelo]

Current SSD = speed penalty due to 6-Men probing.
PCIe SSD = cancel out the already applied penalty due to faster speed?

With PCIe SSD my speed can be 100% with 6-Men probing compared to 95/90% with my current SSD.

Even zero-latency (imaginary) storage give speed penalty, because tbcore.cpp + tbprobe.cpp code is CPU-costly

[yurikvelo]

here is an ATTO bench on my Intel SATA SSD where my EGTB are stored.

please run CrystalDiskMark, IOPS in random read mode.
The same question to Jon Fredrik Åsvang

Code: Select all

-----------------------------------------------------------------------
CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo
                           Crystal Dew World : http://crystalmark.info/
-----------------------------------------------------------------------
* MB/s = 1,000,000 byte/s [SATA/300 = 300,000,000 byte/s]
 
           Sequential Read :   243.968 MB/s
          Sequential Write :   127.766 MB/s
         Random Read 512KB :   223.618 MB/s
        Random Write 512KB :   127.005 MB/s
    Random Read 4KB (QD=1) :    12.803 MB/s [  3125.8 IOPS]
   Random Write 4KB (QD=1) :    28.522 MB/s [  6963.5 IOPS]
   Random Read 4KB (QD=32) :    55.477 MB/s [ 13544.1 IOPS]
  Random Write 4KB (QD=32) :   104.684 MB/s [ 25557.6 IOPS]
 
  Test : 1000 MB [E: 0.1% (0.1/111.7 GB)] (x5)
  Date : 2014/12/19 1:28:15
    OS : Windows 7 Ultimate SP1 [6.1 Build 7601] (x64)

[Nordlandia]

kgburcham: can i ask why you don't store tablebases on the fastest SSD aka PCIe SSD?