EVGA

GAMING ON A LOCAL "NODE"

Author
Wool Clad Wolf
New Member
  • Total Posts : 17
  • Reward points : 0
  • Joined: 2021/05/04 04:55:10
  • Location: Oregon
  • Status: offline
  • Ribbons : 0
Flagged as Spam (2)
2021/05/04 05:34:06 (permalink)
SR/IOV, TB/USB4, PCIE NUC DEV KITS, ??? others???
In , a single system image (SSI) cluster is a  of machines that appears to be one single system,
I'm sure these thoughts have been previously "thunk", are "consumer" level nodes possible/ feasible for daily computing/ gaming  (with AND without consideration of "plug and play" or "It just works" or user friendly Graphical user interface environments 
 

Attached Image(s)

#1

8 Replies Related Threads

    houkom
    FTW Member
    • Total Posts : 1275
    • Reward points : 0
    • Joined: 2007/09/13 13:49:52
    • Location: 01101000011011110110110101100101
    • Status: offline
    • Ribbons : 4
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 06:07:00 (permalink)
    I'm genuinely not sure what you're getting at... but... I'll take a stab at it. 
     
    There's been a few people who have made things like "6 gamers 1 PC" (Linus Tech Tips comes to mind)... where you have 1 physical machine but have provisioned out 6 individual COMPUTERS within it and everyone had pretty stellar performance. The cost was prohibitive but rather the concept behind it was rather cool to watch. That being said.. I'm not sure that the reverse would work so well.... Taking multiple different machines and "stitching them" together to make a "single" computer would present some challenges as you would have multiple different levels of hardware trying to intermingle. I know that companies are able to do this with super computers but to be honest, they are more than likely all the same hardware. Doing this at the consumer level to play games would seem rather inefficient and clunky. You may be able to pull this off fairly easily for actual compute performance (the term Beowulf Cluster comes to mind) but in order to maintain an ideal level of performance for GAMING would probably require far more hardware/cost/ingenuity than just building a single machine to play games and then a computer cluster for other resource intensive tasks.
     
    I hope I got close to what you were talking about. 

    9900k @ 5.0Ghz, 1.30V, no AVX offset | 32GB Trident Z 3200mhz | ASUS Maximus XI CODE | RTX 3090 FTW3 ULTRA GAMING | 1TB Samsung 970 Evo Plus NVMe | Corsair AX1200 | Cooler Master Cosmos II | Corsair H100 | ASUS ROG Swift PG278Q
     

     
    #2
    Wool Clad Wolf
    New Member
    • Total Posts : 17
    • Reward points : 0
    • Joined: 2021/05/04 04:55:10
    • Location: Oregon
    • Status: offline
    • Ribbons : 0
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 06:51:33 (permalink)
    hey sort of the opposite, in that scenario there is one physical machines cut up into a bunch of synthetic little ones 
     
    what I'm interested in is multiple physicals "machines" shared "locally" (in the same location) communicating amongst or within one another running one instance of an OS "~seamlessly~" across/amongst this one "node/computer"
     
    P.s i hope your human and not a bot promoting LTT
    #3
    Brad_Hawthorne
    Insert Custom Title Here
    • Total Posts : 22747
    • Reward points : 0
    • Joined: 2004/06/06 16:13:06
    • Location: Dazed & Confused
    • Status: offline
    • Ribbons : 37
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 06:51:50 (permalink)
    Ok, can someone translate that from word salad to English?
    #4
    Wool Clad Wolf
    New Member
    • Total Posts : 17
    • Reward points : 0
    • Joined: 2021/05/04 04:55:10
    • Location: Oregon
    • Status: offline
    • Ribbons : 0
    Flagged as Spam (1)
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 06:56:17 (permalink)
    in , a single system image (SSI) cluster is a  cluster(group) of machines that appears to be one single system
     
    Some SSI systems allow processes on different nodes to communicate using  mechanisms as if they were running on the same machine
     
    trust me this isn't a ltt kind of topic, any resource for  Computing dictionary's, acronyms of computing? the medical dictionary for doctorates of PCs    and idk where to find "Whetzel" (sorry for the botched name)
    post edited by Wool Clad Wolf - 2021/05/04 07:01:27
    #5
    castrator86
    SSC Member
    • Total Posts : 738
    • Reward points : 0
    • Joined: 2010/07/24 09:33:21
    • Status: online
    • Ribbons : 2
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 07:48:32 (permalink)
    You could do this. It's extremely cost prohibitive. You'd need to build a server that can take the load of all the clients that want to leverage its hardware to run games.

    You'd have to work out a way to remote into one of the VMs hosted by your server through client PCs; you could get away with NUCS, you don't need any processing power on the clients; but you'd need a hell of a good network (fiber) and you'd need to be running Threadripper with SLI GPUs to handle all of it.



    #6
    Wool Clad Wolf
    New Member
    • Total Posts : 17
    • Reward points : 0
    • Joined: 2021/05/04 04:55:10
    • Location: Oregon
    • Status: offline
    • Ribbons : 0
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 08:31:44 (permalink)
    (I'm familiar with) Threadripper, did you suggest it as you knew it could support Sli or because the of the number of pcie lanes over consumer options also be leveraged? 
    I'm currently running on a Asus WS Z390 Pro thunderbolt add in ready and a broadcom chip
    The ExpressLane™ PEX 8747 is a 48-lane, 5-port, PCIe Gen 3 switch device developed on 40nm technology. The PEX 8747 device offers Multi-Host PCI Express switching capability enabling users to connect a host to its respective endpoints via scalable, high bandwidth,
    non-blocking interconnection to a variety of graphics applications. The PEX 8747 is optimized to support high-resolution graphics while supporting peer-to-peer traffic and multicast for maximum performance. Included is PLX's proprietary visionPAK, 
    MOBO block diagram   
     
    brodcom 8747 info    
     
     
    with intel not viable until lga 1700 matures(or later) and AMD mainstream being some, but not justifiable improvement in performance(yet~), and Thredrippers all day attainable OC mhz frequencys/ipc likley are not amazing and might hinder performance on a "(drempt)gaming node)
     
     
    i guess url's are BANNED if not addvertising evgas products,,, and after some time trying tto convert and compress things i could get 1/5 thing to upload so i did none,,, and included (vanishing)urls ://=P
    ill try block again
     
     
    post edited by Wool Clad Wolf - 2021/05/04 08:40:17

    Attached Image(s)

    #7
    castrator86
    SSC Member
    • Total Posts : 738
    • Reward points : 0
    • Joined: 2010/07/24 09:33:21
    • Status: online
    • Ribbons : 2
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 10:36:06 (permalink)
    The PCI-e lanes you'll need to leverage would lead you to Intel HEDT or Threadripper plaftorms. You need the CPU cores AND the multi-GPU to get any sort of respectable framerates if you're going to be hooking up 4-6 clients.



    #8
    Wool Clad Wolf
    New Member
    • Total Posts : 17
    • Reward points : 0
    • Joined: 2021/05/04 04:55:10
    • Location: Oregon
    • Status: offline
    • Ribbons : 0
    Re: GAMING ON A LOCAL "NODE" 2021/05/04 10:44:58 (permalink)
    dang.. i thought i sorted this out after the first guy, multiple physical compute units appearing as one just one selfish gamer gz
     
     

    Working together as an operating system

    The architecture and design of a distributed operating system must realize both individual node and global system goals. Architecture and design must be approached in a manner consistent with separating policy and mechanism. In doing so, a distributed operating system attempts to provide an efficient and reliable distributed computing framework allowing for an absolute minimal user awareness of the underlying command and control efforts.
    The multi-level collaboration between a kernel and the system management components, and in turn between the distinct nodes in a distributed operating system is the functional challenge of the distributed operating system. This is the point in the system that must maintain a perfect harmony of purpose, and simultaneously maintain a complete disconnect of intent from implementation. This challenge is the distributed operating system's opportunity to produce the foundation and framework for a reliable, efficient, available, robust, extensible, and scalable system. However, this opportunity comes at a very high cost in complexity.

    The price of complexity

    In a distributed operating system, the exceptional degree of inherent complexity could easily render the entire system an anathema to any user. As such, the logical price of realizing a distributed operation system must be calculated in terms of overcoming vast amounts of complexity in many areas, and on many levels. This calculation includes the depth, breadth, and range of design investment and architectural planning required in achieving even the most modest implementation.
    These design and development considerations are critical and unforgiving. For instance, a deep understanding of a distributed operating system's overall architectural and design detail is required at an exceptionally early point. An exhausting array of design considerations are inherent in the development of a distributed operating system. Each of these design considerations can potentially affect many of the others to a significant degree. This leads to a massive effort in balanced approach, in terms of the individual design considerations, and many of their permutations. As an aid in this effort, most rely on documented experience and research in distributed computing power.
     
    CREDIT TO .  wikipidia  DOT ORG/wiki/Distributed_operating_system
    #9
    Jump to:
  • Back to Mobile