Overclock.net › Forums › General Hardware › General Processor Discussions › Combining CPU Cores for better performance.
New Posts  All Forums:Forum Nav:

Combining CPU Cores for better performance.

post #1 of 10
Thread Starter 
I was wondering if it is possible, or done already, combining the cores of a CPU to give better performance, just like SLI and CF.

The game developers then can decide whether they want fewer combines yet more powerful cores (combination) or simply more threads.
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
post #2 of 10
Quote:
Originally Posted by Deni View Post

I was wondering if it is possible, or done already, combining the cores of a CPU to give better performance, just like SLI and CF.

The game developers then can decide whether they want fewer combines yet more powerful cores (combination) or simply more threads.

In a sense, this technically already exists. Essentially, a multi-core CPU is nothing but several processors on the same chip. Now, if you're talking about two separate multi-core processors working with each other for more processing power, there are multi-socket motherboards that do this, but those are primarily used in large server towers and GPU farms.
post #3 of 10
Thread Starter 
Quote:
Originally Posted by charlie97 View Post

In a sense, this technically already exists. Essentially, a multi-core CPU is nothing but several processors on the same chip. Now, if you're talking about two separate multi-core processors working with each other for more processing power, there are multi-socket motherboards that do this, but those are primarily used in large server towers and GPU farms.

What I meant is two or more cores splitting a single thread equally in order to finish it in a shorter time, i.e. faster.
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
post #4 of 10
No, that's not how things work.
Gaming
(17 items)
 
Gaming PC
(20 items)
 
 
CPUMotherboardGraphicsRAM
7700K AS Rock Z170 OC Formula Titan X Pascal 2050MHz 64GB DDR4-3200 14-14-14-34-1T 
Hard DriveHard DriveHard DriveCooling
950 EVO m.2 OS drive 850 EVO 1TB games drive Intel 730 series 500GB games drive Custom water cooling 
OSMonitorKeyboardPower
Win 10 Pro x64 AMH A399U E-Element mechanical, black switches, Vortex b... EVGA G3 1kw 
CaseMouseAudioAudio
Lian-Li PC-V1000L Redragon M901 LH Labs Pulse X Infinity DAC Custom built balanced tube amp with SS diamond ... 
Audio
MrSpeakers Alpha Prime 
  hide details  
Reply
Gaming
(17 items)
 
Gaming PC
(20 items)
 
 
CPUMotherboardGraphicsRAM
7700K AS Rock Z170 OC Formula Titan X Pascal 2050MHz 64GB DDR4-3200 14-14-14-34-1T 
Hard DriveHard DriveHard DriveCooling
950 EVO m.2 OS drive 850 EVO 1TB games drive Intel 730 series 500GB games drive Custom water cooling 
OSMonitorKeyboardPower
Win 10 Pro x64 AMH A399U E-Element mechanical, black switches, Vortex b... EVGA G3 1kw 
CaseMouseAudioAudio
Lian-Li PC-V1000L Redragon M901 LH Labs Pulse X Infinity DAC Custom built balanced tube amp with SS diamond ... 
Audio
MrSpeakers Alpha Prime 
  hide details  
Reply
post #5 of 10
Thread Starter 
Quote:
Originally Posted by EniGma1987 View Post

No, that's not how things work.

Why biggrin.gif
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
post #6 of 10
Quote:
Originally Posted by Deni View Post

What I meant is two or more cores splitting a single thread equally in order to finish it in a shorter time, i.e. faster.
The definition of a thread is... "the smallest sequence of programmed instructions that can be managed independently by an operating system scheduler."


Crossfire and SLI work because the GPU received work requests that can easily be broken into hundreds/thousands of threads.




Example..... "Make me a sandwich."
Get bread.
Get meat.
Get cheese.
Get condiments.
Place bread.
Place meat.
Place cheese.
Place meat.
Place condiments.
Place bread.
Deliver.

You can have multiple people get the bread, meat, cheese, and condiments in parallel but the rest of the process is basically sequential. (You CAN parallelize the place of the sandwich but that drives up complexity drastically.) However, you cannot break down "get bread" into a smaller sequence of action.... that's a thread.
Edited by DuckieHo - 11/6/13 at 6:54am
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #7 of 10
Quote:
Originally Posted by Deni View Post

What I meant is two or more cores splitting a single thread equally in order to finish it in a shorter time, i.e. faster.

This is impossible.
Quote:
Originally Posted by Deni View Post

Why biggrin.gif

Coherency.

Duplicating the state of one thread across two (or more) cores, and keeping them updated in real-time, would cause a vastly greater performance hit than one could hope to overcome by the combining of resources.

Think about it, registers and L1 caches have access penalties in the ballpark of 1-10 CPU cycles. Reaching across a shared L2 or L3 to even see exactly what another core is doing takes at least 10-20 times as long (and often much longer). Doing this frequently, and syncing changes, would utterly cripple performance.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.3/3.6GHz core/uncore, 1.225/1.2v Gigabyte X99 SOC Champion (F22n) 2x Sapphire R9 290X Tri-X OC New Edition (10036... 4x4GiB Crucial @ 2667, 12-11-12-27-T1, 1.37v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.05) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Cooler Master Nepton 280L 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Reference R9 290X w/Stilt's MLU 1000e / 1375m E... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.3/3.6GHz core/uncore, 1.225/1.2v Gigabyte X99 SOC Champion (F22n) 2x Sapphire R9 290X Tri-X OC New Edition (10036... 4x4GiB Crucial @ 2667, 12-11-12-27-T1, 1.37v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.05) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Cooler Master Nepton 280L 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Reference R9 290X w/Stilt's MLU 1000e / 1375m E... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #8 of 10
Thread Starter 
Quote:
Originally Posted by DuckieHo View Post

The definition of a thread is... "the smallest sequence of programmed instructions that can be managed independently by an operating system scheduler."


Crossfire and SLI work because the GPU received work requests that can easily be broken into hundreds/thousands of threads.




Example..... "Make me a sandwich."
Get bread.
Get meat.
Get cheese.
Get condiments.
Place bread.
Place meat.
Place cheese.
Place meat.
Place condiments.
Place bread.
Deliver.

You can have multiple people get the bread, meat, cheese, and condiments in parallel but the rest of the process is basically sequential. (You CAN parallelize the place of the sandwich but that drives up complexity drastically.) However, you cannot break down "get bread" into a smaller sequence of action.... that's a thread.

Clear smile.gif. Thank you thumb.gif
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
Current
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon x3363 Asus P5K XFX 5770 @1000/1440 Corsair Dominator 
Hard DriveOptical DriveCoolingOS
1TB Samsung F3 Samsung 24x Cooler Master 212+ Windows 7 Ultimate 
MonitorCaseOther
Acer 22" 1680x1050 Noname Genius SW-G2.1 1250 
  hide details  
Reply
post #9 of 10
Quote:
Originally Posted by Blameless View Post

This is impossible.
Coherency.

Duplicating the state of one thread across two (or more) cores, and keeping them updated in real-time, would cause a vastly greater performance hit than one could hope to overcome by the combining of resources.

Think about it, registers and L1 caches have access penalties in the ballpark of 1-10 CPU cycles. Reaching across a shared L2 or L3 to even see exactly what another core is doing takes at least 10-20 times as long (and often much longer). Doing this frequently, and syncing changes, would utterly cripple performance.

Just a note.... one approach to distributed processing is duplication though. I believe MapReduce uses this duplication concept to get data back faster.
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #10 of 10
Quote:
Originally Posted by DuckieHo View Post

Just a note.... one approach to distributed processing is duplication though. I believe MapReduce uses this duplication concept to get data back faster.

I think this is more for redundancy than anything.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.3/3.6GHz core/uncore, 1.225/1.2v Gigabyte X99 SOC Champion (F22n) 2x Sapphire R9 290X Tri-X OC New Edition (10036... 4x4GiB Crucial @ 2667, 12-11-12-27-T1, 1.37v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.05) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Cooler Master Nepton 280L 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Reference R9 290X w/Stilt's MLU 1000e / 1375m E... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.3/3.6GHz core/uncore, 1.225/1.2v Gigabyte X99 SOC Champion (F22n) 2x Sapphire R9 290X Tri-X OC New Edition (10036... 4x4GiB Crucial @ 2667, 12-11-12-27-T1, 1.37v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.05) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Cooler Master Nepton 280L 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Reference R9 290X w/Stilt's MLU 1000e / 1375m E... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: General Processor Discussions
Overclock.net › Forums › General Hardware › General Processor Discussions › Combining CPU Cores for better performance.