Jump to content

Nehalem (microarchitecture)

From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Jdenyc (talk | contribs) at 14:36, 26 August 2010 (→32 nm processors). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Nehalem (Template:IPA-en[1]) is the codename for an Intel processor microarchitecture, successor to the Core microarchitecture.[2] The first processor released with the Nehalem architecture was the desktop Core i7,[3] which was released in November 2008. It was followed by several Xeon processors and by the i3 and i5.[4]

Initial Nehalem processors use the same 45 nm manufacturing methods as Penryn. A working system with two Nehalem processors was shown at Intel Developer Forum Fall 2007,[5][6] and a large number of Nehalem systems were shown at Computex in June 2008.

The microarchitecture is named after the Nehalem Native American nation in Oregon.[7][8] At that stage it was supposed to be the latest evolution of the NetBurst microarchitecture. Since the abandonment of NetBurst, the codename has been recycled and refers to a completely different project, although Nehalem still has some things in common with NetBurst. Nehalem-based microprocessors utilize higher clock speeds and are more energy-efficient than Penryn microprocessors. Hyper-Threading is reintroduced along with an L3 Cache missing from most Core-based microprocessors.

The first computer to use Nehalem-based Xeon processors was the Apple Mac Pro workstation announced on March 3, 2009.[9] Nehalem-based Xeon EX processors for larger servers were originally expected in Q4 2009[10], However, they were released in April of 2010[11].

Mobile Nehalem-based processors were introduced in September 2009.

Technology

Microarchitecture of the quad-core implementation

Various sources have stated the specifications of processors in the Nehalem family:

Performance and power improvements

It has been reported that Nehalem has a focus on performance, which accounts for the increased core size.[16] Compared to Penryn, Nehalem has:

  • 1.1× to 1.25× the single-threaded performance or 1.2× to 2× the multithreaded performance at the same power level
  • 30% lower power usage for the same performance
  • According to a preview from AnandTech "expect a 20–30% overall advantage over Penryn with a 10% increase in power usage."[17]
  • Per Core, clock-for-clock, Nehalem provides a 15–20% increase in performance compared to Penryn.[18]

PC Watch found that a Nehalem "Gainestown" processor has 1.6× the SPECint_rate2006 integer performance and 2.4× the SPECfp_rate_2006 floating-point performance of a 3.0 GHz Xeon X5365 "Clovertown" quad-core processor.[16]

A 2.93 GHz Nehalem "Bloomfield" system has been used to run a 3DMark Vantage benchmark and gave a CPU score of 17,966.[19] The 2.66 GHz variant scores 16,294. A 2.4 GHz Core 2 Duo E6600 scores 4,300.[20]

AnandTech tested the Intel QuickPath Interconnect ("QPI", 4.8 GT/s version) and found the copy bandwidth using triple-channel 1066 MHz DDR3 was 12.0 GB/s. A 3.0 GHz Core 2 Quad system using dual-channel 1066 MHz DDR3 achieved 6.9 GB/s.[21]

Overclocking is possible with Bloomfield processors and the X58 chipset. The Lynnfield processor uses a PCH removing the need for a northbridge chipset.[22]

The Nehalem processors are the first to incorporate the SSE 4.2 SIMD instructions, adding 7 new instructions to the SSE 4.1 set available in the Core 2 series.

The Nehalem architecture also significantly reduces atomic operation latency; cutting it to 50% in an attempt to eliminate atomic overhead [23].

Code names

Each combination of a Nehalem/Westmere processor die and package has both a separate codename and a product code. Typically, the same dies are used for uniprocessor (UP) and dual-processor (DP) servers, but using an extra QuickPath link for the inter-processor communication in the DP server variant. Like the Core microarchitecture, Nehalem also uses four different processor packages, one for each market segment. The high-end chips with eight cores are only available in the MP server segment, while all others are used in at least two different packages. No details have been released about the Westmere EX processor so far.

Mobile Desktop
UP Server		 DP Server MP Server
Eight-Core 45 nm
Quad-Channel 		Beckton
80604
Quad-Core 45 nm
Triple-Channel 		Bloomfield
80601		 Gainestown
80602
Quad-Core 45 nm
Dual-Channel, PCIe 		Clarksfield
80607		 Lynnfield
80605		 Jasper Forest
80612
Dual-Core 45 nm
Dual-Channel, PCIe, Graphics Core 		Auburndale
canceled		 Havendale
canceled
Twelve/Ten-Core 32 nm
?-channel 		Westmere-EX
?
Six-Core 32 nm
Triple-Channel 		Gulftown
80613		 Westmere-EP
80614
Dual-Core 32 nm
Dual-Channel, PCIe, Graphics Core 		Arrandale
80617		 Clarkdale
80616

Variants

These tables list all the processors of Nehalem microarchitecture to have been leaked so far. The table is ordered roughly by performance, which usually correlates with price and power. Released processors are set in bold.

Notes:

45 nm processors

Codename Market Cores
(Threads) 		Socket Brand Processor No. Clock rate Turbo TDP Interfaces L3 cache Release 1k Unit Price
Base Core Uncore Chipset Memory PCIe
Beckton MP Server, DP Server 8 (16) LGA 1567 Xeon [24] X7560 133 MHz 2.266 GHz Yes 130 W 4× QPI 4× [DDR3 with SMB motherboard] — 24 MB March 30, 2010[25] $3692
X7550 2 GHz 18 MB $2837
Gainestown DP Server[26] 4 (8) LGA 1366 Xeon[27] W5590 133 MHz 3.33 GHz Yes 130 W 2× QPI 6.4 GT/s 3× DDR3-13331 — 8 MB August 9, 2009 $1600
W5580 3.2 GHz March 29, 2009[28] $1500
X5570 2.93 GHz 95 W $1286
X5560 2.8 GHz $1072
X5550 2.66 GHz $858
E5540 2.53 GHz 80 W 2× 5.86 GT/s 3× DDR3-10661 $744
E5530 2.4 GHz $530
E5520 2.26 GHz $373
L5520 2.26 GHz 60 W $530
4 (4) E5506 2.13 GHz No 80 W 2× 4.8 GT/s 3× DDR3-8001 4 MB $266
E5504 2.0 GHz $224
L5506 2.13 GHz 60 W $423
2 (2) E5502 1.86 GHz 80 W $188
Bloomfield UP Server[29] 4 (8) LGA 1366 Xeon[30] W3580 133 MHz 3.33 GHz Yes 130 W 1× QPI 6.4 GT/s 3× DDR3-1333 — 8 MB August 9, 2009 $999
W3570 3.2 GHz March 29, 2009[30] $999
W3550 3.06 GHz 1× QPI 4.8 GT/s 3× DDR3-1066 August 9, 2009 $562
W3540 2.93 GHz March 29, 2009[30] $562
W3520 2.66 GHz $284
Lynnfield LGA 1156 X3470 2.93 GHz 95 W DMI 2× DDR3-1333 September 8, 2009 $589
X3460 2.8 GHz $316
X3450 2.66 GHz $241
X3440 2.53 GHz $215
L3426 1.86 GHz 45 W $284
4 (4) X3430 2.4 GHz 95 W $189
Bloomfield Enthusiast Desktop[31] 4 (8) LGA 1366 Core i7 Extreme 975[32] 133 MHz 3.33 GHz 2.66 GHz Yes 130 W 1× QPI 6.4 GT/s 3× DDR3-1066 — 8 MB May 31, 2009 $999
965 3.2 GHz November 17, 2008 $999
Core i7 960[33] 3.2 GHz 2.13 GHz 1× QPI 4.8 GT/s October 20, 2009 $562
950[32] 3.06 GHz May 31, 2009 $562
940 2.93 GHz November 17, 2008 $562
930 2.8 GHz February 28, 2010 $294
920 2.66 GHz November 17, 2008 $284
Lynnfield Performance Desktop 4 (8) LGA 1156 880 133 MHz 3.06 GHz Yes 95 W DMI 2× DDR3-1333 1 ×16 / 2 ×8 8 MB $583
875K 2.93 GHz $342
870[34] 2.4 GHz September 8, 2009 $562
860 2.8 GHz $284
860S 2.53 GHz 82 W January 7, 2010 $337
4 (4) Core i5 760 2.8 GHz 95 W July 17, 2010 $209
750[35] 2.66 GHz 2.13 GHz 95 W September 8, 2009 $196
750S 2.4 GHz 82 W January 7, 2010 $259
Clarksfield Extreme/Performance Mobile 4 (8) mPGA-989 Core i7 Extreme 940XM 2.13 GHz 55 W Jun.21.2010 $1096
920XM 2.0 GHz Sep.23.2009 $1054
Core i7 840QM 1.86 GHz 45 W Jun.21.2010 $568
820QM 1.73 GHz Sep.23.2009 $546
740QM 6 MB Jun.21.2010 $378
720QM 1.6 GHz Sep.23.2009 $364
  • 1 Though there is only one memory controller and it has only three channels, Intel states the Gainestown processors have six memory channels. Gainestown processors have dual QPI links and have a separate set of memory registers for each link[36][37]; in effect, a multiplexed six-channel system.

The Havendale and Auburndale variants (which contained Gilo and Ironlake) have been cancelled.[38]

32 nm processors

Codename Market Cores
(Threads) 		Socket Brand Processor No. Clock rate Turbo TDP(W) Interfaces L3 cache Release 1k Unit Price
Base Core GPU Chipset Memory PCIe
Westmere-EX[39] MP Server 12 (24) or 10 (20) [40] LGA 1567 [41] Xeon ? ? ? — — ? ?×QPI ?×DDR3 — ? Q2 2011 ?
Gulftown

(Westmere-EP)[42] 		DP Server 6 (12) LGA-1366 Xeon X5690 133 MHz 3.46 GHz — Yes 130 W 2× QPI 6.4 GT/s 3× DDR3-1333 — 12 MB ?? ??
X5680 3.33 GHz March 16, 2010 $1663
X5670 2.93 GHz 95 W $1440
X5660 2.8 GHz $1219
X5650 2.66 GHz $996
E5645 2.4 GHz 80 W 2× QPI 5.86 GT/s $958
L5640 2.26 GHz 60 W $996
L5638 2.0 GHz $958
4 (8) X5677 3.46 GHz 130 W 2× QPI 6.4 GT/s $1663
X5667 3.06 GHz 95 W $1440
E5640 2.66 GHz 80 W 2× QPI 5.86 GT/s 3× DDR3-1066 $774
E5630 2.53 GHz $551
E5620 2.4 GHz $387
L5630 2.13 GHz 40 W $551
L5618 1.86 GHz $530
4 (4) L5609 1.86 GHz No 2× QPI 4.8 GT/s $440
UP Server 6 (12) LGA-1366 Xeon W3680 133 MHz 3.33 GHz — Yes 130 W 1× QPI 6.4 GT/s 3× DDR3-1333 — 12 MB March 16, 2010[43] $999
Extreme/Performance Desktop Core i7 Extreme 980X 3× DDR3-1066
Core i7 970 3.2 GHz 1× QPI 4.8 GT/s July 17,2010 $899
Clarkdale[44] Mainstream/Value Desktop 2 (4) LGA-1156 Core i5 680 133 MHz 3.6 GHz 733 MHz Yes 73 W DMI 2× DDR3-1333 1 ×16 4 MB April 18, 2010 $294
670 3.46 GHz January 7, 2010 $284
661 3.33 GHz 900 MHz 87 W $196
660 733 MHz 73 W
650 3.2 GHz $176
Core i3 540 3.06 GHz No $133
530 2.93 GHz $113
2 (2) Pentium G6950 2.8 GHz 533 MHz 2× DDR3-1066 3 MB $87
Celeron G1101 2.26 GHz 2 MB OEM
Arrandale[45] Mainstream/Value Mobile 2 (4) mPGA-989 Core i7 620M 133 MHz 2.66 GHz 766 MHz Yes 35 W DMI 2× DDR3-1066 1 ×16 4 MB January 7, 2010 $332
640LM 2.13 GHz 566 MHz 25 W $332
620LM 2.0 GHz $300
640UM 1.2 GHz 500 MHz 18 W 2× DDR3-800 $305
620UM 1.06 GHz $278
Core i5 540M 2.53 GHz 766 MHz 35 W 2× DDR3-1066 3 MB $257
520M 2.4 GHz $225
520UM 1.06 GHz 500 MHz 18 W 2× DDR3-800 $241
430M 2.26 GHz 766 MHz 35 W 2× DDR3-1066 OEM
Core i3 350M 2.26 GHz 667 MHz No
330M 2.13 GHz
2 (2) Celeron P4500 / P4505 1.86 Ghz 500 Mhz 2 MB March 28, 2010 $86 / OEM

For the desktop, Gulftown is to be an "Extreme Edition" CPU and so will coexist with Bloomfield.[46] It will have Turbo Boost and similar clock speeds to Bloomfield.[47]

Lynnfield and Clarksfield may make the 32 nm transition in the middle of 2010[45], sometime after Q2[46], while Beckton will move to 32 nm at the end of 2010.[48] The 32 nm CPUs will not have significantly different clock speeds compared to 45 nm CPUs.[49]

Clarkdale and Arrandale contain the 32 nm dual core processor Hillel and the 45 nm integrated graphics device Ironlake, and support switchable graphics.[38][49][50] The lowest-power variant of Arrandale may have a 10 W CPU TDP, and a maximum clock speed of 1.6 GHz.[51]

A successor to Bloomfield and entry level server chips are also expected in Q2 2010.[38][45][48]

Westmere

Westmere (formerly Nehalem-C) is the name given to the 32 nm die shrink of Nehalem. The first Westmere-based processors were launched on January 7, 2010 and branded as members of the Core i3, Core i5, and dual-core mobile Core i7 families.

Westmere's features and improvements from Nehalem have been reported as follows:

  • Native six-core, and possibly dual-die hex-core (12-cores), processors.[52]
    • The successor to Bloomfield and Gainestown is six-core.
  • A new set of instructions that gives over 3x the encryption and decryption rate of Advanced Encryption Standard (AES) processes compared to before.[53]
    • Delivers seven new instructions (AES instruction set or AES-NI) that will be used by the AES algorithm. Also an instruction called PCLMULQDQ (see CLMUL instruction set) that will perform carry-less multiplication for use in cryptography.[54] These instructions will allow the processor to perform hardware-accelerated encryption, not only resulting in faster execution but also protecting against software targeted attacks.
    • AES-NI may be included in the integrated graphics of Westmere.
  • Integrated graphics, released at the same time as the processor.
  • Improved virtualization latency.[55]
  • New virtualization capability: "VMX Unrestricted mode support," which allows 16-bit guests to run (real mode and big real mode).
  • Support for "Huge Pages" of 1GB in size.

Successor

The successor to Nehalem and Westmere will be Sandy Bridge, scheduled for release in late 2010, according to statements by Intel.[56] [57] The successor to Sandy Bridge will be Haswell, scheduled for release in 2012. It will come with a new cache subsystem, a FMA (fused multiply-add) unit, and a vector coprocessor.[18]

Intel CPU core roadmaps from NetBurst to Haswell



References

  1. ^ I Am Nehalem
  2. ^ Intel Details Upcoming New Processor Generations, Intel Corporation, 2007-03-28
  3. ^ Gruener, Wolfgang (2008-08-10), Nehalem = i7: Intel unveils new Core processor brand, TG Daily
  4. ^ Hillis, Scott (2008-05-21), Intel says Nehalem chips on track for '08 2nd half, Reuters {{citation}}: Unknown parameter |coauthor= ignored (|author= suggested) (help)
  5. ^ Lal Shimpi, Anand (2007-09-18), IDF 2007 – Day 1: Nehalem, Intel's GPUs, 32nm and More, AnandTech
  6. ^ Roadmaps of AMD and Intel CPU in 2010, Extra Reading Material, December 24, 2009
  7. ^ Ian, King (2008-10-20), Intel's new faster chip right on AMD's heels, The Seattle Times
  8. ^ Jones, George (2008-02-09), IAMD vs Intel: The future of desktop CPUs, PC Advisor UK
  9. ^ Crothers, Brooke (2009-03-03), Apple beats Intel to Nehalem-EP chip launch
  10. ^ Morgan, Timothy (2009-02-10), Intel confirms Nehalem Xeons imminent
  11. ^ Intel Unveils Plans for Six-Core Nehalem-EX – InternetNews.com
  12. ^ Laird, Jeremy (2007-09-19), IDF: Not all Intel Nehalem CPUs single-die, Tech.co.uk
  13. ^ Laird, Jeremy (2007-07-19), IDF: Intel reveals eight-core PC processor, TechRadar UK
  14. ^ a b Lal Shimpi (2008-03-17), Opening the Kimono: Intel Details Nehalem and Tempts with Larrabee, AnandTech {{citation}}: More than one of |author= and |last= specified (help)
  15. ^ Gelsinger, Patrick P. (2008-03-17), "Intel Architecture Press Briefing" (PDF), Intel
  16. ^ a b "Intel's dual teamed approached to micro-architecture development", PC Watch (in Japanese), 2008-01-29
  17. ^ Lal Shimpi, Anand (2008-06-05), The Nehalem Preview: Intel Does It Again, AnandTech
  18. ^ a b Doc Teraboule (2008-04-08), IDF 2008 Shanghaï – From Nehalem to Haswell (in French), Canard PC, retrieved 2009-03-01
  19. ^ Nguyen, Tuan (2008-07-09), 'Nehalem' 2.93 GHz Benches Revealed, Tom's Hardware
  20. ^ Intel Bloomfield 2.66 GHz: First Comprehensive Evaluation, techPowerUp!, 2008-07-09
  21. ^ Intel does it again, AnandTech, 2008-06-05
  22. ^ Botezatu, Bogdan (2008-04-22), Intel: No Overclocking for Mainstream Nehalems, Softpedia
  23. ^ NO EXECUTE!
  24. ^ AMD launches 12-core Opteron server chips, Intel counters with the 8-core Xeon 7500, engadedget.com, 2010-03-31, retrieved 2010-03-31
  25. ^ Nebojsa Novakovic (2009-02-12), Intel's next bunch of fun CPUs moves to 2010, CNN International, retrieved 2009-03-01
  26. ^ Intel Xeon Processor 5500 Series Product brief, (Document Number: 321579-001US) (PDF), Intel, 2009-04-01, retrieved 2009-04-10
  27. ^ Core i7 to make leap to Xeon in early 2009, Legacy.macnn.com, retrieved 2008-11-24
  28. ^ Fuad Abazovic (2009-01-28), Nehalem based Xeon comes Mar 29th, Fudzilla.com, retrieved 2009-01-28
  29. ^ Intel Xeon Processor 3500 Series, Intel, 2009-04-01, retrieved 2009-04-10
  30. ^ a b c Teglet, Traian (2008-11-13), Intel to Add New Nehalem Xeon Processor, softpedia, retrieved 2008-11-13 Cite error: The named reference "Intel to Add New Nehalem Xeon Processor" was defined multiple times with different content (see the help page).
  31. ^ "Intel Core i7 Processor Extreme Datasheet, Volume 2, (Document Number: 320835-002)" (PDF). Intel. 2008-11-01. Retrieved 2009-03-24.
  32. ^ a b Worrel, Jon (2009-02-09), Core i7 950 and 975 to replace 940 and 965, Fudzilla, retrieved 2009-02-10
  33. ^ Intel to launch new Core i7 960 and Celerons
  34. ^ Intel Core i5-750, AnandTech, 2009-06-17
  35. ^ Intel Quietly Announces Core i5, Xtreview, 2009-06-26
  36. ^ Intel Xeon Processor 5000 SequenceTechnical Documents
  37. ^ Intel Xeon Processor 5500 Series Datasheet Volume 2 (PDF), Intel, March 2009, retrieved 2009-05-01
  38. ^ a b c Jansen, Ng (2009-02-10), Intel Replaces "Havendale" and "Auburndale" With 32nm Die Shrinks, DailyTech, retrieved 2009-03-01
  39. ^ Intel's Nehalem-EX: A Big Chip For Big Servers
  40. ^ 【後藤弘茂のWeekly海外ニュース】 ノートPCからMPサーバーまで投入されるIntelの「Sandy Bridge」 (in Japanese)
  41. ^ Head to head: Intel's Nehalem EX and AMD's Magny Cours
  42. ^ Intel pushes workhorse Xeons to six cores
  43. ^ Intel Launches Its Most Secure Data Center Processor
  44. ^ Intel Clarkdale Processor, XTREVIEW, 2009-02-09, retrieved 2009-03-01
  45. ^ a b c Ng, Jansen (10 February 2009), "Gulftown" is the Flagship of 32nm "Westmere" Line, DailyTech, retrieved 2009-02-10
  46. ^ a b Gulftown 32nm six-core is only an "Extreme Edition"
  47. ^ Gulftown 32nm six core supports Turbo boost
  48. ^ a b Ng, Jansen (2009-02-10), Intel Aims for Efficiency With New Server Roadmap, DailyTech, retrieved 2009-03-01
  49. ^ a b Bell, Brandon (2009-02-10), Intel CPU Roadmap 2009–2010, FS Media, Inc, retrieved 2009-03-01
  50. ^ CPU list
  51. ^ Arrandale to get to Ultra low voltage
  52. ^ Intel says no to 28nm, focuses on 22nm: Ivy Bridge/Haswell & Larrabee
  53. ^ Smalley, Tim (2007-09-19), Westmere is Nehalem's successor, bit-tech.net
  54. ^ Carry-Less Multiplication and Its Usage for Computing The GCM Mode – Intel Software Network, 2008-04-11, retrieved 2009-03-01
  55. ^ Fuad Abazovic (2008-09-16), Westmere 32nm to improve Nehalem features, Fudzilla.com, retrieved 2009-03-01
  56. ^ Update: Intel's First-Quarter Profits Soar
  57. ^ The Intel Tick-Tock Model of Architecture & Silicon Cadence, Intel Corporation

Further reading

Retrieved from "https://en.wikipedia.org/w/index.php?title=Nehalem_(microarchitecture)&oldid=381125457"