Difference between revisions of "Intel Nehalem (microarchitecture)"

From OnnoWiki
Jump to navigation Jump to search
 
(8 intermediate revisions by the same user not shown)
Line 50: Line 50:
  
 
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. Where the Core microarchitecture used four different processor sockets, one for each market segment, Nehalem now uses [[Socket 1366]] for the high-end of both UP and DP machines, and [[Socket 1156]] for the low end UP machines. The name for the UP version of Gulftown is not yet known; its product code is 80613 and can be found in Intel's product database
 
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. Where the Core microarchitecture used four different processor sockets, one for each market segment, Nehalem now uses [[Socket 1366]] for the high-end of both UP and DP machines, and [[Socket 1156]] for the low end UP machines. The name for the UP version of Gulftown is not yet known; its product code is 80613 and can be found in Intel's product database
 +
 +
{| border="1" cellpadding=2 style="border-collapse: collapse"
 +
|-
 +
! !! Mobile !! Desktop UP Server !! DP Server !! MP Server
 +
|- align="center"
 +
| align="left"|'''Dual-Core 45 nm''' Dual-Channel, PCIe, Graphics Core
 +
| Auburndale ''canceled'' || Havendale ''canceled'' || ||
 +
|- align="center"
 +
| align="left"|'''Dual-Core 32 nm''' Dual-Channel, PCIe, Graphics Core
 +
| [[Arrandale (microprocessor)|Arrandale]] 80617 || [[Clarkdale (microprocessor)|Clarkdale]] 80616 || ||
 +
|- align="center"
 +
| align="left"|'''Quad-Core 45 nm''' Dual-Channel, PCIe
 +
| [[Clarksfield (microprocessor)|Clarksfield]] 80607 || [[Lynnfield (microprocessor)|Lynnfield]] 80605 || [[Xeon#Jasper Forest|Jasper Forest]] 80612 ||
 +
|- align="center"
 +
| align="left"|'''Quad-Core 45 nm''' Triple-Channel
 +
| || [[Bloomfield (microprocessor)|Bloomfield]] 80601 || [[Xeon#5500-series "Gainestown"|Gainestown]] 80602 ||
 +
|- align="center"
 +
| align="left"|'''Six-Core 32 nm''' Triple-Channel
 +
| || [[Gulftown (microprocessor)|Gulftown]] 80613 || [[Gulftown (microprocessor)|Gulftown]] 80614 ||
 +
|- align="center"
 +
| align="left"|'''Eight-Core 45 nm''' Triple-Channel
 +
| || || || [[Xeon#Beckton|Beckton]] 80604
 +
|}
 +
 +
 +
==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:
 +
* "Extreme" processors have an [[CPU locking|unlocked]] [[CPU multiplier|clock multiplier]]. [[Thermal design power]] (TDP) values for CPUs with integrated GPUs include the GPU.
 +
* All variants have 64 [[Kibibyte|KiB]] [[CPU_cache#Multi-level_caches|L1 cache]] per core, and 256 KiB [[CPU_cache#Multi-level_caches|L2 cache]] per core.
 +
 +
 +
===45 nm processors===
 +
{| border="1" cellpadding=2 style="border-collapse: collapse"
 +
|-
 +
! rowspan=2 | Codename
 +
! rowspan=2 | Market
 +
! rowspan=2 | [[Multi-core|Cores]]<br />(Threads)
 +
! rowspan=2 | [[CPU socket|Socket]]
 +
! rowspan=2 | Brand
 +
! rowspan=2 | Processor No.
 +
! colspan=3 | [[Clock rate]]
 +
! rowspan=2 | Turbo
 +
! rowspan=2 | [[Thermal Design Power|TDP]]
 +
! colspan=3 | Interfaces
 +
! rowspan=2 | [[CPU_cache#Multi-level_caches|L3 cache]]
 +
! rowspan=2 | Release
 +
! rowspan=2 | 1k Unit Price
 +
|-
 +
! Base
 +
! Core
 +
! [[Uncore]]
 +
! Chipset
 +
! Memory
 +
! [[PCI Express|PCIe]]
 +
|-
 +
| ''Beckton''
 +
| MP Server
 +
| 8 (16)
 +
| [[Land grid array|LGA]]-1567
 +
|
 +
|
 +
|
 +
|
 +
| rowspan=24|
 +
|
 +
| 130/105/90 [[Watt|W]]
 +
| 4x [[Intel QuickPath Interconnect|QPI]]
 +
| 4x [DDR3 with SMB motherboard]
 +
| n/a
 +
| 24 [[Megabyte|MB]]
 +
| Q1 2010
 +
|
 +
|-
 +
| rowspan=12 | ''Gainestown''
 +
| rowspan=12 | DP Server
 +
| rowspan=8 | 4 (8)
 +
| rowspan=12 | LGA-1366
 +
| rowspan=12 | [[Xeon#Gainestown|Xeon]]
 +
| W5580
 +
| rowspan=12 | 133 [[Hertz|MHz]]
 +
| 3.2 [[Hertz|GHz]]
 +
| rowspan=8 | Yes
 +
| 130 W
 +
| rowspan=4 | 2x QPI 6.4 [[Transfer (computing)|GT]]/s
 +
| rowspan=4 | 3x [[DDR3 SDRAM|DDR3]]-1333<sup>1</sup>
 +
| rowspan=12 | n/a
 +
| rowspan=8 | 8 MB
 +
| rowspan=12 |'''March 29, 2009'''
 +
| $1500
 +
|-
 +
| X5570
 +
| 2.93&nbsp;GHz
 +
| rowspan=3 | 95 W
 +
| $1286
 +
|-
 +
| X5560
 +
| 2.8&nbsp;GHz
 +
| $1072
 +
|-
 +
| X5550
 +
| 2.66&nbsp;GHz
 +
| $858
 +
|-
 +
| E5540
 +
| 2.53&nbsp;GHz
 +
| rowspan=3 | 80 W
 +
| rowspan=4 | 2x 5.86 GT/s
 +
| rowspan=4 | 3x DDR3-1066<sup>1</sup>
 +
| $744
 +
|-
 +
| E5530
 +
| 2.4&nbsp;GHz
 +
| $530
 +
|-
 +
| E5520
 +
| 2.26&nbsp;GHz
 +
| $373
 +
|-
 +
| L5520
 +
| 2.26&nbsp;GHz
 +
| 60 W
 +
| $530
 +
|-
 +
| rowspan=3 | 4 (4)
 +
| E5506
 +
| 2.13&nbsp;GHz
 +
| rowspan=4 | No
 +
| rowspan=2 | 80 W
 +
| rowspan=4 | 2x 4.8 GT/s
 +
| rowspan=4 | 3x DDR3-800<sup>1</sup>
 +
| rowspan=4 | 4 MB
 +
| $266
 +
|-
 +
| E5504
 +
| 2.0&nbsp;GHz
 +
| $224
 +
|-
 +
| L5506
 +
| 2.13&nbsp;GHz
 +
| 60 W
 +
| $423
 +
|-
 +
| 2 (2)
 +
| E5502
 +
| 1.86&nbsp;GHz
 +
| 80 W
 +
| $188
 +
|-
 +
| rowspan=5 | ''Bloomfield''
 +
| rowspan=11 | UP Server
 +
| rowspan=10 | 4 (8)
 +
| rowspan=5 | LGA-1366
 +
| rowspan=11 | Xeon
 +
| W3580
 +
| rowspan=11 | 133&nbsp;MHz
 +
| 3.33&nbsp;GHz
 +
| rowspan=11 | Yes
 +
| rowspan=5 | 130 W
 +
| rowspan=2 | 1x QPI 6.4 GT/s
 +
| rowspan=2 | 3x DDR3-1333
 +
| rowspan=5 | n/a
 +
| rowspan=11 | 8 MB
 +
| '''August 9, 2009'''
 +
| $999
 +
|-
 +
| W3570
 +
| 3.2&nbsp;GHz
 +
| '''March 29, 2009'''
 +
| $999
 +
|-
 +
| W3550
 +
| 3.06&nbsp;GHz
 +
| rowspan=3 | 1x QPI 4.8 GT/s
 +
| rowspan=3 | 3x DDR3-1066
 +
| '''August 9, 2009'''
 +
| $562
 +
|-
 +
| W3540
 +
| 2.93&nbsp;GHz
 +
|rowspan=2 | '''March 29, 2009'''
 +
| $562
 +
|-
 +
| W3520
 +
| 2.66&nbsp;GHz
 +
| $284
 +
|-
 +
| rowspan=6|''Lynnfield''
 +
| rowspan=6|LGA 1156
 +
| X3470
 +
| 2.93&nbsp;GHz
 +
| rowspan=4|95 W
 +
| rowspan=6|DMI
 +
| rowspan=6|2x DDR3-1333
 +
| rowspan=6|
 +
| rowspan=6|'''September 8, 2009'''
 +
| $589
 +
|-
 +
| X3460
 +
| 2.8&nbsp;GHz
 +
| $316
 +
|-
 +
| X3450
 +
| 2.66&nbsp;GHz
 +
| $241
 +
|-
 +
| X3440
 +
| 2.53&nbsp;GHz
 +
| $215
 +
|-
 +
| L3426
 +
| 1.86&nbsp;GHz
 +
| 45 W
 +
| $284
 +
|-
 +
| 4 (4)
 +
| X3430
 +
| 2.4&nbsp;GHz
 +
| 95 W
 +
| $189
 +
|-
 +
| rowspan=7 | ''Bloomfield''
 +
| rowspan=7 | Extreme/Performance Desktop
 +
| rowspan=7 | 4 (8)
 +
| rowspan=7 | LGA-1366
 +
| rowspan=2 | [[Intel Core i7|Core i7]] Extreme
 +
| 975
 +
| rowspan=7 | 133&nbsp;MHz
 +
| 3.33&nbsp;GHz
 +
| rowspan=2 | 2.66&nbsp;GHz
 +
| rowspan=7 | Yes
 +
| rowspan=7 | 130 W
 +
| rowspan=2 | 1x QPI 6.4 GT/s
 +
| rowspan=7 | 3x DDR3-1066
 +
| rowspan=7 | n/a
 +
| rowspan=7 | 8 MB
 +
| '''May 31, 2009'''
 +
| $999
 +
|-
 +
| 965
 +
| 3.2&nbsp;GHz
 +
| '''November 17, 2008'''
 +
| $999
 +
|-
 +
| rowspan=8 | Core i7
 +
| 960
 +
| 3.2&nbsp;GHz
 +
| rowspan=5 | 2.13&nbsp;GHz
 +
| rowspan=5 | 1x QPI 4.8 GT/s
 +
| '''October 20, 2009'''
 +
| $562
 +
|-
 +
| 950
 +
| 3.06&nbsp;GHz
 +
| '''May 31, 2009'''
 +
| $562
 +
|-
 +
| 940
 +
| 2.93&nbsp;GHz
 +
| '''November 17, 2008'''
 +
| $562
 +
|-
 +
| 930
 +
| 2.8&nbsp;GHz
 +
| February 28, 2010
 +
| $284
 +
|-
 +
| 920
 +
| 2.66&nbsp;GHz
 +
| '''November 17, 2008'''
 +
| $284
 +
|-
 +
| rowspan=5 | ''Lynnfield''
 +
| rowspan=5 | Performance/Mainstream Desktop
 +
| rowspan=3 | 4 (8)
 +
| rowspan=5 | [[LGA 1156]]
 +
| 870
 +
| rowspan=8 | 133&nbsp;MHz
 +
| 2.93&nbsp;GHz
 +
| rowspan=2 | 2.4&nbsp;GHz
 +
| rowspan=8 | Yes
 +
| rowspan=2 | 95 W
 +
| rowspan=8 | [[Direct Media Interface|DMI]]
 +
| rowspan=8 | 2x DDR3-1333
 +
| rowspan=8 | 1x16 / 2x8
 +
| rowspan=7 | 8 MB
 +
| rowspan=2 | '''September 8, 2009'''
 +
| $562
 +
|-
 +
| 860
 +
| 2.8&nbsp;GHz
 +
| $284
 +
|-
 +
| 860S
 +
| 2.53&nbsp;GHz
 +
|
 +
| 82 W
 +
| '''January 7, 2010'''
 +
| $337
 +
|-
 +
| rowspan=2 | 4 (4)
 +
| rowspan=2 | Core i5
 +
| 750
 +
| 2.66&nbsp;GHz
 +
| 2.13&nbsp;GHz
 +
| 95 W
 +
| '''September 8, 2009'''
 +
| $196
 +
|-
 +
| 750S
 +
| 2.4&nbsp;GHz
 +
|
 +
| 82 W
 +
| '''January 7, 2010'''
 +
| $259
 +
|-
 +
| rowspan=3 | ''Clarksfield''
 +
| rowspan=3 | Extreme/Performance Mobile
 +
| rowspan=3 | 4 (8)
 +
| rowspan=3 | mPGA-989
 +
| Core i7 Extreme
 +
| 920XM
 +
| 2.0&nbsp;GHz
 +
| rowspan=3|
 +
| 55 W
 +
| rowspan=3 | '''September 23, 2009'''
 +
| $1054
 +
|-
 +
| rowspan=2 | Core i7
 +
| 820QM
 +
| 1.73&nbsp;GHz
 +
| rowspan=2 | 45 W
 +
| $546
 +
|-
 +
| 720QM
 +
| 1.6&nbsp;GHz
 +
| 6 MB
 +
| $364
 +
|-
 +
|}
 +
 +
*<sup>1</sup> 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; in effect, a multiplexed six-channel system.<!-- I have no clue if this ref is in correct format; please fix if not. -->
 +
 +
The ''Havendale'' and ''Auburndale'' variants (which contained ''Gilo'' and ''Ironlake'') have been cancelled.
 +
 +
===32 nm processors===
 +
{| border="1" cellpadding=2 style="border-collapse: collapse"
 +
|-
 +
! rowspan=2 | Codename
 +
! rowspan=2 | Market
 +
! rowspan=2 | [[Multi-core|Cores]]<br />(Threads)
 +
! rowspan=2 | [[CPU socket|Socket]]
 +
! rowspan=2 | Brand
 +
! rowspan=2 | Processor No.
 +
! colspan=3 | [[Clock rate]]
 +
! rowspan=2 | Turbo
 +
! rowspan=2 | TDP
 +
! colspan=3 | Interfaces
 +
! rowspan=2 | [[CPU_cache#Multi-level_caches|L3 cache]]
 +
! rowspan=2 | Release
 +
! rowspan=2 | 1k Unit Price
 +
|-
 +
! Base
 +
! Core
 +
! GPU
 +
! Chipset
 +
! Memory
 +
! [[PCI Express|PCIe]]
 +
|-
 +
| ''Gulftown''
 +
| Extreme/Performance Desktop
 +
| 6 (12)
 +
| LGA-1366
 +
| Core i7 Extreme
 +
| 980 XE
 +
| 133 MHz
 +
| 3.33 GHz
 +
| n/a
 +
| Yes
 +
| 130 W
 +
| 2x [[Intel QuickPath Interconnect|QPI]]
 +
| 3x [[DDR3 SDRAM|DDR3]]-1066
 +
| n/a
 +
| 12 MB
 +
| March 16, 2010
 +
| $999
 +
|-
 +
| rowspan=7 | ''Clarkdale''
 +
| rowspan=7 | Mainstream/Value Desktop
 +
| rowspan=6 | 2 (4)
 +
| rowspan=7 | LGA-1156
 +
| rowspan=4 | Core i5
 +
| 670
 +
| rowspan=7 | 133&nbsp;MHz
 +
| 3.46&nbsp;GHz
 +
| 733&nbsp;MHz
 +
| rowspan=4 | Yes
 +
| 73 W
 +
| rowspan=7 | [[Direct Media Interface|DMI]]
 +
| rowspan=6 | 2x DDR3-1333
 +
| rowspan=7 | 1 x16
 +
| rowspan=6 | 4 MB
 +
| rowspan=7 | '''January 7, 2010'''
 +
| $284
 +
|-
 +
| 661
 +
| rowspan=2 | 3.33&nbsp;GHz
 +
| 900&nbsp;MHz
 +
| 87 W
 +
| rowspan=2 | $196
 +
|-
 +
| 660
 +
| rowspan=4 | 733&nbsp;MHz
 +
| rowspan=5 | 73 W
 +
|-
 +
| 650
 +
| 3.2&nbsp;GHz
 +
| $176
 +
|-
 +
| rowspan=2 | Core i3
 +
| 540
 +
| 3.06&nbsp;GHz
 +
| rowspan=3 | No
 +
| $133
 +
|-
 +
| 530
 +
| 2.93&nbsp;GHz
 +
| $113
 +
|-
 +
| 2 (2)
 +
| Pentium
 +
| G6950
 +
| 2.8&nbsp;GHz
 +
| 533&nbsp;MHz
 +
| 2x DDR3-1066
 +
| 3 MB
 +
| $87
 +
|-
 +
| rowspan=12 | ''Arrandale''
 +
| rowspan=12 | Mainstream/Value Mobile
 +
| rowspan=11 | 2 (4)
 +
| rowspan=12 | mPGA-989
 +
| rowspan=5 | Core i7
 +
| 620M
 +
| rowspan=12 | 133&nbsp;MHz
 +
| 2.66&nbsp;GHz
 +
| 766&nbsp;MHz
 +
| rowspan=9 | Yes
 +
| 35 W
 +
| rowspan=12 | DMI
 +
| rowspan=3 | 2x DDR3-1066
 +
| rowspan=12 | 1 x16
 +
| rowspan=5 | 4 MB
 +
| rowspan=11 | '''January 7, 2010'''
 +
| $332
 +
|-
 +
| 640LM
 +
| 2.13&nbsp;GHz
 +
| rowspan=2 | 566&nbsp;MHz
 +
| rowspan=2 | 25 W
 +
| $332
 +
|-
 +
| 620LM
 +
| 2.0&nbsp;GHz
 +
| $300
 +
|-
 +
| 640UM
 +
| 1.2&nbsp;GHz
 +
| rowspan=2 | 500&nbsp;MHz
 +
| rowspan=2 | 18 W
 +
| rowspan=2 | 2x DDR3-800
 +
| $305
 +
|-
 +
| 620UM
 +
| 1.06&nbsp;GHz
 +
| $278
 +
|-
 +
| rowspan=4 | Core i5
 +
| 540M
 +
| 2.53&nbsp;GHz
 +
| rowspan=2 | 766&nbsp;MHz
 +
| rowspan=2 | 35 W
 +
| rowspan=2 | 2x DDR3-1066
 +
| rowspan=6 | 3 MB
 +
| $257
 +
|-
 +
| 520M
 +
| 2.4&nbsp;GHz
 +
| $225
 +
|-
 +
| 520UM
 +
| 1.06&nbsp;GHz
 +
| 500&nbsp;MHz
 +
| 18 W
 +
| 2x DDR3-800
 +
| $241
 +
|-
 +
| 430M
 +
| 2.26&nbsp;GHz
 +
| 766&nbsp;MHz
 +
| rowspan=4 | 35 W
 +
| rowspan=3 | 2x DDR3-1066
 +
| rowspan=3 | OEM
 +
|-
 +
| rowspan=2 | Core i3
 +
| 350M
 +
| 2.26&nbsp;GHz
 +
| rowspan=2 | 667&nbsp;MHz
 +
| rowspan=3 | No
 +
|-
 +
| 330M
 +
| 2.13&nbsp;GHz
 +
|-
 +
| 2 (2)
 +
| Celeron
 +
| P4500
 +
| 1.86&nbsp;Ghz
 +
| 500 Mhz
 +
|
 +
| 2 MB
 +
| Q2, 2010
 +
|
 +
|}
 +
 +
 +
For the desktop, ''Gulftown'' is to be an "Extreme Edition" CPU and so will coexist with ''Bloomfield.'' It will have Turbo Boost and similar clock speeds to Bloomfield.
 +
 +
''Lynnfield'' and ''Clarksfield'' may make the 32&nbsp;nm transition in the middle of 2010, sometime after Q2, while ''Beckton'' will move to 32&nbsp;nm at the end of 2010.
 +
The 32&nbsp;nm CPUs will not have significantly different clock speeds compared to 45&nbsp;nm CPUs.
 +
''Clarkdale'' and ''Arrandale'' contain the 32&nbsp;nm dual core processor ''Hillel'' and the 45&nbsp;nm integrated graphics device ''Ironlake'', and support switchable graphics. The lowest-power variant of Arrandale may have a 10 W CPU TDP, and a maximum clock speed of 1.6&nbsp;GHz.
 +
A successor to ''Bloomfield'' and entry level server chips are also expected in Q2 2010.
 +
 +
 +
====Westmere====
 +
'''Westmere''' (formerly '''Nehalem-C''') is the name given to the [[32 nanometer|32 nm]] [[die shrink]] of ''Nehalem''. ''Westmere'' was to be ready for a Q4 2009 release provided that Intel stayed on target with their roadmap. However, the first Westmere-based processors were launched on January 7, 2010 as the Core i3, Core i5, and dual-core mobile Core i7.
 +
 +
Westmere's features and improvements from Nehalem have been reported as follows:
 +
 +
* Native six-core, and possibly dual-die hex-core (12-cores), processors.
 +
** 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.
 +
** 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. 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.
 +
* New virtualization capability: "VMX Unrestricted mode support," which allows 16-bit guests to run (real mode and big real mode).
 +
 +
 +
 +
==Successor==
 +
 +
The successor to ''Nehalem'' and ''Westmere'' will be ''[[Intel Sandy Bridge (microarchitecture)|Sandy Bridge]]'', scheduled for release in 2011, according to Intel roadmaps. The successor to ''Sandy Bridge'' will be ''[[Intel Haswell (microarchitecture)|Haswell]]'', scheduled for release in 2012. It will come with a new cache subsystem, a FMA ([[Multiply-accumulate|fused multiply-add]]) unit, and a vector coprocessor.
 +
 +
[[Image:IntelProcessorRoadmap.jpg|500px|center|thumb|Intel CPU core roadmaps from ''[[Intel NetBurst|NetBurst]]'' to ''[[Intel Sandy Bridge (microarchitecture)|Sandy Bridge]]'']]
 +
 +
 +
==See Also==
 +
* [[Intel Core i7|Core i7]]
 +
* [[X86]] instruction set architecture
 +
* [[x86-64]]
 +
* [[Xeon]]
 +
* [[Pentium]]
 +
* [[Intel P6 (microarchitecture)|P6]]
 +
* [[Intel NetBurst (microarchitecture)|NetBurst]]
 +
* [[Intel Core (microarchitecture)|Core]]
 +
* [[Intel Sandy Bridge (microarchitecture)|Sandy Bridge]]
 +
* [[Intel Haswell (microarchitecture)|Haswell]]
 +
* [[List of Intel CPU microarchitectures]]
 +
 +
==External links==
 +
 +
*[http://www.intel.com/technology/architecture-silicon/next-gen/ Nehalem processor] at [[Intel]].com

Latest revision as of 08:17, 10 February 2010

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, 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. 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. Nehalem-based Xeon EX processors for larger servers were expected in Q4 2009 based on initial announcements from Intel, but in November 2009 the launch of these processors was pushed back to the first half of 2010.

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 will have a focus on performance, which accounts for the increased core size. Compared to Penryn, Nehalem will have:

  • 1.1x to 1.25x the single-threaded performance or 1.2x to 2x 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."
  • Per Core, clock-for-clock, Nehalem will provide a 15–20% increase in performance compared to Penryn.

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

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

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.

Overclocking will be possible with Bloomfield processors and the X58 chipset. The Lynnfield processor will use a PCH removing the need for a northbridge chipset.

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.

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. Where the Core microarchitecture used four different processor sockets, one for each market segment, Nehalem now uses Socket 1366 for the high-end of both UP and DP machines, and Socket 1156 for the low end UP machines. The name for the UP version of Gulftown is not yet known; its product code is 80613 and can be found in Intel's product database

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


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 8 (16) LGA-1567 130/105/90 W 4x QPI 4x [DDR3 with SMB motherboard] n/a 24 MB Q1 2010
Gainestown DP Server 4 (8) LGA-1366 Xeon W5580 133 MHz 3.2 GHz Yes 130 W 2x QPI 6.4 GT/s 3x DDR3-13331 n/a 8 MB March 29, 2009 $1500
X5570 2.93 GHz 95 W $1286
X5560 2.8 GHz $1072
X5550 2.66 GHz $858
E5540 2.53 GHz 80 W 2x 5.86 GT/s 3x 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 2x 4.8 GT/s 3x 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 4 (8) LGA-1366 Xeon W3580 133 MHz 3.33 GHz Yes 130 W 1x QPI 6.4 GT/s 3x DDR3-1333 n/a 8 MB August 9, 2009 $999
W3570 3.2 GHz March 29, 2009 $999
W3550 3.06 GHz 1x QPI 4.8 GT/s 3x DDR3-1066 August 9, 2009 $562
W3540 2.93 GHz March 29, 2009 $562
W3520 2.66 GHz $284
Lynnfield LGA 1156 X3470 2.93 GHz 95 W DMI 2x 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 Extreme/Performance Desktop 4 (8) LGA-1366 Core i7 Extreme 975 133 MHz 3.33 GHz 2.66 GHz Yes 130 W 1x QPI 6.4 GT/s 3x DDR3-1066 n/a 8 MB May 31, 2009 $999
965 3.2 GHz November 17, 2008 $999
Core i7 960 3.2 GHz 2.13 GHz 1x QPI 4.8 GT/s October 20, 2009 $562
950 3.06 GHz May 31, 2009 $562
940 2.93 GHz November 17, 2008 $562
930 2.8 GHz February 28, 2010 $284
920 2.66 GHz November 17, 2008 $284
Lynnfield Performance/Mainstream Desktop 4 (8) LGA 1156 870 133 MHz 2.93 GHz 2.4 GHz Yes 95 W DMI 2x DDR3-1333 1x16 / 2x8 8 MB September 8, 2009 $562
860 2.8 GHz $284
860S 2.53 GHz 82 W January 7, 2010 $337
4 (4) Core i5 750 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 920XM 2.0 GHz 55 W September 23, 2009 $1054
Core i7 820QM 1.73 GHz 45 W $546
720QM 1.6 GHz 6 MB $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; in effect, a multiplexed six-channel system.

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

32 nm processors

Codename Market Cores
(Threads)
Socket Brand Processor No. Clock rate Turbo TDP Interfaces L3 cache Release 1k Unit Price
Base Core GPU Chipset Memory PCIe
Gulftown Extreme/Performance Desktop 6 (12) LGA-1366 Core i7 Extreme 980 XE 133 MHz 3.33 GHz n/a Yes 130 W 2x QPI 3x DDR3-1066 n/a 12 MB March 16, 2010 $999
Clarkdale Mainstream/Value Desktop 2 (4) LGA-1156 Core i5 670 133 MHz 3.46 GHz 733 MHz Yes 73 W DMI 2x DDR3-1333 1 x16 4 MB 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 2x DDR3-1066 3 MB $87
Arrandale Mainstream/Value Mobile 2 (4) mPGA-989 Core i7 620M 133 MHz 2.66 GHz 766 MHz Yes 35 W DMI 2x DDR3-1066 1 x16 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 2x DDR3-800 $305
620UM 1.06 GHz $278
Core i5 540M 2.53 GHz 766 MHz 35 W 2x DDR3-1066 3 MB $257
520M 2.4 GHz $225
520UM 1.06 GHz 500 MHz 18 W 2x DDR3-800 $241
430M 2.26 GHz 766 MHz 35 W 2x DDR3-1066 OEM
Core i3 350M 2.26 GHz 667 MHz No
330M 2.13 GHz
2 (2) Celeron P4500 1.86 Ghz 500 Mhz 2 MB Q2, 2010


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

Lynnfield and Clarksfield may make the 32 nm transition in the middle of 2010, sometime after Q2, while Beckton will move to 32 nm at the end of 2010. The 32 nm CPUs will not have significantly different clock speeds compared to 45 nm CPUs. Clarkdale and Arrandale contain the 32 nm dual core processor Hillel and the 45 nm integrated graphics device Ironlake, and support switchable graphics. The lowest-power variant of Arrandale may have a 10 W CPU TDP, and a maximum clock speed of 1.6 GHz. A successor to Bloomfield and entry level server chips are also expected in Q2 2010.


Westmere

Westmere (formerly Nehalem-C) is the name given to the 32 nm die shrink of Nehalem. Westmere was to be ready for a Q4 2009 release provided that Intel stayed on target with their roadmap. However, the first Westmere-based processors were launched on January 7, 2010 as the Core i3, Core i5, and dual-core mobile Core i7.

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

  • Native six-core, and possibly dual-die hex-core (12-cores), processors.
    • 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.
    • 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. 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.
  • New virtualization capability: "VMX Unrestricted mode support," which allows 16-bit guests to run (real mode and big real mode).


Successor

The successor to Nehalem and Westmere will be Sandy Bridge, scheduled for release in 2011, according to Intel roadmaps. 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.

Intel CPU core roadmaps from NetBurst to Sandy Bridge


See Also

External links