Xeon

This article is about the Intel microprocessor. For the emulator, see Xeon (emulator).

Xeon

File:Xeon.png
General information
Launched	1998
Discontinued	present
Common manufacturer	Intel
Performance
Max. CPU clock rate	400 MHz to 3.6 GHz
Architecture and classification
Microarchitecture	Intel Core Microarchitecture, NetBurst, P6
Instruction set	x86
Physical specifications
Cores	1, 2, or 4

The Xeon brand refers to many families of Intel's x86 multiprocessing CPUs – for dual-processor (DP) and multi-processor (MP) configuration on a single motherboard targeted at non-consumer markets of server and workstation computers, and also at blade servers and embedded systems. The Xeon brand has been maintained over several generations of x86 and x86-64 processors. Older models added the Xeon moniker to the end of the name of their corresponding desktop processor, but more recent models used the name Xeon on its own. The Xeon CPUs generally have more cache than their desktop counterparts in addition to multiprocessing capabilities. Intel's (non-x86) IA-64 processors are called Itanium, not Xeon.

Pentium II Xeon

The first Xeon branded processor was released in 1998, named the Pentium II Xeon (codenamed "Drake"), as the replacement of the Pentium Pro. It was based on the 0.25 µm "Deschutes" core (P6 microarchitecture) branded Pentium II (sharing its 80523 product code), used either a 440GX (a dual-processor workstation chipset) or 450NX (quad-processor, or oct with additional logic) chipset, and differed from the Pentium II desktop CPU (Deschutes) in that its off-die L2 cache ran at full speed. It also used a larger slot known as slot 2. Cache sizes were 512 KB (512 KiB), 1 MB (1 MiB) and 2 MB, and it used a 100 MT/s front side bus (FSB).

Pentium III Xeon

File:Pentium 3 Xeon Logo.png

Pentium III Xeon logo

In 1999, the Pentium II Xeon was replaced by the Pentium III Xeon. Reflecting the incremental changes from the Pentium II "Deschutes" core to the Pentium III "Katmai" core, the first Pentium III Xeon, named "Tanner", was just like its predecessor except for the addition of Streaming SIMD Extensions (SSE) and a few cache controller improvements. The second version, named "Cascades", was based on the Pentium III "Coppermine" core. The "Cascades" Xeon used a 133 MT/s bus and relatively small 256 KB on-die L2 cache resulting in almost the same capabilities as the Slot 1 Coppermine processors, which were capable of dual-processor operation but not quad-processor operation. To improve this situation, Intel released another version, officially also named "Cascades", but often referred to as "Cascades 2 MB". That came in two variants: with 1 MB or 2 MB of L2 cache. Its bus speed was fixed at 100 MT/s, though in practice the cache was able to offset this. Product codes for Tanner and Cascades mirrored that of Katmai and Coppermine; 80525 and 80526 respectively.

Xeon (DP) & Xeon MP (32-bit)

In mid-2001, the Xeon brand was introduced ("Pentium" was dropped from the name). The initial variant that used the new NetBurst architecture, "Foster", was slightly different from the desktop Pentium 4 ("Willamette"). It was a decent chip for workstations, but for server applications it was almost always outperformed by the older Cascade 2 MB core and AMD's Athlon MP. Combined with the need to use expensive Rambus Dynamic RAM, the Foster's sales were somewhat unimpressive.

At most two Foster processors could be accommodated in an SMP system built with a mainstream chipset, so a second version (Foster MP) was introduced with a 1 MB L3 cache and the Jackson Hyper-Threading capacity. This improved performance slightly, but not enough to lift it out of third place. It was also priced much higher than the dual-processor (DP) versions.

In 2002 Intel released a 130 nm version of Xeon branded CPU, codenamed "Prestonia". It supported Intel's new Hyper-Threading technology and had a 512 KB L2 cache. This was based on the "Northwood" Pentium 4 core. A new server chipset, E7500 (which allowed the use of dual-channel DDR SDRAM) was released to support this processor in servers, and soon the bus speed was boosted to 533 MT/s (accompanied by new chipsets: the E7501 for servers and the E7505 for workstations). The Prestonia performed much better than its predecessor and noticeably better than Athlon MP. The support of new features in the E75xx series also gave it a key advantage over the Pentium III Xeon and Athlon MP branded CPUs (both stuck with rather old chipsets), and it quickly became the top-selling server/workstation processor.

Subsequent to the Prestonia was the "Gallatin", which had an L3 cache of 1 MB or 2 MB. Its Xeon MP version also performed much better than the Foster MP, and was popular in servers. Later experience with the 130 nm process allowed Intel to create the Xeon MP branded Gallatin with 4 MB cache.

The Foster shared the 80528 product code with Willamette; The Xeon branded Prestonia and Gallatin were designated 80532, like Northwood.

Xeon (DP) & Xeon MP (64-bit)

Due to a lack of success with Intel's Itanium and Itanium 2 processors, the 90 nm version of the Pentium 4 ("Prescott") was built with support for 64-bit instructions (called Intel 64 or EM64T, Intel's implementation of x86-64), and a Xeon version codenamed "Nocona" was released in 2004. Released with it were the E7525 (workstation), E7520 and E7320 (both server) chipsets, which added support for PCI Express, DDR-II and Serial ATA. The Xeon was noticeably slower than AMD's Opteron, although it could be faster in situations where Hyper-Threading came into play.

A slightly updated core called "Irwindale" was released in early 2005, with twice the L2 cache of Nocona and able to reduce its clockspeeds during low processor demand. However, independent tests showed that AMD's Opteron still outperformed Irwindale.

64-bit Xeon MPs were introduced in April 2005. The cheaper "Cranford" was an MP version of Nocona, while the more expensive "Potomac" was a Cranford with 8 MB of L3 cache. All these Prescott-derived Xeons have the product code 80546.

Dual-Core Xeon

Xeon processor family
	New Xeon logo
Original logo	2008 New logo

"Paxville DP"

The first dual-core CPU branded Xeon, codenamed Paxville DP, product code 80551, was released by Intel on 10 October 2005. Paxville DP had NetBurst architecture, and was a dual-core equivalent of the single-core Irwindale (related to the Pentium D branded "Smithfield"") with 4 MB of L2 Cache (2 MB per core). The only one Paxville DP model released ran at 2.8 GHz, featured an 800 MT/s front side bus, and was produced using a 90 nm process.

7000-series "Paxville MP"

An MP-capable version of Paxville DP, codenamed Paxville MP, product code 80560, was released on 1 November 2005. There are two versions: one with 2 MB of L2 Cache (1 MB per core), and one with 4 MB of L2 (2 MB per core). Paxville MP, called the dual-core Xeon 7000-series, was produced using a 90 nm process. Paxville MP clock ranges between 2.67 and 3.0 GHz (model numbers 7020-7041), with some models having a 667 MT/s FSB, and others having an 800 MT/s FSB.

LV (ULV), "Sossaman"

On 14 March 2006, Intel released a dual-core processor codenamed Sossaman and branded as Xeon LV (low-voltage). Subsequently a ULV (ultra-low-voltage) version was released. The Sossaman was a low-/ultra-low-power and double-processor capable CPU (like AMD Quad FX), based on the "Yonah" processor, for ultradense non-consumer environment (i.e. targeted at the blade-server and embedded markets), and it was rated at a thermal design power (TDP) of 31 watts (LV: 1.66 and 2 GHz ) and 15 W (ULV: 1.66 GHz). As such, it supported most of the same features as earlier Xeons: Virtualization Technology, 667 MT/s front side bus, and dual-core processing, but it did not support 64-bit operations, so it could not run 64-bit-only server software, such as Microsoft Exchange Server 2007, and therefore it was limited to only 16 GB of memory. A planned successor, codenamed "Merom MP" was to be a drop-in upgrade to allow Sossaman-based servers to upgrade to 64-bit capability. However, this was abandoned in favour of low-voltage versions of the Woodcrest LV processor leaving the Sossaman at a dead-end with no planned upgrades.

5000-series "Dempsey"

On 23 May 2006, Intel released the dual-core CPU (Xeon branded 5000 series) codenamed Dempsey (product code 80555). Released as the Dual-Core Xeon 5000-series, Dempsey is a NetBurst architecture processor produced using a 65 nm process, and is virtually identical to Intel's "Presler" Pentium Extreme Edition, except for the addition of SMP support, which lets Dempsey operate in dual-processor systems. Dempsey ranges between 2.67 and 3.73 GHz (model numbers 5030-5080). Some models have a 667 MT/s FSB, and others have a 1066 MT/s FSB. Dempsey has 4 MB of L2 Cache (2 MB per core). A Medium Voltage model, at 3.2 GHz and 1066 MT/s FSB (model number 5063), has also been released. Dempsey also introduces a new interface for Xeon processors: Socket J, also known as LGA 771.

5100-series "Woodcrest"

On 26 June 2006, Intel released the dual-core CPU (Xeon branded 5100 series) codenamed Woodcrest (product code 80556); it was the first Intel Core microarchitecture processor to be launched on the market. It is a server and workstation version of the Intel Core 2 processor. Intel claims that it provides an 80% boost in performance, while reducing power consumption by 20% relative to the Pentium D.

Most models have a 1333 MT/s FSB, except for the 5110 and 5120, which have a 1066 MT/s FSB. The fastest processor (5160) operates at 3.0 GHz. All Woodcrests use LGA 771 and all except two models have a TDP of 65 watts. The 5160 has a TDP of 80 W and the 5148LV (2.33 GHz), has a TDP of 40 W. The previous generation Xeons had a TDP of 130 W. All models support Intel 64 (Intel's x86-64 implementation), the XD bit, and Virtualization Technology, with the "Demand Based Switching" power management option only on Dual-Core Xeon 5140 or above. Woodcrest has 4 MB of shared L2 Cache.

7100-series "Tulsa"

Released on 29 August 2006, the 7100 series, codenamed Tulsa (product code 80550), is an improved version of Paxville MP, built on a 65 nm process, with 2 MiB of L2 cache (1 MiB per core) and up to 16 MiB of L3 cache. It uses Socket 604 [1]. Tulsa was released in two lines: the N-line uses a 667 MT/s FSB, and the M-line uses an 800 MT/s FSB. The N-line ranges from 2.5 to 3.5 GHz (model numbers 7110N-7150N), and the M-line ranges from 2.6 to 3.4 GHz (model numbers 7110M-7140M). L3 cache ranges from 4 MiB to 16 MiB across the models.

7200-series "Tigerton"

The 7200 series, codenamed Tigerton is an MP-capable processor, similar to the 7300 series, but, in contrast, only one core is active on each silicon chip, and the other one is turned off (blocked), resulting as a dual-core capable processor. [2] [3][4] [5]

Tigerton model	Speed (GHz)	L2 Cache (MB)	FSB (MHz)	TDP (Watts)
E7220	2.93	8	1066	80
E7210	2.40	8	1066	80

3000-series "Conroe"

The 3000 series, codenamed Conroe (product code 80557) dual-core Xeon (branded) CPU, released at the end of September 2006, was just rebranded version of the Intel's mainstream Conroe otherwise branded as Core 2 Duo (for consumer desktops). Unlike most Xeon processors, they only support single-CPU operation. They use Socket T (LGA775), operate on a 1066 MHz front-side bus, and do not support Hyper-Threading. Processors with a number ending in "5" have a 1333 MT/s FSB.

Xeon model	Core 2 Duo model	Speed (GHz)	L2 Cache (MB)	FSB (MHz)	TDP (Watts)
3040	E6300	1.86	2	1066	65
N/A	E6320	1.86	4	1066
3050	E6400	2.13	2	1066	65
3055	E6420	2.13	4	1066
3060	E6600	2.4	4	1066	65
3070	E6700	2.66	4	1066	65
3080	X6800	2.93	4	1066	65
3065	E6550	2.33	4	1333	65
3075	E6750	2.66	4	1333	65
3085	E6850	3.00	4	1333	65

Quad-Core Xeon

3200-series "Kentsfield"

Intel released relabeled versions of its quad-core (2x2) Core 2 Quad processor as the Xeon 3200-series on 7 January 2007. The 2x2 "quad-core" (dual-die dual-core) comprised two separate dual-core die next to each other in one CPU package. The models are the X3210, X3220 and X3230, running at 2.13, 2.4 and 2.66 GHz, respectively. Like the 3000-series, these models only support single-CPU operation and operate on a 1066 MHz front-side bus. It is targeted at the "blade" market.

5300-series "Clovertown"

A quad-core (2x2) successor of the Woodcrest for DP segment, consisting of two dual-core Woodcrest chips in one package similarly to the dual-core Pentium D branded CPUs (two single-core chips) or the quad-core Kentsfield. The Clovertown has been usually implemented with two Woodcrest dies on a multi-chip module, with 8 MB of L2 cache (4 MB per die). Like Woodcrest, lower models use a 1066 MT/s FSB, and higher models use a 1333 MT/s FSB. Intel released Clovertown, product code 80563, on 14 November 2006 with models E5310, E5320, E5335, E5345, and X5355, ranging from 1.6 to 2.66 GHz. The E and X designations are borrowed from Intel's Core 2 model numbering scheme; an ending of -0 implies a 1066 MT/s FSB, and an ending of -5 implies a 1333 MT/s FSB. All models have a TDP of 80 W with the exception of the X5355, which has a TDP of 120 W. A low-voltage version of Clovertown with a TDP of 50 W has a model numbers L5310, L5320 and L5335 (1.6, 1.86 and 2.0 GHz respectively). The 3.0 GHz X5365 arrived in July 2007, and became available in the Apple Mac Pro [6] on 4 April 2007.[7] The X5365 is among the fastest processors, performing up to around 43 GFLOPS, while others say that they can perform up to 60 GFLOPS, possibly with overclocking. [8]

5400-series "Harpertown"

Intel’s Sep 2007 roadmap revealed three new Penryn based Xeon models with a higher front-side bus speed – 1600 MHz. The three new 1600 MHz front-side bus processors were available in dual-core and quad-core models. The quad-core Xeon E5472 and E5462 were the first quad-core models to receive the 1600 MHz front-side bus.

The Xeon E5472 features a 3.0 GHz clock speed while the E5462 features a 2.8 GHz clock speed. These models feature 80-watt thermal ratings as designated with the E moniker. Intel plans to release these 1600 MHz front-side bus processors in Q4 2007 with the rest of the Penryn family.

Intel has one 1600 MHz dual-core Penryn based Xeon processor ready for launch – the E5272. The Xeon E5272 features a 3.4 GHz clock speed and is priced at $1,172. Intel has also pulled in the launch of all dual-core Xeon processors to Q4 2007, with the quad-core processors. The dual-core Xeon E5260 and E5205 will also launch in Q4, not the Q1 2008 date earlier roadmaps showed.

Intel 1600 MHz front-side bus Xeon processors will drop into the upcoming Seaburg chipset. Seaburg features support for dual PCIe 2.0 x16 slots and up to 128 GB of memory.

The 5xxx series is designed for single or dual CPU configurations, in workstations, such as the early 2008 Mac Pro, or small enterprise servers.

7300-series "Tigerton"

The 7300 series, codenamed Tigerton is a four-socket (packaged in Socket 604) and greater capable quad-core processor, consisting of two dual core Core2 architecture silicon chips on a single ceramic module, similar to Intel's Xeon 5300 series Clovertown processor modules. It was announced on 5 September 2007 [9], and is expected to ship in the second half of 2007.

The 7300 series uses Intel's Caneland (Clarksboro) platform.

Intel claims the 7300 series Xeons offer more than twice the performance and more than three times the performance per watt as Intel's previous generation 7100 series. The 7300 series' Caneland chipset provides a point to point interface allowing the full front side bus bandwidth per processor.

The 7xxx series is aimed at the large server market, supporting configurations of up to 32 CPUs per host.

Tigerton model	Speed (GHz)	L2 Cache (MB)	FSB (MHz)	TDP (Watts)
X7350	2.93	8	1066	130
E7340	2.40	8	1066	80
E7330	2.40	6	1066	80
E7320	2.13	4	1066	80
E7310	1.60	4	1066	80
L7345	1.86	8	1066	50

Future versions

Whitefield (cancelled)

A quad-core processor, partially based on Woodcrest, and would have used the new QuickPath Interconnect bus, a bus shared with the Itanium 2 processors of its generation (beginning with the "Tukwila" core). Whitefield would have had 16 MB of L2 cache, manufactured using the 65 nm process initially, and the 45 nm process later.

Whitefield was the first full processor being worked on at Whitefield, Bangalore, India. It was cancelled from the processor roadmap and replaced with Tigerton.

Aliceton

Aliceton was a successor to Tigerton. It has effectively been renamed Dunnington as the original Dunnington was based on the now cancelled Whitefield.

Dunnington - with 6 cores

File:Dunnington Xeon.jpg

Dunnington die micrograph.

Dunnington - the last CPU of the Penryn generation and the Intel's first multi-core (above two) die - will feature a single-die six core design with three unified 3 MiB L2 caches (resembling three merged 45 nm dual-core Wolfdale dies), and 96 KiB L1 cache (Data) and 16 MiB of L3 cache. It is expected to feature 1066 MHz FSB, fit into the Tigerton's mPGA604 socket, and be compatible with the Caneland chipset. These processors are expected to support DDR2-1066 (266 MHz), and to have a max. power consumption (TDP) below 130 watts. They are intended for blades and other stacked computer systems. Availability is scheduled for the second half of 2008. It will be followed shortly by the Nehalem microarchitecture.

5400-series Harpertown and 5200-series Wolfdale

Harpertown is said to be a 45 nm, quad-core processor based on Intel's Penryn microarchitecture with 12 MiB of L2 cache. An older rumour stated that it was simply the 45 nm shrink of Woodcrest, but that has since changed.

Harpertown, which will succeed the current 65 nm Clovertown processors (Xeon 5300 series), will receive 5400 sequence number, with X, E, and L letters indicating performance, regular and low-power versions of the CPU. A LINPACK benchmark shows that a dual Harpertown system could sustain 80 GFLOPS and peak at 102 GFLOPS. [10] [11]

The mainstream lineup (80 watts) will reach from the E5405 with a clock speed in the low 2 GHz range up to the E5450 with 3.0 GHz. The X5460 will clock in at 3.16 GHz and will be rated at a thermal design power of 120 watts. The two highest models E5462 and E5472 will clock in at 2.8 and 3.0 GHz at new 1600MT/s FSB (provided with Seaburg chipset in the Stoakly platform planned at 2007-H2) rated at TDP of 80 watts. Intel also plans to introduce two low-power versions codenamed Harpertown LV, rated at 50 watts, with 2.33 and 2.66 GHz speeds (L5410 and L5430). All Harpertown processors will include a 12 MiB L2 cache, up from 8 MiB in Clovertown. The front side bus besides E5462 and E5472 is expected to be an FSB1333 version across the board, while the slide published by VR-Zone still indicates that the E5405 could run at a slower clock speed.

The dual-core version of the CPU, code-named Wolfdale-DP, apparently will be available with processor speeds of 1.86 GHz, 3.33 GHz and 3.4 GHz at 1066/1333/1600 MT/s FSB, rated at 65/80/80 watts TDP respectively. There will also be a 3.16 GHz low-power version of the processor codenamed Wolfdale-LV, rated at 40 watts.

The socket type is expected to be 771, but it could be something else at higher frequencies.

Harpertown is expected to be launched on 11 November 2007. Wolfdale is expected to launch late in Q4 of this year(2007).

"Harpertown" processors are now available in the Apple Mac Pro line, with configurations as follows: One 2.8 GHz Quad Core, Two 2.8 GHz Quad Core, Two 3.0 GHz Quad Core, and Two 3.2 GHz Quad Core processors.

The entire Intel® Xeon® Processor 5000 family specifications, including "Harpertown" processors are presented at the following Intel web page: http://www.intel.com/products/processor/xeon5000/specifications.htm?iid=products_xeon5000+tab_specs

Codename	Model	Speed (GHz)	L2 Cache (MB)	FSB (MHz)	TDP (Watts)
Harpertown	X5482	3.2	12	1600	150
	E5472	3.0			80
	E5462	2.8			80
	X5460	3.16		1333	120
	X5450	3.0			120
	E5450	3.0			80
	E5440	2.83			80
	E5430	2.66			80
	E5420	2.5			80
	E5410	2.33			80
	E5405	2.0			80
Harpertown LV	L5430	2.66			50
	L5410	2.33			50
Wolfdale-DP	E5272	3.4	6	1600	80
	E5260	3.33		1333	65
	E5205	1.86		1066	65
Wolfdale-LV	L5250	3.16		1333	40

[12]

Gainestown

Gainestown is a quad-core processor based on Intel's upcoming Nehalem microarchitecture. [13]

Beckton

This Nehalem-based, MP-capable processor with eight or more cores is expected to be launched within 2008.

Supercomputers

Supercomputers based on Xeon processors in the top ten of the Top500 fastest supercomputers in the world:

• an Intel Xeon system at SGI in Chippewa Falls, Wisconsin. Machine: SGI Altix ICE 8200 system with 3584 Quad-Core Clovertown processors at 3.0 GHz and InfiniBand interconnect. This supercomputer is listed in third place as of November 2007, ahead of the fastest Itanium and Opteron-based supercomputers but behind two PowerPC-based BlueGene systems.

See also

• List of future Intel Xeon microprocessors
• List of Intel Xeon microprocessors
• List of Intel microprocessors

External links

• Server Processors at the Intel website