Processors and processing speed has changed tremendously in the past two decades . According to moores law microprocessors will double their capacity and become smaller every four years. The two leading processor manufacturing companies in the world are Intel and AMD and both of them have advanced and changed rapidly to meet the global demands and customer needs for faster,better and scalable processor technologies.
The various classifications of computer archietecture arises from the need of the consumer and enterprise users. From Home computing, desktops, workstations, medium range servers and high end servers which are now moving into the era of cloud computing. Critical enhancements have been done in terms of increasing performance,efficiency, clock speed and higher multitasking capabilities is the need of the hour. AMD and Intel have leap forged into the adaptive age with their enhancements in processors.
AMD: The processor speed is one of the prime reasons why processor competition is heating up and the need for each player to be able to compete aggressively in terms of improvements in processors speed, core count multiplication and thereby enhancing performance parameters. Due to the race between AMD and Intel in terms of who is the better one, everything else has been developed and focused on , except for the processor speed which took a back seat . Advanced Micro Devices launched its latest series of Opteron processors, codenamed Magny Cours. These processors which are cooly called as Magny Cours have firmly improved architecture and promise an overwhelming change in the performance speed . This one is called the new Opteron 6000 series offering from the AMD stable and has the 6000 series chips which enhance the performance of the processor by two times in comparison to the older generation six core chips. From the conventional six core to shift to the two core 6000 , the direct impact is on the performance and the design which has been developed in mind to adapt to this this new change in the CPU architecture and the introduction of the this new processor archietecture provides a choice for a 8 core or 12 core CPU's , designed to specifically boost performance and increase memory and computing speeds for enormous capacity workloads.
CPU archietecture changes have been subjected only to the changing demand from the consumers in the various segments and the most critical being the server and storage space in which data centres are being changed every year with the increasing need on space and speed doubling every two years. Enterprise applications are also one of the criteria where the CPU archietectureal changes have been constantly asked for . AMD with its Opteron 6000 Series platform signals a new era of server value, which significantly impacts and influences server computing for the future as a benchmark and moves it performance benchmarks with its alignment to the enterprise applications apce and the server economics by providing end value in terms of performance-per-watt, value and consistency customers demand for their real-world data center workloads.
AMD with its foray with the new 6000 series processors also gave prominence to the new age technologies keeping in mind the cloud computing and virtualization era. By aligning its product innovations AMDS also announced the release of the new 5600 Series chipset, featuring I/O virtualization. The new platform offers a number of benefits, including more memory, thanks to the 12 memory DIMMs per processors, an enhanced memory controller with support for 4 channels of DDR3 memory. This chipset has been designed to complement the advancements in the CPU archietecture and provides state of the art computing advantages with the Quad 16-bit HyperTransport 3 technology which links with up to 6.4 GT/s per link and PCIe 2.0 support.
The new processors are designed to accommodate the 2- and 4-socket enterprise servers, providing a choice for the industry's first 12 core and 8-core processors, featuring L2 of 512K/core and 12MB of shared L3 cache. There's also AMD's CoolCore, PowerNow, CoolSpeed, APML technologies, Enhanced C1 State and more at AMD, the development a team leading the server platform design.
Over the last few years, CPU archietecture enhancements in AMD - commenced first with "Shanghai", a phenomenal quad-core processor that delivered up to 35% more performance while consuming 35% less power and then with "Istanbul", the world's first 6-core x86 processor
The new AMD Opteronâ„¢ 6000 Series platform (aka "Magny-Cours") that is helping redefine today's data center. When compared to the earlier categories and the three generations of computer chips and the traditional archietecture , the demand was for a finite number of transactions and storage was limited. But with business demanding and becoming global the need for space and the performance also changed. The new AMD's Magny-Cours chips offer twice the performance of the company's Istanbul chips, support for four channels of DDR3 memory for up to 2.5 times the overall memory bandwidth and 33 percent more memory channels than Intel's two-socket offerings.
Intel Westmere
Intel's perceived value as a leader stems out from the intel legacy which is virtually unchallendab has been designed for maximum performance and reliability, these massively scalable 2-way to 256-way, 64-bit multi-core servers are designed to provide exceptional scalable performance and mission-critical class reliability for your most data-demanding applications and virtualization projects.
Intel made strides in innovation and subsequent enhancements with its new release and with its next line of Xeon server chips , which positions intel as ramps up its production to the as Intel ramps up with its double focus on improving performance and the change in the architecture to cater to the various segments the westmere chips with the advancements and improvements with the chip production to the 32 nm process configuration and westmere will be and is the answer to cater to the various different segments from the desktops to westmere 32 nm focused on the enterprise server segment and the introduction of a new range of performance enhanced powerfull computing capacities with the introduction of the Xeon processor range and is different from the earlier generations which were using the 45 nm process and were called the 5500 and 3500 series .
Intel's new offering with the 32 nm westmere is highly anticipated by consumers and competeros alike. Since the need for speed is more than ever with AMD's new offerings on the CPU archietecture in the server economics. Intel's move on Nehalem server chip technology which will have eight-core Nehalem-EX server chip and is Intel's fastest server chip to date. Intel is targeting Nehalem-EX at high-end systems running data-intensive applications such as databases. Focussing on the end customer value the Nehalem microarchitecture will drastically cut data bottlenecks that have plagued earlier architectures to improve system performance. By communicating faster with the integrated memory controllers with systems components with the intergration into one wholistic system inside the CPU will meet the specific needs and cater to high-end servers. Itanium chips which are state of the art have the new Error-correction technology which will cut down on the data corruption by increasing security levels and hence position westmere into the reliability factor in terms of the arechietecture of the chip and deliver highly reliable server performance derived from its high-end Itanium chips will help reduce data corruption and ensure reliable server performance. With the need for technology distinctions and uniqueness, Intel with it's separate buffered memory chips which can store temporary data by isolating this memory pieces can help faster processing speeds and therby increasing performance and contributes to faster task execution. Improved performance benefits and power usage which is optimized in the Westmere CPU archietecture which include an intergrated graphics processor with the CPU in a new multichip package which is integrated and hence due to the integration with the chips archietecture which enables the provision with the usage of the advanced 32-nm process. Westmere chip package will include an integrated graphics processor with the CPU in a multi-chip package. The high levels of integration will lead to smaller chip sizes that deliver better graphics and processing performance while consuming less power.
Security also needs to evaluated when performance is provided. Westmere with its new encryption advancements in data security , has the and the decryption called Advanced Encryption Standard (AES),. This will help secure data residing in servers or virtualized environments. Power usage Both the Nehalem-EX and Westmere-EP chips will be capable of shutting down idle cores to save power and provided optimum value and performance in terms of costs and efficiencies in terms of operational costs.
Smaller process core size which is new and multi core and provides better performance. With the Intel core 2 enables processors with the six cores and moves away from the Intel Core ads on the desktop roadmap with the new advanced graphics processor card.
Key Features of the 32nm westmere archietecture are
The power packed new age technology feature called the Intel Turbo Boost technology which helps in enhancing performance features and the overall clock speeds and data transfer. With the developments in the core and thread archietecture
Intel Hyper Hyper-Threading technology (2 Cores, 4 threads)
Integrated graphics, discrete /
Key Features of the Westmere processor are
Intel microarchitecture codename Nehalem on 32nm
Multi-Chip Package (MCP) processor with: processor cores built on 32nm 2nd generationhigh-k metal gate process
integrated graphics controller & memorycontroller built on 45nm high-k metal gate process with graphics processor and more secure encryption and decryption to provide higher security features and reliability.
Analysis
The various classification processors are provided below for the 32 nm to Instanbul
32nm
Intel Core i7 980X Extreme Edition Processor
Magny-Cours
AMD Opteron 6134 Magny-Cours 2.3GHz 8 x L2 Cache
Six-core
Intel Corp. Processor Xeon HC E7450 2.40GHz 12MB L3 Cache
Intel Xeon / 2.4 GHz processor
45nm
AMD Phenom II X4 945 Quad Core Processor
Istanbul
Aperture Istanbul Alex PHT / Pamuk Webb ISBN RKY7SL 1-597-11034-5
Its important to know the process which goes into making the new offerings from the stable of AMD. The process is implementation of the Opteron are both baked into the 45 nanometer silicon and using the insulation process which goes into making the Opteron 4100 and 6100 using the 45nm process. With the L1 data of of 64KB the Magny cours gets an additional 512KB of L2 cache. This means that there is additional capacity into the cache memory which helps in contributing to the overall performance of cache per core and the data transfer speeds also increase due the expanded capacity in the L2 cache and L3 cache and impacts the overall capabilities which delivers significant speeds and performance capabilities using the 45nm process. This makes AMD move up a notch till the 6 MB of L3 cache per processor and its double stuffed which helps the Magny-Cours to 12 MB of L3 cache per socket. So the Magny-Cours therefore come with either eight or twelve cores activated. The core distinctive features of theAMD "Magny-Cores" Opteron 6100 processors. Clock speeds, higher configuration and other double packed processors all come together for the Magny cores Opteron 6100s which have been classified to cater to the various segments . From the regular standard ones to the highly efficient versions. With the four core and six core predeccors come in standard, Special Edition, and Highly Efficient versions, With the four-core and six-core predecessors to the Lisbon and Magny-Cours Opterons, the standard thermal envelope parts were rated at 75 watts using AMD's Average CPU Power or ACP test. This test is a critical element in the quality check process and this helps in checking all the related performance parameters of the Magny Cours. From this has evolved the chips which form into the SE and HE versions archietecture. The major difference between these two are the clockspeeds and power usage. With SE chips the clock speeds show a slight increase in the transactional speeds and burn or use upto 95 watts . While the HE versions are also equally very good, but in comparison to the SE , the HE ones have slower clock speeds and also lesser power consumption at 55 watts and can be qualified efficient in terms of power optimization. From the Highly Efficient category came in the Extremely Efficient, or EE, Opteron parts and are better in terms of the power usage and the clock speeds. The EE one's showed a mere usage of power consumption of 40 watts but they are not available with the 6100's but are an intergral part of the Magny cours in the future versions.
AMD has experimented with the processor dyes by making provisons for faster processing speed and multitasking and has done tremendous improvements for faster processing speed and better multitasking with the AMD Socket G34 CPU the Socket G34 with improvements in the archietecture AMD has done good improvements in the CPU in comparison to Intel. Being the runner at the forefront in the dense memory category and memory cell technologies like the Z -RAM. From four to six cores and then to the eight cores everything has moved for the AMD to have which is Unlike Core i7 and Gainestown Nehalems [Nehalem-EP] have 8 MB L3 per four cores, fed by a 216-bit Triple-Channel DDR3 while Beckton [Nehalem-EX] will have 8 cores and 24 MB L3, fed by a 288-bit Quad-Channel memory controller. Latency also needs to be managed , with these advancements its important that the connection between the processor and the cache memory has to be intergrated well with the sharing of the FSB . This helps in the transfer of data into the CPU archietecture as a single package which has load balancing capabilities from the data speeds to increase transfer rate and thereby improve performance many times with technologies like the Hyper Transport and impact the speed of data transfer from the standard to the current improvised modified one.
With this the memory also gets a facelift with the 12 DIMMS per CPU socket and with two of the DDR3 channels exposed . This arrangement will help in the access of the die and the memory channels at the same time. With the increase in the dynamic memory , which is primararily looking at numbers, with DDR3-1333 memory which gives the opertron the powerful bandwidth.
With Intel Hyper Threading Technology which can now go upto Up to 8 cores and 16 threads per socket. The new Intel Xeon processor 7500 series offers increased performance with 45nm manufacturing technology and increased headroom for multi-threaded applications and data demanding applications . This perfectly works in sync and alignment with the enterprise needs of being enabled to provide improved performance for virtualization and significantly impacting system utilization. With the Intel Westmere microarchitecture codename Nehalem which not only helps integrate user environments in multiple applications but also positively boosts performance and enables better multitasking . This is perfect in terms of the data centre management and needs in terms of deployment thru improved performance-per-watt45nm. Processor technology has also changed with Intel's hi-k process technology . This technogloy is unique and provides larger cache on the die which impacts and provides better performance and reduced and reduced transistor gate leakage which improves the overall energy efficiency24 MB of L3 cache.
With the westmere CPU archietecture there is a direct impact on the data transfer and increased bandwidth. Efficiency is increased considerably with the cache to core data transfer Increases the flow of information and helps in large computational capabilities.
One of the key issues with older generation CPU's was the latency , which used to affect the data processing and processor performance. With the new Intel Westmere the latency is reduced considerably with storing larger data sets which are now set very close to processor. This helps in reducing the number of trips to main memory and increases the performance and speed.
A new technology innovation called the Intel Quickpath Interconnect is the technology which establishes point to point connections and increases the connectivity between the Input output Hub and the processor and therby increases speed of transactions by contributing the quickpath methodlogy.
Memory capactity and speed has also increased performance capabilities and flexibility which are scalable . This puts the westmere in the reliable space with the RAS which enables scaling of the processor from 2 to 8 sockets and delivers outstanding interconnectivity, reliability, availability, and serviceability (RAS) with design-scalable configurations for optimal balance of price, performance, and energy efficiency
From the AMD stable when compared to the earlier generations of the processors, the most powerful is the Opteron 6100s. They are very close and similar to the new Xeon 5600s and the low-end of tomorrow's Xeon 7500s. The base of the archietecture which is the two socket boxes which are theXeon 5600 and the new era four socket ones are aimed at the server segment which provide enhanced features and performance. Multiple server manufacturers are jumping on to the bandwagon to provide with servers from two socket to four socket which are focused on the provision for extended memory compatible servers which focus on the extended memory they offer. This is one of the primary reasons which paved the way for the need to produce better chips and server memory . Since higher computation and processing seed is brn and because in a lot of cases, server buyers are more constrained by memory capacity and bandwidth than processing capacity. This is yet another way that the uptake of virtualization puts pressure on the chip and server makers.
With the market for eight-socket and larger boxes dwindling and the four-socket market expected to see a similar decline, it is no wonder that AMD decided to bifurcate its product line in a different way from the old Opteron 1000, 2000, and 8000 series and focus on creating platforms with one, two, or four sockets with two distinct processors and memory and energy profiles. Those looking for the densest and most energy efficient server platforms with one or two sockets are expected to go for the Opteron 4100s, while those looking for more cores and more memory per socket as well as up to four sockets.
AMD wants one set of chip cores and chipsets that get implemented in two slightly different packages to address the bulk of the market. Intel still has two distinct processors and chipsets. AMD wants to compete on price and ramp up its volumes.
Intel Westmere Core i5 32nm Processors
Targetting is one of the primary need to ensure that the introduction of a new line of server archietecture will land successfully. With the Core i5 launch being done and positioning itself as a Westmere is only going to help Intel in capturing the desktop segment. Code named "Clarkdale" which is primarily for the desktop segment will now for the first time include a graphics core which is intergrated into the two and four cores. This will impact the performance and boost the graphics capability for the CPU archietecture. With this distinction on the two and four cores the memory takes a notch up with the dual-channel memory support which provides more memory and helps in handling multiple transactions.
With advanced features in the Westmere , they are also perfect to provide optimum power usage for the chips and are aptly positioned for the mobile usage. With power consumptions regulated and enhanced management of power supply and based on the usage, only paves the way for westmere to lead the mobile segment with its state of the art, new age, high performance and optimized archiectecture . Memory usage is framed into the two core and four core which provides increases data transfer speeds and hence improves the overall performance of the westmere CPU and the devices it has been deployed for,
The new Intel 5-series chipsets
Not so long ago, chipsets used to be split and isolated and then managed. Since manageability was the biggest factor Intel chipsets used to be split into two separate parts, with the Northbridge controlling interfacing between the processor, memory and graphics subsystems, while a secondary chip, called the Southbridge, is used to control the I/O storage functions of the motherboard. With the change in the demands and the computing needs and levels, Intel changed the and brought out the new Intel 5-series of chipsets which solve the problem of the integration functionality. This helps in directly contributing to the ease of use, performance driven in terms of the memory and the storage functionality and in a single chip provides all the controlling capabilities and features of the controlling the memory graphics and extra features to interconnect peripheral devices into a single chip and provides a unified way to manage and increase memory and graphics performance.
Replacement and change is constant and it's but evident that the Intel P55 Express chipsets will soon replace the earlier versions. Since they are proven to increase the performance in terms of enhanced memory , larger storage and capabilities to handle complex and multiple transcations and computations. This will Intel's mainstream chipset line and it will retain the proven technology feature which is no change in the northbridge and southbridge
Conclusions
The Intel systems have only 24GB ram against the 32GB ram on the 2S magny cours. That's why the 100GB database test favors the Magny cours by a large margin. It's interesting to acknowledge that the Xeon Westmere EP, something was missing: a challenger, a competitor to make things more exiting. It's very clear based on the above analysis that the Xeon X5670 is on eye to eye with the AMD. But the AMD Magny cour has advantages like larger bandwidth, DDR Memory controller, Hyper Transport which indicate that its better in terms of speed, performance and with the extra cores it's positioned to deliver better performance for advanced and mammoth computing needs.
One of the main contentions is the comparison in the terms of the graphic processors chipsets . The chipsets in the Magny cour and Westmere almost come very close to the individual graphic processors . Intel has done more than what it's processors can deliver by upgrading the intergration with the cache memory dyes and the FSB into 1333 MHZ in the core 2 Due intel processor and this extend the performance of Westmere over the Magny cour in terms of sheer performance and how it can actually contribute to the rising needs of the enterprise server space and the various performance enhancement issues
Compatability is key and hence the need for the data compatibility and security is enhanced for the new PC's and the data centre's needs which are important and hence the need for their improvements in the performance architecture .Performance lead of Intel over the AMD's Athlon64, The new 1333 MHz FSB Intel Core 2 Duo processors support Intel 64 technology, Enhanced Intel Speed Step (EIST), Execute Disable (XD), Intel's Virtualization Technology (IVT) and Intel's Trusted Execution Technology.
AMD has distinctive advantages by which the Magny cours have some advantages clearly over the Intel servers. In terms of performance of the Magny cours because of its architecture to have integrated the FSB and providing higher transfer speeds into the cache memory and this is one of the advantages in transferring packets into the processing.
With intel's advanced encryption technology security is one of the major distinctions in terms of intel being able to scale itself to meet the data centre's crunching need for the security of the data.
Processing Speeds: this has been a heated war with all the major runners and hence the need to be able to go farther lengths with speed which is Magny cours have so far been able to improve with it's changes in the architecture and new enhancements in the DDR and the various combinations from the earlier generations have made the Magny cours a little faster and better than the westmere's. But finally the Intel Xeon Westmere will lead the race since the segmentation and the target segment with its value.
http://www.google.co.in/search?hl=en&source=hp&q=amd+magny+cours&aq=0&aqi=g10&aql=&oq=AMD+Magny+&gs_rfai=
http://www.amd.com/us/products/embedded/processors/opteron/Pages/opteron.aspx
http://www.amd.com/us/aboutamd/newsroom/Pages/newsroom.aspx
http://www.amd.com/us/press-releases/Pages/amd-sets-the-new-standard-29mar2010.aspx
http://download.intel.com/pressroom/kits/32nm/westmere/32nm_WSM_Press.pdf
http://download.intel.com/pressroom/kits/32nm/westmere/News_Fact_Sheet.pdf
http://www.zdnet.com/blog/computers/first-intel-westmere-32nm-reviews/1149
http://www.amd.com/us/products/embedded/processors/opteron/Pages/opteron-6100-series.aspx
