Transcript dual-core software licensing

Multi Core Roadmap
Dual Core Processors
Dual Core
• This term refers to integrated circuit (IC) chips that contain
two complete physical computer processors (cores) in the
same IC package.
•Typically, this means that two identical processors are
manufactured so they reside side-by-side on the same die. It
is also possible to (vertically) stack two separate processor
die and place them in the same IC package.
•Each of the physical processor cores has its own resources
(architectural state, registers, execution units, etc.).
•The multiple cores on-die may or may not share several
layers of the on-die cache.
Intel Pentium Extreme Edition 965Specifications & Features
•3.73GHz Dual Core Processor
•1066MHz "Quad-Pumped" front side bus
•.065-micron manufacturing process
•Hyper-Threading Technology
•4MB on-chip, full-speed L2 cache - 2MB for each core
•Intel EM64T Extensions - 64-bit computing
•Execute Disable Bit - For enhanced security
•1.20 - 1.33V operating voltage range
95 - 130 watts TPD (Thermal Design Power)
•Die Size: Approximately 140mm2
Approximately 376M Transistors
(cont)
•We should also note that the Pentium Extreme Edition 965 has full
support for Intel's Virtualization technology, which gives users the ability to
run multiple operating systems in independent environments,
• full support for EM64T to run 64-bit operating systems, and the Execute
Disable Bit to prevent certain types of malware driven buffer-overflow
attacks.
•Red indicates dual core speeds vs single core
Multi Core
•The multi core system is an extension to the dual core system except that it would
consist of more than 2 processors.
•The current trends in processor technology indicate that the number of processor
cores in one IC chip will continue to increase.
•If we assume that the number of transistors per processor core remains relatively
fixed, it is reasonable to assume that the number of processor cores could follow
Moore's Law, which states that the number of transistors per a certain area on the
chip will double approximately every 18 months.
• The optimal number of processors is yet to be determined, but will probably change
over time as software adapts to effectively use many processors, simultaneously.
•However, a software program that is only capable of running on one processor (or
very few processors) will be unable to take full advantage of future processors that
contain many processors cores. For example, an application running on a 4-processor
system with each socket containing quad-core processors has 16 processor cores
available to schedule 16 program threads simultaneously.
• At the Intel Developer Forum Intel also publicly announced its intention to develop
and manufacture multi-core processors in its Itanium Processor Family.
OS Support
Software Licensing Models
•
In order for multiprocessor systems to be effective the operating
system must be able to detect multiple processors and provide a
mechanism to schedule separate processes and/or threads on
the physical and logical processors present.
•Microsoft provides several versions of Windows that have this
capability, as do many Unix/Linux-based operating systems
Price/Performance
•There are many factors that affect the cost of processors and computer systems and therefore
affect the price that a computer system vendor will charge for a system.
•Traditional dual processor systems require two separate processors in two sockets. Both may be
physically located on one main computer board or they may each have their own separate board.
Using separate boards may make systems more modular (and possibly easier to maintain).
•Dual processor systems provide excellent performance because both processors can operate
independently since each has their own computing resources.
•Power consumption goes down because the electrical pathway between chips has been
dramatically reduced.
•Dual core systems are capable of taking advantage of parallelism for improving performance
rather than trying to increase performance of a single thread using frequency bumps and processor
complexity.
•Since power consumption increases exponentially with processor speed so a small drop in
processor speed allows significant savings in power consumption..
Clock Speed:
•It is the flattening of the clock speed curve that some are reasoning why a shift to dual core?
•Some have surmised that AMD and INTEL have hit clock speed walls and another route is
being taken to continue the performance curve and stay top of mind with new product releases.
• The problem with winding up clock speeds is heat.
•At present the processor engine can operate at only so much RPM before the engine will seize.
• Heat is the enemy of any processor and high clock speeds mean high heat and that means
errors. A Windows PC running at 10GHz isn't much good if it can't make it past booting up
before crashing.
•That heat comes from power. It takes a lot of juice to crank up a processor to high clock speeds
and a processor with that much electricity running around the die is prone to noise. It's not
audible noise like a high RPM cooling fan but electrical noise otherwise akin to interference.
• The pathways on a processor are microscopically close together. The more power that runs
through these pathways due to the requirement of higher clock speeds means that there will be
a small amount of electrical radiation from one pathway to the next.
•That leakage could corrupt the data in another pathway. Corrupted data means errors which
means a program could get cranky.
DUAL-CORE SOFTWARE LICENSING
•Many major software companies, including Microsoft and Sun, have publicly
stated that a dual-core microprocessor will require only a single software license.
•Other companies, particularly Oracle, are hazier on that issue, and haven't yet
made a definitive statement as to whether you'll need a dual-processor license to
use a dual-core chip.
•But for every piece of consumer desktop software that I know of, and all
•Microsoft server operating systems and applications, users will be able deploy
dual-core microprocessors for the same licensing costs that they currently incur.
While that's not a big deal for the desktop, that could be a huge savings for multicore servers.
Software
The next big question is: How do application developers take advantage of dualcore processors?
Answer: The same way you exploit multi-chip SMP, by instituting threading.
Programmatically, a dual-core system implemented on a single microprocessor is
identical to a dual-core system that uses two single-core microprocessors.
•As long as the operating system's scheduler is thread-aware, it can allocate
threads and processes efficiently. All modern versions of Linux, Solaris, and
Windows are completely thread-aware.
•Within specific applications, today's best practices already suggest that you code
in threads, whether you're writing native code, or managed code running in a Java
Virtual Machine or on the .NET Common Language Runtime.
AMD64 Dual Core Physical Design
• 90nm
– Approximately same die size as 130nm single-core AMD
Opteron processor*
– ~205 million transistors*
• 95 watt power envelope
– Fits into 90nm power infrastructure
• Socket 940 compatible
*Based on current revisions of the design
One die with 2 CPU cores
each core has its own 1MB L2 cache
up to 16 gigabytes of DDR400
Quad-Core Chips
AMD to Demo Quad-Core Chips in Mid-2006 [UPDATED].
AMD’s Quad-Core Processors to Show Up Soon
Advanced Micro Devices’ quad-core processors will be demonstrated as
early as in the middle of this year and, perhaps, will be unveiled even
earlier than expected according to some analysts. The forthcoming chips
with four processing engines will be demonstrated on the next-generaion
AMD server platforms that will ship this year.
Server roadmaps shown by company executives confirmed that the
company's Opteron processor will move to a quad-core architecture
beginning in 2007, together with the ability to scale to 32 processors
and above. Similar roadmaps presented for the company's desktop
processor roadmap lacked the quad-core disclosure, however, which
could indicate that quad-core, AMD-based PCs will arrive in 2008 or
later.
AMD is also eyeing "cluster based multithreading," a
technology specifically designed for large enterprise
clusters. The technology would allow other
microprocessors to share specific blocks on the chip, such
as the floating-point unit. Designing this technology in will
cost an additional 50 percent in die resources, but yield an
additional 80 percent improvement in performance, Moore
said.
Although AMD currently manufactures its microprocessors at a
single fab, a second, Fab 36, is nearly complete next door to
AMD's Fab 30 in Dresden, Germany. Scheduled to come on line
early next year, the fab will initially produce 90-nm wafers but
quickly shift over to 65-nm lines, said Daryl Ostrander, senior
vice president of logic and manufacturing.
Intel Quad Core Processors
Just as the bragging rights for dual-core chip supremacy are dying
down, Intel gave the first glimpse of a quad-core chip coming next
year.
Clovertown, a four-core processor, will start shipping to computer
manufacturers late this year and hit the market in early 2007. Clovertown
will be made for dual-processor servers, which means that these servers
will essentially be eight-processor servers (two processors x four cores
each).
Core expansion will be a dominant theme for Intel over the next few
years, said Chief Technology Officer Justin Rattner. By the end of the
decade, chips with tens of cores will be possible, while in 10 years, it's
theoretically possible that chips with hundreds of cores will come out, he
added.
Multiplying the number of cores brings distinct advantages
• First, it cuts down overall energy consumption for equivalent levels of
performance. If the recent Core Duo chips released for notebooks from
Intel had only one core, the chips would consume far more power, he
said.
•Integrating processor cores into the same piece of silicon or same
processor package also increases performance by reducing the data
pathways
"To go from core to core can be a matter of nanoseconds," Rattner said.
"As soon as you move cores together you get an automatic improvement
in available bandwidth."
Nonetheless, adding cores requires careful planning. Energy efficiency,
data input/output and memory latency (the time it takes data to go from
memory and the processor and vice versa) will be major issues with
each level of core expansion.
The 45-nanometer process is right on time, according to Intel.
The Santa Clara, Calif.-based chip giant has created test chips made
on the 45-nanometer process and will likely begin shipping
processors, flash, and other chips based on that process in the
second half of 2007, according to Mark Bohr, director of process
architecture and integration at Intel.
One clear part of the process, however, is that Intel will use "dry," or
standard, lithography techniques for 45-nanometer chips.
Lithography is the art of drawing circuit patterns on chips through
optical and chemical processes. "We are committed to dry
lithography for this (the test chips) and manufacturing," said Bohr.