My New Website

aaliyan.home.blog

Search This Blog

Saturday, 29 April 2017

The Miracle Of AMD: Zen Architecture is best AMD Released Ever

post1
The "Zen" design
AMD says its new x86 design known as "Zen" may be a rare event within the semiconductor trade as a result of it is a "clean-sheet" for the corporate. Performance-wise, the Zen microarchitecture sets intent on not solely tackle heavily multi-threaded tasks however moreover improve-single thread performance by using better instruction-level parallelism (ILP).
The design features a way larger instruction scheduler window, permitting the C.P.U. to schedule and send an additional value into the executions units. a brand new micro-op cache means the L2 and L3 cache are often bypassed once utilizing frequently-accessed micro-operations.
The company additionally speaks of a "neural network-based" branch predictor which is employed to reduce prediction errors by permitting Zen to be a lot more intelligent when it comes to initializing optimum instructions and pathways for future work.
Finally, the last performance feature to be enforced by AMD's latest design is SMT (simultaneous multithreading), which allows 2 threads per core. this is often a feature we have seen in some IBM, Intel and Oracle processors and marks a departure from the CMT (clustered multi-thread) style used antecedently in Bulldozer architecture.

Whereas bulldozer used modules that performed very like a dual-core, dual-thread processor, Zen adopts SMT for what AMD calls a C.P.U. complex (CCX) style. CCX may be a four-core/eight-thread module that includes its own L1, L2 and L3 cache.
The cache hierarchy appears like this: every core receives a 64KB L1 cache for instructions and data, a 512KB L2 cache and a shared 8MB L3 cache that services four cores. The L1 cache has been modified from write-through to write-back, giving lower latency and better bandwidth. AMD is advertising up to 5 times larger cache bandwidth and I believe this claim is for the L3 cache whereas the L1 and L2 are nearer to 2 times quicker.
The Ryzen seven series features 2 CCX units, enabling eight cores and sixteen threads. It's doable for AMD to disable individual cores with the CCX and undoubtedly little question about this is that How they are going to produce the six-core/12-thread models.
For those questioning, the CCXs communicate across the high-speed 'Infinity Fabric', a successor to HyperTransport, that itself had replaced the front-side bus in AMD's Opteron, Athlon 64, Athlon II, Sempron 64, Turion 64, Phenom, Phenom II and FX families of microprocessors.
Infinity cloth (essentially HyperTransport two.0) may be a versatile interface/bus that facilitates the exchange of information between CCXs, system memory moreover as I/O and PCIe controllers. This provides Zen with powerful command and control capabilities for period estimations and changes to core voltage, temperature, socket power draw, clock speed, and more.
Those command and control capabilities are essential for AMD's SenseMI technology, that is essentially a package of 5 connected "senses" that are dependent on subtle learning algorithms and it uses the same C&C methodology to fine-tune performance, power characteristics of the cores, manage speculative cache fetches, and perform AI-based branch prediction.
Each Ryzen processor features a grid of interconnected sensors that are fine-tuned to 1mA, 1mV, 1mW and 1-degree C with a polling rate of 1000/sec. These sensors give telemetry information that feeds into the infinity fabric control loop and permits the C.P.U. to create changes that are determined by current and expected operating conditions.
The 5 senses are represented as the text following:
Pure Power A closed-loop control system that optimizes clock speeds and frequencies to deliver the best performance at the least power consumption. AMD says Pure Power is used to enable "lower power for same performance."
Precision Boost Works in parallel with Pure Power but instead offers higher performance at the same power consumption level through incremental clock speed boosts of 25MHz.
Extended Frequency Range (XFR) Perhaps the most interesting SenseMI feature, XFR allows the CPU to increase its clock speed beyond rated boost clock frequencies depending on the temperature levels. This sounds similar to NVidia's GPU Boost 3.0 technology, which often pushed Pascal GPUs well above their rated frequencies. AMD says XFR will reward enthusiast coolers by allowing Ryzen's clock speeds to scale with cooling solutions.
Neural Net Prediction The use of the term "neural net" is a bit loose here, however, this technology will pre-load instructions by anticipating the actions a user can perform ahead of time.
Smart Pre-fetch Learns data access patterns to pre-fetch necessary data into the CPU cache so it can be immediately accessed where necessary.
The AMD AM4 Platform
Along with the Ryzen processors is AMD's all-new AM4 platform which will launch with three key chipsets: the entry-level A320, the mid-range or mainstream B350, and then enthusiast-grade X370.
Chipset    USB (3.1 G2 + 3.1 G1 + 2.0)    SATA    SATAe    PCIe Gen 2    Multi-GPU    Overclocking
X370    2 + 6 + 6    4    2    8 Lanes    Yes    Unlocked
B350    2 + 2 + 6    2    2    6 Lanes    No    Unlocked
A320    1 + 2 + 6    2    2    4 Lanes    No    Locked
All of the chipsets support NVMe PCIe SSDs, SATA, SATA Express, dual-channel DDR4 memory, native USB 3.1 Gen 2 and more.
The X370 differs from the B350 model by offering more USB 3.1 Gen1 ports, more SATA ports, more PCIe 2.0 lanes and support for multi-GPU technologies. As you can see in the table above, the A320 sacrifices another USB 3.1 port and two more PCIe 2.0 lanes in addition to being locked (no overclocking).
The great thing about the AM4 platform is how flexible it is. Socket 1331 streamlines AMD's socket infrastructure (AM3 and FM2+) into a single interface that can support both Ryzen and seventh-generation APUs. AMD also claims that future Raven Bridge and Zen processors will be supporting AMD’s AM4 platform. The company plans to keep this socket for several years through 2020 and will integrate future technologies such as PCIe 4.0 and DDR5.

No comments:

Post a Comment

Comments are welcome

Our Advertisers