Apple may be considering a more powerful Mac Pro: that’s what we know

0


For a few days Apple iphone 13 With a new processor Bionic chip A15With 15 billion transistors, that’s a 27% increase over the 2020 A14.

Most of these additional transistors may have been used for new GPU (Graphics Processing Unit), new AI neural engine and a few more.

Desktop version of A15 with faster clock speeds is a newer version Macbook Air, Macbook Pro, iMac When Mac Mini Later this year, we doubled the cache (32MB). But what about Mac Pro Workplace??

The more power you have, the more cache you get

Now Apple released an intriguing job listing on September 10. There he announced a CPU vacation. Microarchitecture RTL cache Based in the United States.

This is the fifth offer published by the Cupertino giant mentioning “CPU cache” and the third mention mentioning multiprocessor systems. Apple is now considering a product that uses more than one processor. Currently, the Mac Pro is the only candidate.

(Note that Apple may refer to multiple processor families in the same SoC, such as central processing units, graphics processing units, neural processing units, etc.).

The final job description is “CPU Multi-Level Cache Subsystem Architecture and RTL Development for Multiprocessor Systems”, and how to manage cache (ultra-fast memory acting as the first port) for multiple cores and multiple physical processors. Explain about. The number of calls between the core of the processor and the rest of the system is very important.

another Job information (Embedded Linux Engineer), Apple said a successful candidate was “one of the most visible teams verifying complex multidisciplinary systems-on-chip in a multiprocessor environment for future Apple products. It will be a department. “Before adding, they” will develop Linux Next-generation Mac product environment that enables new cutting-edge technologies (sic). ”

(Image credit: Apple)

No more cache, a different cache?

It turns out that each of the Apple M1’s two high-performance cores has a combined 320KB L1 cache and shares a 12MB L2 cache. The remaining four power-efficient cores each have a combined 192KB L2 cache, sharing a 4MB L2 cache. M1 is basically inspired by A14, so there is no L3 cache.

However, to start Xeon from a Mac Pro, Apple must have a different type of processor. It has a different cache architecture, has a faster clock speed, and can support significantly more memory than the 16 GB currently available (located at: SoC with system configuration in package).

Due to the fact that it refers to multiprocessors rather than focusing on multicores, Apple believes that adding more processors instead can improve performance while keeping the core count low.

The more cores you have, the greater the need for L3 cache. Each core of 64 cores Pro 3995WX Chipper For example, 64K L1, 512K L2, and 256MB L3 are shared (4MB each). By the way, this is slightly higher than Apple’s basic fuel-efficient L1 + L2 cash quota. This may indicate that Apple is reluctant to add another complex layer (i.e. L3 shared cache).

So you could end up with two dummy M2Xs that provide 64GB of RAM (32GB each) or four Mac Pros with 128GB of memory (and in some cases DDR5). That covers three of the eight current memory configurations, up to 1.5TB on current Xeon-based Mac Pros, but I still find it insufficient.

If you want to take it seriously, Apple has to find a trick that will allow the M1 successor to take over a lot more system memory.

No Xserve server

But one thing that’s unlikely is that Apple will bring back the Xserve. server A brand that supplies rack servers to businesses around the world. About 13 years after the launch of the last Xserve, the market has changed to the point of being unrecognizable. Dell, HP, Lenovo, etc. still exist, but market dynamics have been transformed by hyperscalers such as Google, Facebook, Microsoft, Alibaba, and Amazon.

These are companies that have a huge desire for computing power and aren’t afraid to set the agenda when it comes to what they want (hence the reason AMD and Intel purchased). FPGA companies in recent years). I don’t want Apple to be competitive in this innovative low-margin environment.

However, it’s no surprise that Apple is following all the other hyperscalers and launching its own dedicated server chip for internal consumption. After all, with hundreds of millions of iCloud users and plans to become a service giant, it’s in Apple’s best interest to do what the customer does on the infrastructure side.

It reduces reliance on third parties by owning the entire vertical stack and providing end users with unique features that are not available anywhere else. Apple Privacy Relay, Super efficient video encoding technology, realistic vmeeting of ideas A tool that works with low bandwidth. A pipe dream (line)? Let’s see.


Share.

About Author

Leave A Reply