Although Intel still holds the stronghold on CPU markets with its XEON processors. But since the Zen-based EPYC processors by AMD which offers impressive performance in fewer prices than Intel’s has changed that. Now AMD has launched AMD Zen 2 EPYC CPU which is based on 7nm. This provides 2x performance as compared to last year’s Zen processor. These new processors are the parts of ROME family and currently being sampled to consumers.
CEO of AMD Lisa Su said
the company is “betting big on 7nm” and the innovations that come with it. AMD claims Rome will be the world’s first 7nm datacenter CPU, will offer increased instructions per clock (IPC) throughput and a big overall performance uplift.
AMD started to design Zen series with a 14nm architecture. Then after further optimizations, they brought Zen+ which was based on 12nm architecture. AMD now decided to jump to 7nm to boost the performance instead of making it on 10nm architecture. This could provide long lasting life cycle. This step let AMD boost the performance by 2X, reduce power consumption and boost in IPC performance by 25% of AMD Zen 2 EPYC as of the previous one Zen+.
The following are the advances that 7nm allowed AMD to make with its Rome-based EPYC processor core micro-architecture.
AMD Zen 2 EPYC Advancements
- 2X Performance per Socket
- 4X Floating Point performance per socket
- Previous gen socket compatible
- Re-optimized micro-op instruction cache
- Increased the size of the micro-op cache
- Doubled floating point width to 256-bit
- Doubled load store bandwidth
- Increased dispatch and retire bandwidth
- Forward compatible with next gen Milan platform that supports PCI Express 4 connectivity
- Improved branch prediction
- Better instructions prefetch
- Results in maintained high throughput for all modes
The CEO said that they are continuously updating the specs and features of AMD Zen 2 EPYC as they will come in next horizon event. Moreover, she said that
the company’s 7nm Second Gen EPYC Rome processor top-end CPU will be 64-core, 128-thread beasts that are socket-compatible to the previous gen EPYC platforms as well as the company’s next-gen Milan server platform that supports PCIe 4. As a result, 7nm Rome 2nd Gen EYPC server processors will offer 2X overall performance per socket and 4X the floating point performance.
AMD showed a demo of single socket AMD Rome EPYC 64 core server running CRay with dual-socket Intel Xeon Platinum 8180M Server (56 cores total, 28 per socket, 112 threads) to show ray-tracing benchmark designed floating point CPU performance.
The AMD processor was at stock frequency and was air cooled. This was in prototype system as AMD yet has to optimize it further. The single socket, 64-core EPYC server edged-out Intel’s Platinum-class dual socket server with ease. Kindly take a look at the video below to check.
Considering CPU vulnerabilities like Spectre and hardware mitigation AMD has added some security enhancements in Zen 2. AMD also has increased the number of encryption keys for virtualization to allow companies to support more virtual machines.
AMD’s latest EPYC processors will face tough competition in form of Intel’s Cascade Lake SP and Cascade Lake AP processors. Both of them are built in 14nm architecture and Cascade Lake-SP is expected to ship in 2019. Cascade Lake AP will be shipped in the 1st half of 2019 and will be available in two-socket configurations. These will be available with up to 40 cores. This means that a dual socket system will be having the potential for 192 total threads and 96 physical cores on board.
Via Hothardware Via AMD
Get real time update about this post categories directly on your device, subscribe now.