Intel has just found out information of their upcoming xeon scalable platform circle of relatives named it cascade lake. This new platform changed into demoed nowadays at sapphire conference wherein chip giant unveiled their new platform this is scheduled for launch in next yr.
Circle of relatives demoed – fifty nine% faster, utilises up to six tb of optane dimms, launching in 2018
Intel cascade lake xeon scalable platform starts in 2018 with optane dual inline memory module assist, and intel but has to launch their skylake-sp xeon circle of relatives but they may be already talking approximately subsequent refresh nearly 12 months earlier but it sounds regular of intel speaking approximately future merchandise so soon, however it can additionally be due to amd’s epyc platform. Based on zeppelin middle so first however it is able to additionally be resulting from amd’s epyc platform. Primarily based on zeppelin middle structure. Epyc information centre processors are aimed to give intel a difficult time also in servers market.
While intel hasn’t launched any public benchmarks of skylake-sp xeons, amd showcased new statistics on epyc’s overall performance against broadwell-ep (now not skylake-sp) components. It is going to be exciting to peer the competing systems pit up in opposition to each other in the important workloads. But amd additionally found out a protracted-time period strategic roadmap for servers at their economic analyst day. So after epyc (previously known as naples), amd plans on liberating rome with 7nm zen 2 cores in 2018 and milan with 7nm+ zen three.0 cores in 2019. This further puts intel in a difficult role, and here they’re, talking approximately their modern day and finest destiny tech so that it will compete in opposition to amd rome server chips in 2018. Those are codenamed intel cascade lake and could be changing the modern-day xeon scalable platform own family.
While Intel has not yet released public benchmarks of Skylake-SP Xeons, the AMD showcased new data on EPYC performance against Broadwell-EP (not Skylake-SP). You will find it interesting to see the computing platforms pit up against each other in the main workloads. But AMD also revealed a long-term strategic roadmap for servers at their Financial Analyst day.
Upcoming Intel and AMD Server Platform Comparison:
|Intel Xeon E5 Bronze / Silver||Intel Xeon E7 Gold / Platinum||AMD Naples Platform (2P)|
|Family Branding||Skylake-SP||Skylake-SP||AMD EPYC|
|PCH||Lewisburg PCH||Lewisburg PCH||SOC|
|Socket||Socket P (LGA 3647)||Socket P (LGA 3647)||SP3 LGA socket|
|Max Core Count||Up To 26||Up To 32||Up To 32|
|Max Thread Count||Up To 52||Up To 64||Up To 64|
|Max L3 Cache||35.75 MB L3||38.5 MB L3||64 MB L3|
|DDR4 Memory Support||6-Channel DDR4||6-Channel DDR4||8-Channel DDR4|
Intel Cascade Lake-SP is Skylake-SP Refresh Based on 14nm+ Node – Up To 6 TB Optane DIMM Support
Intel cascade lake own family of processors additionally to be known as cascade lake-sp may be refresh based on skylake-sp this new chip will maintain intact equal architecture however based totally on barely new 14nm+node on the way to virtually boom strength efficiency and clock speed.
However the key upgrade to cascade lake could be its disruptive reminiscence guide. Intel is going to feature guide for his or her next-gen 3-D xpoint primarily based optane dimms, allowing for the platform to assist up to 6x reminiscence potential. So we’re searching at up to 3 tb of memory for 4-socket and 6 tb of ability for eight-socket platforms.
In a performance demonstration with sap hana, intel used a xeon e7-8890 v4 (broadwell-en) processor in quad configuration against a cascade lake-sp processor in a quad config. The cascade lake-sp platform featured a performance increase of one.59x over the broadwell-e based totally machine which is a huge improvement. Intel expects to launch cascade lake with optane dimms later in 2018 so we can must wait a bit greater for exact facts at the platform.