Intel 4th Gen Xeon: Difference between revisions

From HPCWIKI
Jump to navigation Jump to search
No edit summary
(리터럴 ** 볼드 마커 제거 (MediaWiki는 ''' 볼드, **는 리터럴))
 
(13 intermediate revisions by 2 users not shown)
Line 1: Line 1:
4th Gen Intel® Xeon® Scalable Processors list [https://www.colfax-intl.com/downloads/4th-gen-xeon-acclerators-eguide-with-4th-gen-benchmarks.pdf]
= Intel 4th Gen Xeon =
 
== Overview ==
 
Intel 4th Gen Xeon Scalable processors(Sapphire Rapids)는 2023년 1월 출시된 데이터센터용 프로세서로, AI, 클라우드 컴퓨팅, 네트워킹, 에지 컴퓨팅, 슈퍼컴퓨팅을 위해 설계되었습니다.
 
=== Summary ===
 
* 무엇인가? — Intel 4th Gen Xeon Scalable processors (Sapphire Rapids)
* 왜 필요한가? — 데이터센터 성능, 효율성, 보안, 기능 향상
* 언제 사용하는가? — AI, 클라우드, 네트워킹, 에지, 슈퍼컴퓨팅
 
== Purpose ==
 
이 문서가 존재하는 이유
 
* Goal: Intel 4th Gen Xeon Scalable processors의 제품군, 아키텍처, 성능 특징을 한 곳에서 제공
* Scope: Xeon 모델 카테고리, PCIe 5.0, DDR5, CXL, XCC/MCC/HBM, 내장 가속기
* Non-goals: 이전 세대 Xeon 상세 비교, [[AMD EPYC]] 상세 비교는 별도 문서
 
== Key Concepts ==
 
{| class="wikitable"
{| class="wikitable"
!SKU
! Concept
!CORES
! Description
!BASE
! Related
!ALL CORE TURBO (GHz)
!Max TURBO (GHz)
!CACHE (MB)
!TDP (Watts)
!Maximum Scalability
!DDR5 Memory Speed
!UPI Links Enabled
!Default DSA Devices
!Default IAA Devices
!Default QAT Devices
!Default DLB Devices
!Intel SGX Enclave Capacity (Per Processor)
!Long Life Use
!Intel® On Demand Capable
|-
! colspan="17" |2S PERFORMANCE GENERAL PURPOSE
|-
|8480+
|56
|2.0
|3.0
|3.8
|105
|350
|2S
|4800
|4
|1
|1
|1
|1
|512GB
|
|✓
|-
|8470
|52
|2.0
|3.0
|3.8
|105
|350
|2S
|4800
|4
|1
|0
|0
|0
|512GB
|✓
|✓
|-
|8468
|48
|2.1
|3.1
|3.8
|105
|350
|2S
|4800
|4
|1
|0
|0
|0
|512GB
|
|✓
|-
|8460Y+
|40
|2.0
|2.8
|3.7
|105
|300
|2S
|4800
|4
|1
|1
|1
|1
|128GB
|
|✓
|-
|8462Y+
|32
|2.8
|3.6
|4.1
|60
|300
|2S
|4800
|3
|1
|1
|1
|1
|128GB
|✓
|✓
|-
|6448Y
|32
|2.1
|3.0
|4.1
|60
|225
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|6442Y
|24
|2.6
|3.3
|4.0
|60
|225
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|
|✓
|-
|6444Y
|16
|3.6
|4.0
|4.1
|45
|270
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|
|✓
|-
|6426Y
|16
|2.5
|3.3
|4.1
|37.5
|185
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|6434
|8
|3.7
|4.1
|4.1
|22.5
|195
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|
|✓
|-
|5415+
|8
|2.9
|3.6
|4.1
|22.5
|150
|2S
|4400
|3
|1
|1
|1
|1
|128GB
|✓
|✓
|-
! colspan="17" |2S MAINLINE GENERAL PURPOSE
|-
|8452Y
|36
|2.0
|2.8
|3.2
|67.5
|300
|2S
|4800
|4
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|6438Y+
|32
|2.0
|2.8
|4.0
|60
|205
|2S
|4800
|3
|1
|1
|1
|1
|128GB
|
|✓
|-
|6430
|32
|2.1
|2.6
|3.4
|60
|270
|2S
|4400
|3
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|5420+
|28
|2.0
|2.7
|4.1
|52.5
|205
|2S
|4400
|3
|1
|1
|1
|1
|128GB
|✓
|✓
|-
|5418Y
|24
|2.0
|2.8
|3.8
|45
|185
|2S
|4400
|3
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|4416+
|20
|2.0
|2.9
|3.9
|37.5
|165
|2S
|4000
|2
|1
|1
|1
|1
|64GB
|✓
|✓
|-
|4410Y
|12
|2.0
|2.8
|3.9
|30
|150
|2S
|4000
|2
|1
|0
|0
|0
|64GB
|✓
|✓
|-
! colspan="17" |LIQUID COOLED GENERAL PURPOSE(-Q)
|-
|8470Q
|52
|2.1
|3.2
|3.8
|105
|350
|2S
|4800
|4
|1
|0
|0
|0
|512GB
|
|✓
|-
|6458Q
|32
|3.1
|4.0
|4.0
|60
|350
|2S
|4800
|3
|1
|0
|0
|0
|128GB
|
|✓
|-
! colspan="17" |SINGLE SOCKET GENERAL PURPOSE(-U)
|-
|6414U
|32
|2.0
|2.6
|3.4
|60
|350
|1S
|4800
|0
|1
|0
|0
|0
|128GB
|
|✓
|-
|5412U
|24
|2.1
|2.9
|3.9
|45
|185
|1S
|4400
|0
|1
|0
|0
|0
|128GB
|
|✓
|-
|3408U
|8
|1.8
|1.9
|1.9
|22.5
|125
|1S
|4000
|0
|1
|0
|0
|0
|64GB
|✓
|
|-
! colspan="17" |LONG - LIFE USE (IOT) GENERAL PURPOSE (-T)
|-
|4410T
|10
|2.7
|3.4
|4.0
|26.25
|150
|2S
|4000
|2
|1
|0
|0
|0
|64GB
|✓
|✓
|-
! colspan="17" |IMDB/ANALYTICS/VIRTUALIZATION OPTIMIZED (-H) – SOCKET SCALABLE
|-
|8490H
|60
|1.9
|2.9
|3.5
|112.5
|350
|8S
|4800
|4
|4
|4
|4
|4
|512GB
|
|
|-
|8468H
|48
|2.1
|3.0
|3.8
|105
|330
|8S
|4800
|4
|4
|4
|4
|4
|512GB
|
|
|-
|8460H
|40
|2.2
|3.1
|3.8
|105
|330
|8S
|4800
|4
|4
|4
|0
|0
|512GB
|
|
|-
|8454H
|32
|2.1
|2.7
|3.4
|82.5
|270
|8S
|4800
|4
|4
|4
|4
|4
|512GB
|
|
|-
|8450H
|28
|2.0
|2.6
|3.5
|75
|250
|8S
|4800
|4
|4
|4
|0
|0
|512GB
|
|
|-
|8444H
|16
|2.9
|3.2
|4.0
|45
|270
|8S
|4800
|4
|4
|4
|0
|0
|512GB
|
|
|-
|6448H
|32
|2.4
|3.2
|4.1
|60
|250
|4S
|4800
|3
|1
|1
|2
|2
|512GB
|
|
|-
|6418H
|24
|2.1
|2.9
|4.0
|60
|185
|4S
|4800
|3
|1
|1
|0
|0
|512GB
|✓
|
|-
|6416H
|18
|2.2
|2.9
|4.2
|45
|165
|4S
|4800
|3
|1
|1
|0
|0
|512GB
|
|
|-
|6434H
|8
|3.7
|4.1
|4.1
|22.5
|195
|4S
|4800
|3
|1
|1
|0
|0
|512GB
|
|
|-
! colspan="17" |5G / NETWORKING OPTIMIZED (-N)
|-
|8470N
|52
|1.7
|2.7
|3.6
|97.5
|300
|2S
|4800
|4
|4
|0
|4
|4
|128GB
|✓
|✓
|-
|8471N
|52
|1.8
|2.8
|3.6
|97.5
|300
|1S
|4800
|0
|4
|0
|4
|4
|128GB
|✓
|✓
|-
|6438N
|32
|2.0
|2.7
|3.6
|60
|205
|2S
|4800
|3
|1
|0
|2
|2
|128GB
|✓
|✓
|-
|6428N
|32
|1.8
|2.5
|3.8
|60
|185
|2S
|4000
|3
|1
|0
|2
|2
|128GB
|✓
|✓
|-
|6421N
|32
|1.8
|2.6
|3.6
|60
|185
|1S
|4400
|0
|1
|0
|0
|0
|128GB
|✓
|✓
|-
|5418N
|24
|1.8
|2.6
|3.8
|45
|165
|2S
|4000
|3
|1
|0
|2
|2
|128GB
|✓
|✓
|-
|5411N
|24
|1.9
|2.8
|3.9
|45
|165
|1S
|4400
|0
|1
|0
|2
|2
|128GB
|✓
|✓
|-
! colspan="17" |CLOUD OPTIMIZED IaaS (-P) / SaaS(-V) / Media(-M)
|-
|8468V
|48
|2.4
|2.9
|3.8
|97.5
|330
|2S
|4800
|3
|1
|1
|1
|1
|128GB
|
|✓
|-
|8458P
|44
|2.7
|3.2
|3.8
|82.5
|350
|2S
|4800
|3
|1
|1
|1
|1
|512GB
|
|✓
|-
|8461V
|48
|2.2
|2.8
|3.7
|97.5
|300
|1S
|4800
|0
|1
|1
|1
|1
|128GB
|
|✓
|-
|6438M
|32
|2.2
|2.8
|3.9
|60
|205
|2S
|4800
|3
|1
|1
|0
|0
|128GB
|
|✓
|-
! colspan="17" |STORAGE & HYPERCONVERGED INFRASTRUCTURE (HCI) OPTIMIZED (-S)
|-
|6454S
|32
|2.2
|2.8
|3.4
|60
|270
|2S
|4800
|4
|4
|0
|4
|4
|128GB
|
|✓
|-
|5416S
|16
|2.0
|2.8
|4.0
|30
|150
|2S
|4400
|3
|1
|0
|2
|2
|128GB
|
|✓
|-
|-
! colspan="17" |HPC OPTMIZED (Intel® Xeon® CPU MAX Series)
| Sapphire Rapids
| 4th Gen Xeon의 코드명
| [[Intel 4th Gen Xeon]]
|-
|-
|9480
| PCIe 5.0
|56
| 최대 80개 레인, 대역폭 2배 향상
|1.9
| [[PCIe]]
|2.6
|3.5
|112.5
|350
|2S
|4800
|4
|4
|0
|0
|0
|128GB
|
|
|-
|-
|9470
| DDR5
|52
| DDR4 대비 1.5x 대역폭 향상
|2
| [[Intel 4th Gen Xeon]]
|2.7
|3.5
|105
|350
|2S
|4800
|4
|4
|0
|0
|0
|512GB
|
|
|-
|-
|9468
| CXL
|48
| CPU-가속기 간 메모리 공간 통합
|2.1
| [[Intel 4th Gen Xeon]]
|2.6
|3.5
|105
|350
|2S
|4800
|4
|4
|0
|0
|0
|128GB
|
|
|-
|-
|9460
| XCC
|40
| 4개 컴퓨트 타일 통합 (최대 60코어)
|2.2
| [[Intel 4th Gen Xeon]]
|2.7
|3.5
|97.5
|350
|2S
|4800
|3
|4
|0
|0
|0
|128GB
|
|
|-
|-
|9462
| MCC
|32
| 단일 다이 (최대 32코어)
|2.7
| [[Intel 4th Gen Xeon]]
|3.1
|3.5
|75
|350
|2S
|4800
|3
|4
|0
|0
|0
|128GB
|
|
|-
|-
! colspan="17" |Y Supports Intel Speed Select Technology – Performance Profile 2.0 (Intel SST-PP)
| HBM
| 64GB HBM2e 메모리 내장
| [[Intel 4th Gen Xeon]]
|}
|}


=== built-in accelerators<ref>https://www.colfax-intl.com/downloads/4th-gen-xeon-acclerators-eguide-with-4th-gen-benchmarks.pdf</ref> ===
== Architecture ==
integrated accelerators support today’s most demanding workloads spanning AI, security, HPC, networking, analytics and storage
{| class="wikitable"
|+
!Name
!Purpose
!Notes
!
|-
| rowspan="2" |Intel® AI Engine
|Intel® Advanced Matrix Extensions (Intel® AMX)
|Intel AMX is Intel’s next-generation advancement for deep-learning training and
inference on 4th Gen Intel Xeon Scalable processors. Ideal for workloads like natural


language processing, recommendation systems and image recognition, this new
Intel 4th Gen Xeon은 Tile 기반 Chiplet 설계와 XCC/MCC/HBM 패키징을 제공합니다:


feature extends the built-in AI acceleration capabilities of previous Intel Xeon Scalable
<syntaxhighlight lang="mermaid">
graph TD
    A[Sapphire Rapids] --> B[XCC - 4 Tiles]
    A --> C[MCC - Monolithic]
    A --> D[HBM - High Bandwidth]
    B --> E[PCIe 5.0]
    B --> F[DDR5 Memory]
    B --> G[CXL]
    B --> H[Integrated Accelerators]
</syntaxhighlight>


processors while also offering significant performance gains
== Workflow ==
|
|-
|Intel® Advanced Vector Extensions 512 (Intel® AVX-512
|A continuing feature of Intel Xeon Scalable processors, Intel AVX-512 is a general-
purpose performance-enhancing accelerator with a wide range of uses. When it


comes to AI, Intel AVX-512 can accelerate machine learning workloads for training
{| class="wikitable"
 
! Stage
and inferencing. 4th Gen Intel Xeon Scalable processors with Intel AVX-512 are also
! Input
 
! Output
designed to speed up data preprocessin
|
|-
|-
| rowspan="3" |Intel® Security Engines
| Selection
|Intel® Software Guard Extensions (Intel® SGX)
| [[Workload]] requirements
 
| Xeon model selection
 
|Intel SGX is the most researched, updated and deployed confidential computing
technology in data centers on the market today. This continuing feature of Intel Xeon
 
Scalable processors provides the foundation for confidential computing solutions
 
across edge and multi-cloud infrastructures.
 
Intel SGX offers a hardware-based security solution that is designed to prohibit
 
access to protected data in use via unique application-isolation technology. By
 
helping protect selected code and data from inspection or modification, Intel SGX
 
allows developers to run sensitive data operations inside enclaves to help increase
 
application security and protect data confidentiality
|
|-
|-
|Intel® Trust Domain Extensions (Intel® TDX)
| Deployment
|Intel TDX is a new capability available through select cloud providers in 2023 that
| Xeon CPU + [[Motherboard]]
offers increased confidentiality at the virtual machine (VM) level, enhancing privacy
| System boot
 
and control over your data. Within an Intel TDX confidential VM, the guest OS and
 
VM applications are isolated from access by the cloud host, hypervisor and other
 
VMs on the platform.
|
|-
|-
|Intel® Crypto Acceleration
| [[Optimization]]
|Intel Crypto Acceleration uses single instruction, multiple data (SIMD) techniques to
| [[BIOS]]/sysctl settings
process more encryption operations in every clock cycle. This can help increase the
| Optimized performance
|}


total throughput of applications that require strong data encryption, with minimal
== Detailed Explanation ==


impact on performance and user experience
=== Product Categories ===
|
|-
| rowspan="3" |Intel® HPC Engines
|Intel® Advanced Vector Extensions 512 (Intel® AVX-512)


4th Gen Xeon은 6개 카테고리로 제공됩니다:


|With ultrawide 512-bit vector operations capabilities, Intel AVX-512 is especially
* Max 9400 — HPC 최적화 (Intel Xeon CPU Max Series)
suited to handle the most demanding computational tasks commonly encountered in
* Platinum 8000 — 고급 데이터 분석, AI, 하이브리드 클라우드
* Gold 6000/Gold 5000 — 데이터센터 및 멀티클라우드 워크로드
* Silver 4000 — 엔트리 레벨 데이터센터 컴퓨팅, 네트워킹, 스토리지
* Bronze 3000 — 기본 컴퓨팅
* Max Series — HPC 최적화


HPC applications. It’s used by organizations across educational, financial, enterprise,
=== Performance Features ===


engineering and medical industries. By enabling users to run complex workloads
* Optimized Power Mode — 최대 20% 소켓 전력 절약, 5% 미만 성능 영향
* Air/Liquid Cooling — 데이터센터 에너지 소비 감소


on existing hardware, Intel AVX-512 accelerates performance for tasks like financial
=== Use Case Categories ===


analytics, 3D modeling and scientific simulations
* Performance general purpose
|
* Mainline general purpose
|-
* [[Liquid cooling]] general purpose
|Intel® Advanced Matrix Extensions (Intel® AMX)
* Single socket general purpose (Q Series)
|In addition to accelerating AI workloads, Intel AMX is also designed to deliver
* Long-life Use (IoT) general purpose (T Series)
performance gains across popular HPC workloads. This new built-in accelerator
* IMDB/Analytics/virtualization optimized (H Series)
* 5G/Networking optimized (N Series)
* Cloud-optimized IaaS (P, V, M Series)
* Storage and HCI optimized (S series)
* HPC optimized (Intel Xeon CPU Max Series)


transforms large matrix math calculations into a single operation and uses a two-
=== PCI Express Gen5 (PCIe 5.0) ===


dimensional register file to store large chunks of data
최대 80개 레인의 PCIe 5.0 지원. PCIe 4.0 대비 I/O 대역폭 2배 향상, 후방 호환성 유지.
|
|-
|Intel® Data Streaming Accelerator (Intel® DSA
|Intel DSA is a new feature designed to optimize and speed up streaming data
movement and transformation operations common in networking, data-


processing-intensive applications and high-performance storage. Intel DSA
=== DDR5 ===


accelerates HPC workloads by offloading the most common data movement tasks
DDR4 대비 최대 1.5x 대역폭 향상. 4th Gen Xeon은 최대 4,800 MT/s (1 DPC) 또는 4,400 MT/s (2 DPC) 지원.


that cause CPU overhead in data-center-scale deployments
=== CXL ===
|
|-
| rowspan="3" |Intel® Network Engines
|Intel® QuickAssist Technology (Intel® QAT)


CXL 1.1 지원. CPU와 가속기 간 통합 일관성 메모리 공간 생성. 데이터센터 서버 아키텍처 혁신.


|Intel QAT boosts performance to meet the demands of today’s networking workloads,
=== XCC, MCC, HBM Variants ===
helping systems serve more clients. It can deliver significant workload acceleration for


cryptography, including symmetric and asymmetric encryption and decryption.
* XCC (E1A) — 4개 컴퓨트 타일 통합, 각 타일 최대 15코어 (최대 60코어)
* MCC (E1B) — 단일 다이, TDP 125W-350W, 최대 32코어
* HBM (E1C) — 16GB HBM2e 스택 (총 64GB), 고속 메모리 풀


Intel QAT using RSA4K can increase client density on an open-source NGINX web
=== Built-in Accelerators ===


server compared to software running on CPU cores without acceleration
* AI Engine — Intel AMX, Intel AVX-512
|
* Security Engines — Intel SGX, Intel TDX, Intel Crypto Acceleration
|-
* HPC Engines — Intel AVX-512, Intel AMX, Intel DSA
|Intel® vRAN Boost
* [[Network]] Engines — Intel QAT, Intel vRAN Boost, Intel DLB
|Intel vRAN Boost is a new feature that eliminates the need for an external accelerator
* Analytics Engines — Intel IAA, Intel DSA, Intel QAT
card by integrating vRAN acceleration directly into the Intel 4th Gen Xeon Scalable
* Storage Engines — Intel DSA, Intel QAT, Intel VMD


processor. By reducing vRAN component requirements, it reduces overall vRAN
== Configuration ==


solution complexity and provides power savings for customers
<syntaxhighlight lang="bash">
|
# CPU 정보 확인
|-
lscpu | grep -i intel
|Intel® Dynamic Load Balancer (Intel® DLB
cat /proc/cpuinfo | grep -i model
|Intel DLB is a new feature that enables the efficient distribution of network
processing across multiple CPU cores. It also restores the order of networking data


packets processed simultaneously on CPU cores.
# PCIe 정보 확인
lspci | grep -i pci


With Intel DLB, customers can gain higher performance on packet forwarding
# 메모리 정보 확인
free -h
</syntaxhighlight>


compared to software queue management on cores without acceleration.
== Examples ==


Additionally, applications can achieve higher packet processing performance than
=== Example 1: Xeon 모델 선택 ===


the previous generation
{| class="wikitable"
|
! Use Case
! Recommended Series
! Example Model
|-
|-
| rowspan="3" |Intel® Analytics Engines
| AI/Analytics
|Intel® In-Memory Analytics Accelerator (Intel® IAA)
| Platinum 8000
 
| 8480+
 
|Intel IAA is designed to accelerate database and analytics performance. By increasing
query throughput and decreasing the memory footprint for in-memory databases
 
and advanced analytics workloads, this new feature can provide faster data
 
movement and improve CPU core utilization by reducing dependency on CPU cores.
 
Intel IAA is ideal for in-memory databases, open-source databases and data stores
 
like RocksDB, Redis, Cassandra and MySQL. Customers using 4th Gen Intel Xeon
 
Platinum 8490H with Intel IAA will gain up to 3x higher RocksDB performance
 
compared to the previous generation
|
|-
|-
|Intel® Data Streaming Accelerator (Intel® DSA)
| Cloud IaaS
|Intel DSA is a new feature designed to optimize and speed up streaming data
| Gold 6000
movement and transformation operations common in data-intensive applications,
| 6438Y+
 
driving better business outcomes. By offloading tasks like data movement, data
 
copying and error checking, Intel DSA enables the CPU to focus on business-
 
critical database functions or other analytics workloads. This reduces query
 
latencies and increases throughput, delivering faster data processing
|
|-
|-
|Intel® QuickAssist Technology (Intel® QAT)
| Networking
|Intel QAT can be used to accelerate database backups. When QAT was enabled
| Silver 4000
as a new feature in SQL Server 2022, it delivered more efficient and higher
| 6428N
 
performance applications.
 
With Intel QAT, SQL Server customers achieve up to 2.3x faster backup operations
 
and a 6% reduction in backup storage capacity
|
|-
|-
| rowspan="3" |Intel® Storage Engines
| HPC
|Intel® Data Streaming Accelerator (Intel® DSA)
| Max Series
| 9480
|}


== Best Practices ==


|Intel DSA works on the CPU — between DRAM, caches and processor cores — and
* workload에 맞는 Xeon 시리즈 선택
extends across I/Os to attached memory and storage, plus networked resources.
* PCIe 5.0 장치 활용
* DDR5 메모리 최적화
* CXL 지원 장치 고려
* 내장 가속기 활성화 (AMX, QAT, DSA 등)
* Optimized Power Mode 활용


As Intel’s next-generation direct memory access (DMA) engine, it speeds transfers
== Limitations ==


between volatile and persistent memory and supports virtualized memory and I/Os.
* 특정 모델은 2S/4S/8S 제한
* HBM 모델은 제한적 가용성
* CXL 장치 호환성 확인 필요


Customers using 4th Gen Intel Xeon Platinum 8490H with Intel DSA will gain up to
== References ==


1.6x higher IOPS and up to 37% latency reduction for large packet sequential read
* https://www.intel.com/content/www/us/en/products/docs/processors/xeon/scalable/4th-gen-overview.html
* https://www.colfax-intl.com/downloads/4th-gen-xeon-acclerators-eguide-with-4th-gen-benchmarks.pdf


compared to the previous generation
== Related Pages ==
|
|-
|Intel® QuickAssist Technology (Intel® QAT)
|Intel QAT increases performance of storage workloads and applications by
accelerating cryptography and data compression/decompression. Customers


using 4th Gen Intel Xeon Platinum 8490H with Intel QAT will gain up to 95%
* [[AMD EPYC 9004 series vs Intel 4th gen XEON]]
* [[Intel Xeon 6 processor]]
* [[PCIe]]
* [[DDR5]]
* [[CXL]]


fewer cores and 2x higher level 1 compression throughput compared to the
[[Category:Hardware]]
== Knowledge Graph ==


previous generation
Related
|
|-
|Intel® Volume Management Device (Intel® VMD)
|Intel VMD is a legacy feature that enables direct control and management of
NVMe SSDs from the PCIe bus without the need for additional hardware adapters.


It allows for a smoother and lower-cost transition to fast NVMe storage while
→ [[AMD CPUs]]
→ [[Server]]
→ [[AMD EPYC 9004 series vs Intel 4th gen XEON]]


limiting the downtime of critical infrastructure. With benefits like bootable RAID,
[[Category:Reference]]
 
robust surprise hot-plug and blink status LED, Intel VMD increases serviceability
 
and enables you to deploy next-generation storage with confidence
|
|}

Latest revision as of 13:53, 2 October 2026

Intel 4th Gen Xeon

Overview

Intel 4th Gen Xeon Scalable processors(Sapphire Rapids)는 2023년 1월 출시된 데이터센터용 프로세서로, AI, 클라우드 컴퓨팅, 네트워킹, 에지 컴퓨팅, 슈퍼컴퓨팅을 위해 설계되었습니다.

Summary

  • 무엇인가? — Intel 4th Gen Xeon Scalable processors (Sapphire Rapids)
  • 왜 필요한가? — 데이터센터 성능, 효율성, 보안, 기능 향상
  • 언제 사용하는가? — AI, 클라우드, 네트워킹, 에지, 슈퍼컴퓨팅

Purpose

이 문서가 존재하는 이유

  • Goal: Intel 4th Gen Xeon Scalable processors의 제품군, 아키텍처, 성능 특징을 한 곳에서 제공
  • Scope: Xeon 모델 카테고리, PCIe 5.0, DDR5, CXL, XCC/MCC/HBM, 내장 가속기
  • Non-goals: 이전 세대 Xeon 상세 비교, AMD EPYC 상세 비교는 별도 문서

Key Concepts

Concept Description Related
Sapphire Rapids 4th Gen Xeon의 코드명 Intel 4th Gen Xeon
PCIe 5.0 최대 80개 레인, 대역폭 2배 향상 PCIe
DDR5 DDR4 대비 1.5x 대역폭 향상 Intel 4th Gen Xeon
CXL CPU-가속기 간 메모리 공간 통합 Intel 4th Gen Xeon
XCC 4개 컴퓨트 타일 통합 (최대 60코어) Intel 4th Gen Xeon
MCC 단일 다이 (최대 32코어) Intel 4th Gen Xeon
HBM 64GB HBM2e 메모리 내장 Intel 4th Gen Xeon

Architecture

Intel 4th Gen Xeon은 Tile 기반 Chiplet 설계와 XCC/MCC/HBM 패키징을 제공합니다: 

graph TD
    A[Sapphire Rapids] --> B[XCC - 4 Tiles]
    A --> C[MCC - Monolithic]
    A --> D[HBM - High Bandwidth]
    B --> E[PCIe 5.0]
    B --> F[DDR5 Memory]
    B --> G[CXL]
    B --> H[Integrated Accelerators]

Workflow

Stage Input Output
Selection Workload requirements Xeon model selection
Deployment Xeon CPU + Motherboard System boot
Optimization BIOS/sysctl settings Optimized performance

Detailed Explanation

Product Categories

4th Gen Xeon은 6개 카테고리로 제공됩니다:

  • Max 9400 — HPC 최적화 (Intel Xeon CPU Max Series)
  • Platinum 8000 — 고급 데이터 분석, AI, 하이브리드 클라우드
  • Gold 6000/Gold 5000 — 데이터센터 및 멀티클라우드 워크로드
  • Silver 4000 — 엔트리 레벨 데이터센터 컴퓨팅, 네트워킹, 스토리지
  • Bronze 3000 — 기본 컴퓨팅
  • Max Series — HPC 최적화

Performance Features

  • Optimized Power Mode — 최대 20% 소켓 전력 절약, 5% 미만 성능 영향
  • Air/Liquid Cooling — 데이터센터 에너지 소비 감소

Use Case Categories

  • Performance general purpose
  • Mainline general purpose
  • Liquid cooling general purpose
  • Single socket general purpose (Q Series)
  • Long-life Use (IoT) general purpose (T Series)
  • IMDB/Analytics/virtualization optimized (H Series)
  • 5G/Networking optimized (N Series)
  • Cloud-optimized IaaS (P, V, M Series)
  • Storage and HCI optimized (S series)
  • HPC optimized (Intel Xeon CPU Max Series)

PCI Express Gen5 (PCIe 5.0)

최대 80개 레인의 PCIe 5.0 지원. PCIe 4.0 대비 I/O 대역폭 2배 향상, 후방 호환성 유지.

DDR5

DDR4 대비 최대 1.5x 대역폭 향상. 4th Gen Xeon은 최대 4,800 MT/s (1 DPC) 또는 4,400 MT/s (2 DPC) 지원.

CXL

CXL 1.1 지원. CPU와 가속기 간 통합 일관성 메모리 공간 생성. 데이터센터 서버 아키텍처 혁신.

XCC, MCC, HBM Variants

  • XCC (E1A) — 4개 컴퓨트 타일 통합, 각 타일 최대 15코어 (최대 60코어)
  • MCC (E1B) — 단일 다이, TDP 125W-350W, 최대 32코어
  • HBM (E1C) — 16GB HBM2e 스택 (총 64GB), 고속 메모리 풀

Built-in Accelerators

  • AI Engine — Intel AMX, Intel AVX-512
  • Security Engines — Intel SGX, Intel TDX, Intel Crypto Acceleration
  • HPC Engines — Intel AVX-512, Intel AMX, Intel DSA
  • Network Engines — Intel QAT, Intel vRAN Boost, Intel DLB
  • Analytics Engines — Intel IAA, Intel DSA, Intel QAT
  • Storage Engines — Intel DSA, Intel QAT, Intel VMD

Configuration

# CPU 정보 확인
lscpu | grep -i intel
cat /proc/cpuinfo | grep -i model

# PCIe 정보 확인
lspci | grep -i pci

# 메모리 정보 확인
free -h

Examples

Example 1: Xeon 모델 선택

Use Case Recommended Series Example Model
AI/Analytics Platinum 8000 8480+
Cloud IaaS Gold 6000 6438Y+
Networking Silver 4000 6428N
HPC Max Series 9480

Best Practices

  • workload에 맞는 Xeon 시리즈 선택
  • PCIe 5.0 장치 활용
  • DDR5 메모리 최적화
  • CXL 지원 장치 고려
  • 내장 가속기 활성화 (AMX, QAT, DSA 등)
  • Optimized Power Mode 활용

Limitations

  • 특정 모델은 2S/4S/8S 제한
  • HBM 모델은 제한적 가용성
  • CXL 장치 호환성 확인 필요

References

Related Pages

Knowledge Graph

Related

→ AMD CPUs → Server → AMD EPYC 9004 series vs Intel 4th gen XEON

Retrieved from "http://wiki.hpcmate.com/index.php?title=Intel_4th_Gen_Xeon&oldid=3971"