Author: Jimmy Leon, Engineering Manager, Intel
Contributor: Nadezhda Ten, Performance Engineer, Intel
With the introduction of Intel® Xeon® 6+ processors (formerly codenamed Clearwater Forest), Intel delivers a 2-socket platform built to help customers scale smarter with up to 288 cores per chip. Optimized for hyperscale data centers, Intel Xeon 6+ processor boosts throughput, enhances energy efficiency, and provides targeted acceleration for networking and security, enabling customers to run more workloads with greater performance and cost efficiency.
Key Intel Xeon 6+ platform features include:
- 288 E-cores with up to 17% higher IPC than the previous generation.
- Intel 18A process and RibbonFET transistor technology provide improved performance and efficiency.
- Up to 576MBs combined last-level cache
- 12 DDR5 channels at 8000 MT/s, 96 PCIe 5.0 lanes, and 64 CXL 2.0 lanes
- New crypto acceleration for SHA-512, SM3, and SM4 algorithms
- Socket compatibility with Intel® Xeon® 6 6900P series platforms
Top-SKU Performance Leader: Intel Xeon 6990E+ vs. Intel Xeon 6980P
These platform advancements translate directly into measurable top-end throughput gains for hyperscale deployments. As shown in Figure 1 below, the new Intel Xeon 6990E+ delivers up to 22% more performance than the previous highest performing Intel Xeon 6980P processor.(1) Hyperscale deployments such as cloud, AI inference, and web services can, in turn, benefit from faster response times and higher capacity, improving service quality.
Figure 1: Intel Xeon 6990E+ delivers up to 23% higher top-end performance on average than the Intel Xeon 6980P.(1)
Performance Per Watt at Typical Data Center Utilization
Raw performance, however, is only part of the equation in hyperscale environments, where power efficiency and rack-level utilization are equally important. Data centers typically operate at approximately 40% utilization to preserve buffer capacity for unexpected traffic surges or failover events. At this standard operating point, the Intel Xeon 6990E+ provides up to 30% greater performance per watt (perf/W) than the Intel Xeon 6780E and Intel Xeon 6980P, improving operational efficiency while maintaining strong throughput.(2)
Figure 2: At a typical 40% data center utilization level, the new Intel Xeon 6990E+ delivers up to 30% better perf/W compared to prior Intel Xeon 6780E/6980P.(2)
Generational Performance Uplift: Intel Xeon 6990E+ vs. Intel Xeon 6780E
Beyond improving efficiency at typical utilization levels, Intel Xeon 6+ processors also deliver substantial value for customers planning a direct platform refresh. Compared with the prior generation Intel Xeon 6780E, the new Intel Xeon 6990E+ delivers up to 2.2x higher average performance across a broad set of workloads.(3)
Figure 3: In a comparison between the Intel Xeon 6990E+ vs. Intel Xeon 6780E, the new Intel Xeon 6+ processor performs up to 2.2x better.(3)
Targeted Acceleration for Security and Networking Workloads
In addition to broad compute and efficiency gains, the Intel Xeon 6900E+ processor family extends its value with targeted acceleration for networking and telecommunications workloads. The Intel Xeon 6+ CPU series introduces advanced SHA-512 and SM3 cryptographic instructions to help accelerate hash algorithms, along with SM4 instructions to improve block-cipher encryption and decryption efficiency. Compared to the Intel Xeon 6780E, the Intel Xeon 6990E+ achieves up to 15.2x greater average performance on SM3 and SM4, and up to 5.2x higher average performance on SHA-512.(4)
These enhancements enable companies to process security workloads faster and more efficiently, helping reduce latency in critical applications while lowering the compute overhead associated with encryption. This helps translate into stronger data protection without sacrificing performance. Faster hashing improves operations such as password validation, file integrity checks, and digital signatures, enabling more responsive applications and more scalable identity and security services. At the same time, accelerated block cipher performance enhances the efficiency of encryption used in VPNs, HTTPS traffic, messaging platforms, and financial transactions, enabling businesses to secure more data with lower infrastructure cost.
Figure 4: Intel Xeon 6990E+ delivers up to 15.2x higher average performance on SM3 and SM4, and up to 5.2x higher average performance on SHA-512 vs. the Intel Xeon 6780.(4)
Additional Built-In Intel Xeon 6+ Platform Cryptographic Benefits
Beyond improving on-core efficiency, the Intel Xeon 6+ platform integrates Intel® QuickAssist Technology (Intel® QAT) hardware accelerators to offload cryptographic and compression workloads, reducing CPU demand. For example, Intel QAT accelerates NGINX TLS handshakes for secure HTTPS connections, freeing CPU resources for application workloads.
The Intel Xeon 6990E+ processor includes four Intel QAT devices, each capable of up to 16.9K connections per second (CPS), for a total of 67.8K CPS.(5) These accelerators provide higher throughput for secure connections, improved application performance, and more consistent responsiveness. By using all four Intel QAT hardware accelerators to offload cryptographic processing, customers can achieve comparable performance with roughly half the cores—helping reduce infrastructure footprint, lower power consumption, and optimize total cost of ownership.(5)
Figure 5: Four Intel QAT devices deliver near-equivalent TLS 1.3 handshake throughput while reducing CPU core usage by approximately 50%, significantly improving efficiency and freeing compute resources.(5)
These results illustrate a meaningful architectural advantage. Cryptographic workloads can be offloaded to Intel QAT while maintaining essentially equivalent TLS handshake throughput, indicating a negligible performance trade-off. At the same time, CPU core utilization drops significantly, from approximately 66 cores to around 32 cores. That improvement can create additional headroom for application processing, service consolidation, or higher workload density on the same server. In environments where TLS handshakes are a primary scaling constraint—such as edge proxies, API gateways, and microservices ingress layers—this reduction in CPU overhead enables each system to sustain higher connection density without becoming CPU-bound.
Intel Xeon 6+ Platform: High Performance for Data Centers
All together, these results show that Intel Xeon 6+ processors strengthen Intel’s position in hyperscale infrastructure by combining higher top-end performance, better performance per watt at typical operating levels, strong generational uplift, and targeted acceleration for security-focused workloads. For operators balancing throughput, efficiency, and workload specialization, Intel Xeon 6+ processors offer a compelling platform for scale-out services, dense deployments, and modern networked applications.
Learn more about the Intel Xeon 6+ processor and how it is designed to support next-generation hyperscale, networking, and security workloads.
Endnotes:
- See [9G26] at intel.com/processorclaims: Intel® Xeon® 6+. Results may vary.
- See [9T26] at intel.com/processorclaims: Intel® Xeon® 6+. Results may vary.
- See [9G20] at intel.com/processorclaims: Intel® Xeon® 6+. Results may vary.
- See [9W32] at intel.com/processorclaims: Intel® Xeon® 6+. Results may vary.
- See [9W37] at intel.com/processorclaims: Intel® Xeon® 6+. Results may vary.
Notices and Disclaimers
Performance varies by use, configuration, and other factors. Learn more on the Performance Index site.
Performance results are based on testing as of the dates shown in configurations and may not reflect all publicly available updates. See backup for configuration details. No product or component can be absolutely secure.
Your costs and results may vary.
Intel technologies may require enabled hardware, software, or service activation.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.