Why Intel and Tencent: Quick Access to the Latest Cloud Technologies

The close collaboration between Intel and Tencent gives customers fast access to the latest technologies and optimizations. This includes access to 3rd Gen Intel® Xeon® Scalable processors with Intel® Software Guard Extensions (Intel® SGX) for hardware-accelerated AI and cryptography. Customers benefit from low TCO for cloud instances, robust security features and peace of mind, and fast results for cloud workloads.

Tencent Realizes Dramatic Performance Increases with Intel® Optimizations

Up to

1.34x

more single-instance memory capacity for Redis with Intel® Optane™ persistent memory1

Up to

4.23x

faster AI inference performance with 3rd Gen Intel® Xeon® Scalable processors vs. previous generation2

Up to

2M

IOPS for Tencent Cloud Block Storage with Intel® Optane™ technology3

Up to

1.34x

more single-instance memory capacity for Redis with Intel® Optane™ persistent memory1

Up to

4.23x

faster AI inference performance with 3rd Gen Intel® Xeon® Scalable processors vs. previous generation2

Up to

2M

IOPS for Tencent Cloud Block Storage with Intel® Optane™ technology3

Fast and Flexible Compute, Storage, and AI Workloads in the Tencent Cloud

Ultrafast Cloud Storage

Intel® Optane™ technology optimizes Tencent Cloud Block Storage (CBS) engines.

Read the story

Explore Tencent CBS

A More Affordable, Intelligent Cloud

Tencent boosts memory capacity and analysis speed with key Intel® architecture upgrades.

Read the story

Find Intel instances

Better AI Models for Banking

Tencent Research Institute deploys federated learning with Intel® SGX.

Read the case study

Enable Intel® SGX on Tencent

Show more Show less

Flexibility and Innovation in the Cloud

Intel® technology helps improve the quality of services and speed of delivery for Tencent Cloud.

Intel® Xeon® Scalable processors

Intel® Software Guard Extensions

Intel® Optane™ persistent memory

Intel in the cloud

Notices & Disclaimers

Performance varies by use, configuration, and other factors. Learn more on the Performance Index site.

Performance results are based on testing as of 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.

Product and Performance Information

1The data is quoted from Tencent Cloud’s Redis cloud service test based on 2nd Gen Intel® Xeon® Scalable processors. Test configuration: 2-socket (2S) Intel® Xeon® Platinum 8260 processor, 24 cores/48 threads, TurboBoost on, hyper-threading on; BIOS 1.018, 96 GB DRAM and 512 GB Intel® Optane™ persistent memory per socket, single 25GbE network adapter, Linux kernel 4.14.68-1-tlinux3-nvdimm-0005, Redis 4.10, raw data volume 11.27 GB, total 88 VM instances vs. 2-socket Intel® Xeon® Platinum 8260 processor, 24 cores/48 threads, TurboBoost on, hyper-threading on; BIOS 1.018, 384 GB DRAM physical memory per socket, single 25GbE network adapter, Linux kernel 4.14.68-1-tlinux3-nvdimm-0005, Redis 4.10, raw data volume 8.36 GB, total 88 VM instances launched.
2Test by Intel as of March 19, 2021. Baseline config: processor: 2S Intel® Xeon® Platinum 82XX processor (Cascade Lake), 24 cores/48 threads, TurboBoost on, hyper-threading on; memory: 12x 16 GB DDR4 2,933 MHz; storage: Intel® SSD; NIC: Intel® Ethernet Network Adapter X722; BIOS: SE5C620.86B.0D.01.0438.032620191658(ucode:0x5003003); OS: CentOS Linux 8.3; Linux kernel: 4.18.0-240.1.1.el8_3.x86_64; network model: PRN network; data set: Dummy and 300W_LP，12 instances/2 sockets; deep learning framework: Intel® Optimization for TensorFlow 2.4.0; oneDNN: 1.6.4; Intel® Neural Compressor: 1.1; model data format: FP32; test config 1: processor: 2S Intel® Xeon® Platinum 83XX processor (Ice Lake), 36 cores/72 threads, TurboBoost on, hyper-threading on; memory: 16*16 GB DDR4 3200; storage: Intel® SSD *1; NIC: Intel® Ethernet Controller X550T *1; BIOS: SE5C6200.86B.3020.P19.2103170131 (ucode: 0x8d05a260); OS: CentOS Linux 8.3; kernel: 4.18.0-240.1.1.el8_3.x86_64; network model: PRN network; data set: Dummy and 300W_LP，18 instances/2 sockets; deep learning framework: Intel® Optimization for TensorFlow 2.4.0; oneDNN: 1.6.4; Intel® Neural Compressor: 1.1; model data format: FP32; test config 2: processor: 2S Intel® Xeon® Platinum 83XX processor (Ice Lake), 36 cores /72 threads, TurboBoost on, hyper-threading on; memory: 16*16 GB DDR4 3200; storage: Intel® SSD *1; NIC: Intel® Ethernet Controller X550T *1; BIOS: SE5C6200.86B.3020.P19.2103170131 (ucode: 0x8d05a260); OS: CentOS Linux 8.3; kernel: 4.18.0-240.1.1.el8_3.x86_64; network model: PRN network; data set: Dummy and 300W_LP，18 instances/2 sockets; deep learning framework: Intel® Optimization for TensorFlow 2.4.0; oneDNN: 1.6.4; Intel® Neural Compressor: 1.1; model data format: INT8.
3Performance results based on testing as of April 27, 2021, and may not reflect all publicly available updates. Testing configuration: processor: 2S Intel® Xeon® Platinum 8255C; memory 1: 384 GB (32 GB x 12 @ 2,666 MHz); memory 2: Intel® Optane™ persistent memory 128 GB x 12; storage configuration 1: Intel® SSD 480 GB; storage configuration 2: Intel® NVMe SSD 3.84 TB x 12; network adapter: 100GbE x 2; OS kernel version: 5.4.110-1el7.elrepo.x86_64. Results may vary. For more complete information about performance and benchmark results, visit intel.com/benchmarks.