Whether you’re a silicon chip manufacturer like Intel, or interested in other high-performance computing (HPC) environments like the life sciences, oil and gas, aeronautical, and automobile industries, faster throughput is an important metric when evaluating a computing solution. Intel IT recently conducted tests of a dual-socket server based on the latest Intel® Xeon® Platinum 8168 processor (2.7 GHz, 24 cores) running single-threaded and multi-threaded electronic design automation (EDA) applications. This latest-generation processor yielded the following results:
- Up to 86x faster than a server based on a 64-bit Intel® Xeon® processor (3.6 GHz) with a single core
- Up to 24x faster than a server based on the Intel® Xeon® processor 5160 (3.0 GHz) with two cores
- Up to 1.37x faster than a previous-generation Intel® Xeon® processor E5-2699 v4 (2.2 GHz) with 22 cores
Faster Throughput and Cost Savings
Using a weighted performance measure of end-to-end EDA applications based on Intel® silicon design tests, we found that the effective refresh ratio to replace servers based on previous-generation Intel® Xeon® processors with servers based on the Intel® Xeon® processor Scalable family is approximately 3.29:1. By consolidating servers, we can realize operational benefits such as cost avoidance of data center construction and reduced power consumption.
We also have found that, by adhering to a regular refresh cycle in our data centers, we avoid extended warranty expenses as well as fewer system failures.
Building the Infrastructure of the Future
While deploying the Intel® Xeon® processor Scalable family can result in faster throughput, and it provides direct benefits in terms of price/performance and power consumption, it also helps the enterprise to position its infrastructure to meet future computing needs.
The new servers offer new technologies that support hybrid cloud infrastructures, high-performance workloads, and fast, deep insights into big data through machine learning—all of which are important across the enterprise. Here are a few examples of how next-generation technology can benefit business transformation:
- Boosted performance of new HPC applications, scientific simulations, financial analytics, video and image analytics, and data compression with Intel® Advanced Vector Extension 512 (Intel® AVX-512)
- Accelerated data compression and cryptographic workloads with Intel® QuickAssist Technology (Intel® QAT) by offloading those functions to capable hardware on the Intel® Xeon® processor Scalable family-based server’s chipset
- Delivery of the performance for tomorrow’s HPC workloads and the ability to scale to tens of thousands of nodes with integrated Intel® Omni-Path Architecture (Intel® OPA) fabric on the Intel® Xeon® processor Scalable family, which provides up to 100 Gbps port bandwidth and low-latency fabric
- Unleashed RAID storage performance of high-speed NVMe* NAND and Intel® Optane™ Solid State Drives with Intel® Virtual RAID on Chip (Intel® VROC) technology—which directly attaches the SSDs to the CPU PCIe* lanes to improve bandwidth, latency, SSD management, and error handling
- Improved server communication speed, reduced transfer latency of large storage blocks, and virtual machine migration with integrated Intel® 10 Gbe Ethernet (up to 4x10 GbE)
With servers based on the Intel® Xeon® processor Scalable family deployed across our data centers, Intel’s business units will be able to quickly take advantage of these next-generation technologies when they need them.
Read the IT@Intel White Paper, “Increasing EDA Throughput with the Intel® Xeon® Processor Scalable Family,” to find out more about our throughput tests and the benefits the Intel® Xeon® processor Scalable family can bring to your data center.